Africa and The Discovery of America III

CORNELL
UNIVERSITY
LIBRARY
BOUGHT WITH THE INCOME

OF THE SAGE ENDOWMENT
FUND GIVEN IN 1891 BY
HENRY WILLIAMS SAGE
Cornell University
Library
The original of this book is in
the Cornell University Library.
There are no known copyright restrictions in

the United States on the use of the text.
http://www.archive.org/details/cu31924088419829
1
AFRICA
AND THE
DISCOVERY of AMERICA
VOLUME III
By LEO WIENER
PROFESSOR OF SLAVIC LANGUAGES AND LITERATURES AT
HARVARD UNIVERSITY; AUTHOR OF "A COMMENTARY TO
THE GERMANIC LAWS AND MEDIAEVAL DOCUMENTS."
"CONTRIBUTIONS TOWARD A HISTORY OF ARABICO"
GOTHIC CULTURE," "HISTORY OF YIDDISH LITERATURE."
"HISTORY OF THE CONTEMPORARY RUSSIAN DRAMA,"
"ANTHOLOGY OP RUSSIAN LITERATURE," "INTERPRETA-
TION OF THE RUSSIAN PEOPLE;" TRANSLATOR OF THE
WORKS OF TOLSTOY; CONTRIBUTOR TO GERMAN, RUSSIAN,
FRENCH. ENGLISH. AND AMERICAN PHILOLOGICAL
PERIODICALS. ETC., ETC.
INNES & SONS

129-135 N. TWELFTH ST., PHILADELPHIA, PA.
MCMXXII
\3
Copyright, 1922, by Innes & Sons
TABLE OP CONTENTS
Page
FOREWORD IX-XII
SOURCES QUOTED .... XlII-XXI
I. THE HISTORY OF COPPER AND IRON 1- 53
II. THE GYPSIES IN EUROPE .54- 77
III. THE GYPSIES IN AFRICA . . 78-115
IV. AFRICAN FETISHISM AND TOTEMISM 116-141
V. THE BORI 142-162
VI. FETISHISM AND SUFISM . 163-179 .
VII. THE CARAIBS ... 180-198

VIII. THE AREYTO . . 199-212
. .
IX. FEATHERS AND MASKS 213-217

X. THE CARAIB SOCIAL ORDER 218-221
XI. THE BORATIO . 222-227
XII. THE MANDINGO ELEMENTS IN THE MEXI-
CAN CIVILIZATION ... 228-322
XIII. THE MEXICAN NEW YEAR 323-351
. .
XIV. CONCLUSIONS 352-370

WORD INDEX .... 872-394
SUBJECT INDEX 395-402
LIST OF ILLUSTRATIONS
Opp.
Page
Griots 100
African Dancing, 2 plates : J 208
Tarring and Feathering in Brazil, 2 plates 216
Asante Stool 218
African Almaizar 232
Sudanic Armor 234
Arabic Blazonry, 2 plates 236
Mexican Blazonry 236
Fetish Nama 250
The Long-Nosed God 258
Kuare Ceremonies 264
African Chastisement 264
Dasiri Tree 264
Plate of the Bacabs 266
Arabic 6adwal 268
Mexican 6adwal 268
Sudanic Rock Inscriptions, 2 plates 270
Tuxtla Statuette, 2 plates 270
The Spider in Mound-builder gorgets 272
6adwal Design in Africa, 2 plates 272
Cadwal Design in Mound-builder gorgets, 2 plates 272
6adwal Design in Mexico, 2 plates 272
Mexican Game of PatoUi 278
Nanauatzin, the Mexican God of Syphilis 284
African Three-Pointed Altar, 2 plates 296
The Rain God Tlaloc 306
Conical Hat, 2 plates 320
African Ball Game 335
Arabic Hockey Sticks 34O
Mexican Ballground, 2 plates 344
FOREWORD.
No archaeologist, no historian, no philologist will be
more startled by the data collected in this book than I
have been in their discovery. While I to a certain ex-
tent foresaw the end toward which the presence of
Africans in America before Columbus must ultimately
lead in the social and religious orders, I did not allow
myself in my first two volumes to be influenced by any
such considerations, but confined myself to an analysis
of the documentary evidence as to the American origin
of cotton, tobacco, the bread roots, and wampum.
When it became necessary similarly to subject the
spiritual culture of the New World to a comparative
study, turned out that the difficulties in the way were
it
far moreserious than when I undertook to brush aside
the accumulated misconceptions in regard to the mater-
ial civihzation of pre-Columbian times. Not only was
the documentary proof scanty for America and fre-
quently distorted by the monks and later by those who
had theories to defend, but the parallel material for
Africa, especially for the Western Sudan, turned out to
be in a more fragmentary condition and even more dis-
torted by investigators totally unacquainted with the
Arabic antecedents of the Sudanic beliefs and customs.
With the exception of the more or less objective attitude
and cautious work of Delafosse and a very few others,
the African material bearing on fetishism and kindred
subjects is a mass of extravagances of which it is not
possible to avail oneself seriously. In archaeology the
Sudan represents almost a blank. Except for the
commendable field work of Frobenius and Desplagnes,
whose conclusions are unfortunately impossible, as has
X AFRICA AND THE DISCOVERY OF AMERICA
been pointed out by Arabioists, the whole region, in
which a dozen powerful states have arisen within the
last millennium, might as well be off the map, —
it cer-
tainly furnishes to the student almost nothing whatso-
ever for a proper comprehension of fetishism, totemism,
the social structure, the mediaeval trade routes, the
organization of the state.
The task seemed hopeless. But it soon became clear
that the prospect was brighter than it had appeared,
when the Sudanic languages were examined for the
Arabic element contained in them. This foreign in-
trusion, as regards Moslem conceptions, had long been
known and studied, but there was a residue of cultural
ideas in nearly every intimate relation in life which had
not even been suspected. Steinthal, in his study of the
Mande languages, pointed out the fact that the "all-
devouring" tendency of the Negro languages often
completely obliterated the borrowed prototype; but,
by including a study of the Arabic influence through
the oases and in the Berber languages, especially in
Zenaga, in the languages of the Niger Bend, such as
Songay and Peul, down to the furthest outposts of the
Arabic trader and magician, among the Yoruba, Asante,
Dahome tribes, and even further, to the Congo, it was
possible to overcome this "all-devouring" tendency and
lay the foundation for an African philology and then
to trace the religious conceptions of the greatest part of
Northern Africa back to Islamic religion and magic.
This study cast a new aspect upon the religious ideas
of the Negroes, heretofore contemptuously denominated
as fetishism, and the delusive totemism, which has led to
a prolixity of theories, became simple and intelligible.
In fact, the spiritual culture of the Sudan appears not
very different from the popular undercurrent of belief
and practice among Europeans or Asiatics, while its
connection with the Moslem folk religion is still closer.
1)
FOREWORD xi
The powerful Moslem interpenetration in spiritual

matters among tlie pagan Negroes became as clear as it
had been in the case of the Moslem Negroes, hence the
thought suggested itself that the caste system of the
Blacks, with their contempt for the blacksmith, which
they share with the Arabs, might itself be of Arabic
origin. In the attempt to solve this question the dis-
covery was made that the treatment of the blacksmiths
was due to the analogous Moslem treatment of the
nomad Gypsies, who found their way to the Niger valley
possibly as early as the Vll. century. While pursuing
the status of the Hindu metal-workers, the history of
iron expanded into a longer chapter than was originally
intended, but it serves to accentuate the fact that the
westward movement through Africa of Asiatic culture,
with its cotton and steel, did not take place on any ap-
preciable scale before the Arabic occupation.
With these necessary preliminary studies, the task of
coordinating the American religious, social, and political
orders with the Mandingo civilization became simple,
although the fragmentary condition of information
seemed to preclude any definite deduction. The two
civilizations are not merely similar, —
they are identical,
in concept, in form, in ritualistic observances, in no-
menclature, and in the Arabic origin of the terms em-
ployed. The matter of chance is mathematically ex-
cluded. If chance can play such pranks, then all his-
torical, archaeological, and philological conclusions are
null and void, and the respective science must be rele-
gated to the lumber-room.
Only the surface has been scratched. Many more
analogies and identities are known to me, but it has
seemed best to spurn any data which are capable of
further elaborations and need the long patient labor of
many men. The Peruvian civilization has barely been
touched upon, because that of Mexico more easily fur-
—
xii AFRICA AND THE DISCOVERY OF AMERICA

nished the direct evidence of the Mandingo origin. The
archaeologist will be disappointed not to find a nicely
worked out chronology, but this is impossible at the
present stage of our knowledge. Only this much is
certain, — the civilization so far investigated cannot be
earlier than of the XII. century, and in all probability
is not older than of the XIV. century.
It will be asked whether an older civilization for
America is denied. It is neither denied nor affirmed,
because it is beyond the scope of the present investiga-
tion. All that is attempted is the separation of late
accretions from what may have existed before. When
the top layer has been thoroughly worked over, there
will be time to work the archaeological ground with a
subsoil plow, without danger of destroying its fertility.
The reader will want to know how to account for the
stone structures and hieroglyphic writing in America,
which do not seem to exist in Africa. To this the an-
swer is that we know almost nothing of the archaeology
of Africa, that recent excavations have revealed mon-
oUths and inscriptions on stone, that amazing sculptures
have been unearthed in Yoruba-land, that the remark-
able Benin culture is still a puzzle. When the Western
Sudan has received the thorough attention from ex-
plorers and archaeologists which it richly deserves, and
we have coordinated all the Sudanic cultures, we shall
be able to answer the questions which are now baffling
us. Such work can be done only by the cooperation
of many men, men free from the virus of specific
"schools," free from academic arrogance, free to follow
the dictates of reason, no matter where it may lead,
seekers of the truth, indeed.
The Author.
SOURCES QUOTED.
[Abiven]. Dictionnaire francais-malinkd et malinkfr-fran-
Sais, Conakry 1906.
Acta sanctorum, July IV.
AUdridge, T. J. A Transformed Colony, London 1910.
Allen, W. and A Narrative of the Expedition to the River Niger,
Thomson, T. R. H. vol. I, London 1848.
Annual of the British School at Athens, The, vol. VIIL
Arevalo, F. M. Avreli dementis Prvdenti V. C. Carmina,
vols. I, II, Romae 1788.
Artin Pacha, Yacoub Contribution k I'^tude du blason en Orient, Lon-
dres 1902.
Avelot, R. Note sur les pratiques religieuses de Ba-Kal€, in
Bulletins et m^moires de la Soci6t6 d'An-
thropologie de Paris, series "VI, vol. II.
Bacon, R. See Brewer.

Baron, R. See Sibree.
Basset, R. Mission au S6n6gal, vol. I, part I, in Publications de
des Lettres d' Alger, vol. XXXIX.
I'ficole
Beck, L. Die Geschichte des Eisens, vol. I, Braunschweig
1884.
Becker, C. H. Der Islam, vol. IV, Strassburg 1913.
Berend, W. See Lepsius.
B&enger-F6raud, L. Les peuplades de la Sln^gambie, Paris 1879.
Berthelot, M. Collection des anciens alchimistes grecs, vols.
I-III, Paris 1887-1888.
La chimie au moyen Sge, vol. I, Paris 1893.
Bezold, C. Die Schatzhohle, aus dem syrischen Texte dreier
unedierten Handschriften in's Deutsche uber-
setzt, Leipzig 1883.
Black, G. F. A Gypsy Bibliography, London 1914.
Blaydes, F. Aristophanis comoediae, Halis Saxonum 1886.
Bork, F. Tierkreise auf westafrikanischen Kalebassen, in
Mitteilungen der vorderasiatischen Gesellschaft,
vol. XXI.
Bosman, W. A New and Accurate Description of the Coast of
Guinea, London 1721.
Bostock, J. and The Natural History of Pliny, London 1855-1857.
Riley, H. T.
Bosworth, J. A Compendious Anglo-Saxon and English Dic-
tionary, London 1852.
Breton, R. Dictionaire caraibe-frangais, Leipzig 1892.
Brewer, J. S. Ft. Rogeri Bacon opera inedita, vol. I, London 1859.
Bricchetti, R. Somalia e Benadir, Milano 1899.
Brinton, D. G. A Primer of Mayan Hieroglyphics, [Philadelphia]
1895.
xiv AFRICA AND THE DISCOVERY OF AMERICA
Nagualism, Philadelphia 1894.
Brugsch, H. Hieroglyphisch-demotisches WOrterbuch, vol. II,
Leipzig 1868.
Brun, J. Le totfimisme chez quelques peuples du Soudan
Occidentale, in Anthropos, vol. V.
Brun, S. Schiffarten, Basel 1624, reprinte4 in Werken
uitgegeven door de Linschoten-Vereeniging,
vol. VI, 's-Gravenhage 1913.
Brunet, L. et Dahomey et d6pendances, Paris 1901.
Giethlen, L.
Burnam, J. M. Recipes from Codex Matritensis A 16, in Univer-
sity of Cincinnati Studies, vol. VIII.
Burton, R. A Mission to Gelele, King of Dahome, London
1864.
See Tootal.
Cardinall, A.W. The Natives of the Northern Territories of the

Gold Coast, London, New York.
Carra de Vatix, B. See Tannery.
Caspari, C. P. Kirchenhistorische Anecdota, vol. I, Christiania
1883.
Charon, G. La society noire de I'Afrique Occidentale Fran-
Caise, Paris 1908.
Christaller, J. G. A Dictionary of the Asante and Fante Language,
Basel 1881.
Clavigero, D. F. S. The History of Mexico, trans, by C. CuUen,
London 1787.
Clement-Mullet, M. Essai sur la min^ralogie arabe, in Journal Asiatique,
series VI, vol.XL
Coleccion de documentos in^ditos de Ultramar, series II, vol. V, series
I.vol.L
Colecci6n de documentos in§ditos relativos al descubrimiento, conquista y
colonizacion de las posesiones espanolas en America y Occeania, series I,
vols. I, III.
Colecci6n de libros y documentos referentes & la historia del Peril, vol. III.
Coppolani, X. See Depont.
Corpus scriptorum historiae Byzantinae, vols. VIII, X, XIV, XXV,
XXVI, XXXVIII.
Crowther, S. See Schon.
Cullen, C. See Clavigero.
Cumont, F. Textes et monuments figures relatifs aux mys-
thres de Mithra, vol. I, Bruxelles 1899.
Defrfimery, C. and Voyages d'Ibn Batoutah, vol. IV, Paris 1879.

Sanguinetti, B.
Deherme, G. L'Afrique Occidentale Franjaise, Paris 1908.
Delafosse, M. Haut-S§n6gal-Niger, series I, vols. I, III, Paris
1912.
Manuel dahomlen, Paris 1894.
Vocabulaires comparatifs de plus de 60 langues
ou dialectes parlfe k la C6te d' I voire, Paris
1904.
SOURCES QUOTED XV
Depont, O. et Les confr6ries religieuses musulmanes, Alger
Coppolani, X. 1897.
Desplagnes, L. Le plateau central nig&ien, Paris 1907.
Destaing, E. FStes et coutumes saisonni^res chez les Beni
Snoiis, in Revue Africaine, vol. L.
Diemer, J. Genesis und Exodus naeh der Milstater Hand-
schrift, vol. I, Wien 1862.
Dindorf, G. Harpocrationis lexicon in decern oratores atticos,
vol. I, Oxonii 1853.
Doughty, C. Travels in Arabia Deserta, vol. I, Cambridge
1888.
Doutt6, E. Magie et religion dans I'Afrique du Nord, Alger
1909.
Merrakech, Paris 1905.
Dozy, R. Dictionnaire d€taill€ des noms des vStements
chez les Arabes, Amsterdam 1845.
Supplement aux dictionnaires arabes, vol. I,
Leyde 1881.
Dlibner, F. Scholia graeca in Aristophanem, Parisiis 1843.
Ducange. Glossarium mediae et infimae latinitatis, Niort
1883-1887.
Duchesne, L. Le liber pontiflcalis, vol. II, Paris 1892.
Duran, D. Historia de las Indias de Nueva Espana, vol. II,
M&dco 1880.
Duveyrier, H. Les Touareg du Nord, Paris 1864.
Ellis, A. B. The Tshi-Speaking Peoples of the Gold Coast of

West Africa, London 1887.
The Yoruba-Speaking Peoples of the Slave Coast
of West Africa, London 1894.
Espana sagrada, vols. XXXVI, XXXVII.
Ewald, P. Gregorii I Papae registrum epistolarum, vol. I,
part I, Berolini 1887.
Perrand, G. Les Musulmans a Madagascar, vol. I, Paris 1891.

Foa, E. Le Dahomey, Paris 1895.
Forcellini. Totius latinitatis lexicon, Patavii 1771.
Forstermann, E. Altdeutsches Namenbuch, vol. I, Bonn 1900.
Ftihner, H. Bezoarsteine, in Janus, vol. VI.
Garcia Icazbalceta, J. Memoriales de Fray Toribio de Motolinia, Mljico

1903.
Gennep, A. van L'etat actuel du problfeme totdmique, Paris 1920.
Gibson, M. Apocrypha arabica, in Studia sinaitica, vol. VIII.
Giethlen, L. See Brunet.
Goeje, M. de M^moire sur les migrations des Tsiganes a travers
I'Asie,Leide 1903.
Goetz, G. Corpus glossariorum latinorum, vols. II-V,
Lipsiae 1888-1894.
G6mara, F. Lopez de La historia general delas Indias, Anvers 1654.
Graff, E. G. Althochdeutscher Sprachschatz, vol. VI, Berlin
1842.
xvi AFRICA AND THE DISCOVERY OF AMERICA
Granada, D. Vocabulario rioplatense razonado, Montevideo
1890.
Gregory the Great. See Ewald.
Grimm, J. Teutonic Mythology, vol. II, London 1883.
Hall, H. See King.

Harpocration. See Dindorf.
Harris, J. R. The Cult of the Heavenly Twins, Cambridge 1906.
Hatzidakis, G. N. Zur Wortbildungslehre des Mittel- und Neugriech-
ischen, in Byzantinische Zeitschritt, vol. II.
Haywood, A. Through Timbuctu and across the Great Sahara,
London 1912.
Hefele, C. J. von Coneiliengeschichte, vol. Ill, Freiburg im Breis-
gau, 1877.
Henry, J. Les Bambara, Mtinster 1910.
Herodotus.
Herrera, A. de Historia de las Indias, Madrid 1730.
Hessels, J. H. A Latin-Anglo-Saxon Glossary, Cambridge 1906.
Hilprecht, H. V. Zur Lapislazuli-Frage im Babylonischen, in
Zeitschrift fur Assyriologie und verwandte
Gebiete, vol. VIII, Berlin 1893.
Hofmann, K. Archiv fiir lateinische Lexikographie, vol. II,
Leipzig 1885.
Holmes, W. H. Art in Shell of the Ancient Americans, in Second
Annual Report of the Bureau of Ethnology to
the Secretary of the Smithsonian Institution,
1880-'81, Washington 1883.
Horstmann, C. The Three Kings of Cologne, London 1886.
Houtsma, M. Th., etc. The Encyclopaedia of Islam, Leyden, London
1913.
Hughes, T. P. A Dictionary of Islam, London 1885.
Hutton, W. A Voyage to Africa, London 1821.
Ibn-al-Baitar. See Leclerc.

Ibn-Batutah. See Defremery.
Isidore.
Jaubert, H. F. Glossaire du centre de la France, Paris 1864.

Jayakar, A. S. G. Ad-Damlrl's HaySt Al-HayawSn, vol. I, London
1906.
Jewish Encyclopedia, The
Jobson, R. The Golden Trade: or, A Discouery of the Riuer
Gambra, and the Golden Trade of the Aethiop-
ians, London 1623.
Johnston, H. A Comparative Study of the Bantu and Semi-
Bantu Languages, Oxford 1919.
JuynboU, Th. W. Handbuch des islamischen Gesetzes, Leiden,
Leipzig 1910.
SOURCES QUOTED xvii
Kersaint-Gilly, F. de Le Nama, in Bulletin du Comit6 d'Etudes his-

toriques et scientifiques, No. 4, Paris 1919.
King, L. and Egypt and Western Asia in the Light of Recent
Hall, H. Discoveries, London 1907.
Kingsborough, E. K. Antiquities of Mexico, vol. II, London 1830.
KoeUe, S. W. Polyglotta afrieana, London 1854.
Lane, E. W. An Account of the Manners and Customs of the

Modern Egyptians, London 1837.
An Arabic-English Lexicon, book I, part V, Lon-
don 1874.
Laoust, E. Mots et choses berblres, Paris 1920.
Largeau, V. Encyclop^die pahouine, Paris 1901.
Las Casas, B. de Historia de las Indias, vol. Ill, Madrid 1875.
Laufer, B. Die Sage von den goldgrabenden Ameisen, in
T'oung Pao, series II, vol. IX.
Leclerc, L. Traits des simples par Ibn El- Beithar, in Notices
et extraits des manuscrits de la Bibliothfeque
Nationale, vol. XXIII.
Leonard, A. G. The Lower Niger and its Tribes, London 1906.
Lepsius, C. R. and Les mStaux dans les inscriptions egyptiennes,
Berend, W. in Bibliothfeque de I'Ecole des hautes Etudes,
part XXX.
Lery, J. de Histoire d'un voyage faict en la terre du Br&il,
vol. II, Paris 1880.
Lippmann, E. O. von Entstehung und Ausbreitung der Alchemie, Berlin
1919.
Lucan. Pharsalia.
Luring, H. Die liber die medicinischen Kenntnisse der alten
Agypter berichtenden Papyri, Leipzig 1888.
D. Joannis Praefationes, Venetiis 1740.

Mabillonii,
MacDonald, G. The Gold Coast, London, New York, Bombay
1898.
Mannhardt, W. Wald- und Feldkulte, vol. I, Berlin 1904.
Mappae Clavicula, in Archaeologia, vol. XXXII.
Marees, P. de Beschryvinghe ende historische verhael van het
Gout Koninckrijck van Gunea, 's-CJravenhage
1912.
Meyer, L. Handbuch der griechischen Etymologie, vol. Ill,
Leipzig 1901.
Migne, J. P. Patrologia graeca, vol. LXXXVI. 1.
Patrologia latina, vol. CLXXVII.
Miklosich, F. fiber die Mundarten und die Wanderungen der
Zigeuner Europa's, part VI, Wien 1876.
Modi, J. J. The Game of Ball-Bat among the Ancient
Persians, as Described in the Epic of Firdousi,
in The Journal of the Bombay Branch of the
Royal Asiatic Society, vol. XVIII.
Molina, A. de Vocabulario de la lengua mexicana, Leipzig 1880.
Montgomery, J. A. The Samaritans, Philadelphia 1907.
Monumenta Germaniae historica, Capitularia, Concilia, vol. I.
xviii AFRICA AND THE DISCOVERY OF AMERICA
Morley, S. G. The Inscriptions at Copan, Washington 1920,
Morton, J. The Ancren Riwle, London 1853.
Muratori, L. A. Antiquitates italicae medii aevi, vol. II, Mediolani
1739.
Nasmith, J. Itinerarium Symonis Simeonis, et Hugonis Illum-

inatoris ad Terram Sanctam, Cambridge 1778.
Newbold, Capt. The Gypsies of Egypt, in The Journal of the Royal
Asiatic Society of Great Britain and Ireland,
vol. XVI, part II.
Nicholson, R. A. The kit&b al-luma' fi 'l-taâwwuf, in "E. J. W.
Gibb Memorial" Series, vol. XXII.
Nicole, J. L'edit de I'empereur L§on le Sage, in M^moires
de I'Institut national genevois, vol. XVIII.
Oppenheim, M. Vom Mittelmeer zum Persischen GoU, vol. I,

Berlin 1899.
Orozco y Berra, M. Historia antigua y de la conquista de M&rico,
vol. I, Mexico 1880.
Oviedo, F. de Historia general y natural de las Indias, vols.
I-IV, Madrid 1851-1855.
Palmer, H. R. Notes on the Kororofawa and Jukon, in Journal

of the African Society, vol. XI.
Parkhurst, J. An Hebrew and English Lexicon, London 1813.
Pelly, L. A Visit to the Wahabee Capital, Central Arabia,
in The Journal of the Royal Geographical
Society, vol. XXXV.
Perregaux, E. Chez les Achanti, in Bulletin de la Societe neu-
chSteloise de Gôgraphie, vol. XVII.
Pliny. Historia Naturalis.
See Bostock.
Portugaliae monumenta diplomatica, vol. I.
Proceedings of the Society of Biblical Archaeology, vol. XIII.
Purchas, S. Hakluytus Posthumus or Purchas His Pilgrimes,
vol. VI, Glasgow 1905.
Quatremfere, M. Histoire dea sultans mamlouks, vol. I, parts I,

II, vol. II, part I, Paris 1837.
Raccolta di docimienti dalla R. Commissione Colombiana, part I, vol. I,

Roma 1892, part III, vol. II, Roma 1893.
RadSenko, K. Einige Bemerkungen zur neugefundenen Abschrift
des Lebens des heil. Barbar in bulgarischer
Ubersetzung, in Archiv ftir slavische Philolorie,
vol. XXII.
Raflfenel, A. Nouveau voyage dans le pays des Nfegres, vol. I,
Paris 1856.
Rheinisches Museum, vol. XLV.
SOURCES QUOTED XIX
Kiley,H. T. See Bostock.

Robert, A. L'Arabe tel qu'il est, Alger 1900.
Rochefort, C. de Histoire naturelle et morale des Isles Antilles de
rAmfirique, Roterdam 1658.
Roger, F. E. La Terre Saincte, Paris 1646.
Roth, H. L. Great Benin, its Customs, Art and Horrors, Hal-
ifax 1903.
Ruelle, E. Notes anthropologiques, ethnographiques et soc-
iologiques sur quelques populations noires du
2e territoire militaire de I'Afrique Occidentale
Franeaise, in L'Anthropologie, vol. XV.
Ruska, J. Untersuchungen uber das Steinbuch des Aristoteles,
Heidelberg 1911.
Rymer, T. Foedera, vol. I, London 1727.
Sachau, E. Muhammedanisches Recht, in Lehrbucher des

Seminars fiir orientalische Sprachen zu Berlin,
vol. XVII.
Sahagun, B. de Histoiregdn^rale des choses de la Nouvelle-
Espagne, Paris 1880.
Saint Lo, A. de Relation dv voyage dv Cap-Verd, Paris 1637.
Sanguinetti, B. See Defr^mery.
Sauvant, P. Grammaire bambara, Maison-Carrfe (Alger) 1913.
Scheepstra, T. Van den heilighen drien coninghen, Groningen
1914.
Schellhas, P. Representation of Deities of the Maya Manu-
scripts, in Papers of the Peabody Museum of
American Archaeology and Ethnology, Harvard
University, vol. IV, No. I.
Schon, J. F. and Journals of the Expedition up the Niger, London
Crowther, S. 1842.
Schwarzlose, F. W. Die Waffen der alten Araber aus ihren Dichtern
dargestellt, Leipzig 1886.
Scriptores rerum merovingicarum, vols. Ill, IV.
Seler, E. Codex Borgia, vols. I-III, Berlin 1904^-1909.
Gesammelte Abhandlungen zur amerikanischen
Sprach- und Alterthumskunde, vol. II, Berlin
1904.
Seybold, C. Glossarium latino-arabicum, Berolini 1900.
Sibree, J. and The Antananarivo Annual and Madagascar Mag-
Baron, R. azine, vol. IX.
Sievers, E. See Steinmeyer.
Simeon, R. Dictionnaire de la langue nahuatl, Paris 1885.
Smith, R. P. Thesaurus syriacus, Oxonii 1868-1879.
Smith, T. B. Relation of Alvar Nunez Cabeja de Vaca, New
York 1871.
Spiess, C. Fabeln iiber die Spinne bei den Ewe am Unterlauf
des Volta in Westafrika, and Fortsetzung der
Fabeln iiber die Spinne bei den Ewe am Unter-
lauf des Volta in Westafrika, in Mitteilungen des
Seminars fur orientalische Sprachen, vol. XXI,
part III, vol. XXII, part III.
Stade, H. See Tootal.
XX AFRICA AND THE DISCOVERY OF AMERICA
Steinen, K. von den Unter den Naturvolkern Zentral-Brasiliens, Berlin
1894.
Steinmeyer, E. und Die althochdeutschen Glossen, vols. I-IV, Berlin
Sievers, E. 1879-1899.
Steinschneider, M. Die ,,Skidy" oder geomantischen Figuren, in Zeit-
schrift der deutschen morgenlandischen Gesell-
schaft, vol. XXXI.
Steinthal, H. Die Mande-Neger-Sprachen, Berlin 1867.
Strabo.
Sudheim, L. de De itinere Terre Sancte, in Archives de 1' Orient
latin, vol. II.
und
Tafel, G. L. Fr. Urkunden zur alteren Handels- und Staatsges-
Thomas, G. M. chichte der Republik Venedig, vol. I, Wien 1856.
Tannery, P. and Le Rabolion, in M^moires scientifiques, published
Carra de Vaux, B. by J. L. Heiberg, vol. IV, Toulouse, Paris 1920.
Terrien de Lacoup- Western Origin of the Early Chinese Civilization,
erie, A. London 1894.
Theophilus. Schedula diversarum artium, in Quellenschriften
fur Kunstgeschichte und Kunsttechnik des
Mittelalters und der Renaissance, vol. VII.
Thevet, A. La cosmographie universelle, vol. II, Paris 1575.
Thomas, C. Notes on Certain Maya and Mexican Manuscripts,
in Third Annual Report of the Bureau of Ethnol-
ogy to the Secretary of the Smithsonian Institu-
tion, 1881-'82, Washington 1884.
Thomas, G. M. See Tafel.
Thomas, N. W. Timne-English Dictionary, in Anthropological
Report on Sierra Leone, part II, London 1916.
Thomson, T. R. H. See Allen.
Tootal, A. and The Captivity of Hans Stade of Hesse (The Hak-
Burton, R. luyt Society), London 1874.
Tremearne, A. J. N. Hausa Superstitions and Customs, London 1913.
The Ban of the Bori, London 1914.
Trilles, H. Le totSmisme chez les Fan, Miinster 1912.
Torquemada, J. de De la monarquia indiana, vol. II, Madrid 1723.
V&mb&y, H. Etymologisches Worterbuch der turko-tatarischen

Sprachen, Leipzig 1878.
Vincent of Beauvais. Speculum doctrinale, Venetiis 1494.
Speculum naturale, Venetiis 1494.
Wallis, C. B. The Advance of our West African Empire, London

1903.
Wellhausen, J. Skizzen und Vorarbeiten, vol. Ill, Berlin 1887.
Westermann, D. Worterbuch der Ewe-Sprache, Berlin 1905.
Westermarck, E. Ceremonies and Beliefs Connected with Agricul-
ture, Certain Dates of the Solar Year, and the
Weather, in Morocco, Helsingfors 1913.
The Magic Origin of Moorish Designs, in The
Journal of the Anthropological Institute of
Great Britain and Ireland, vol. XXXIV.
SOURCES QUOTED XXI
Wherry, E. M. A Comprehensive Commentary on the Quran,

London 1882-1886.
vols. I-IV,
Wiener, L. Africa and the Discovery of America, vols. I, II,
Philadelphia 1920, 1922.
Commentary to the Germanic Laws and Mediaeval
Documents, Cambridge 1915.
Contributions toward a History of Arabico-Gothic
Culture, vol. I, New York 1917, vol. IV, Phila-
delphia 1921.
Gypsies as Fortune-Tellers and as Blacksmiths, in
Journal of the Gypsy Lore Society, new series,
Nos. I, IV.
vol. Ill,
Ismaelites, in Journal of the Gypsy Lore Society,
new series, vol, IV, No. II.
Wilson, C.W. On the Tribes of the Nile valley, North of Khartfim,
in The Journal of the Anthropological Institute
Great Britain and Ireland, vol. XVII.
of
Wright, T. and Anglo-Saxon and Old English Vocabularies, vol. I,
Wulcker, R. London 1884.
Wiilcker, R. See Wright.
Wylde, A. Modern Abyssinia, London 1901.
Zayas y Alfonso, A. Lexicograffa antillana, Habana 1914.

Zeitschrift der deutschen morgenlandischen Gesellschaft, vol. XXXIII.
Zeitschrift ftir die Geschichte des Oberrheins, vol. IV, Karlsruhe 1853.
Zimmermann, J. A Grammatical Sketch of the Akra- or Ga-Lan-
guage, vol. II, Stuttgart 1858.
CHAPTER I.
The Histoey op Copper and Iron.
(a) Sumerian urudu.
The Assyrian name for "copper" or "bronze" is

eru, for which the Sumerian equivalent is urudu. Re-
lated to the first is Assjrrian eru "to engrave, carve,
be strong, " but this is, no doubt, derived from the con-
notation "bronze," since the root is absent from the
other Semitic languages. Sumerian urudu is unques-
tionably a development of the root ur, which in Sum-
erian seems to have had the original connotation "to
be strong," to judge from ur "lion, man, dog, enemy,
bone, enclose. " Unfortunately the fragmentary condi-
tion of the Sumerian vocabulary does not permit to
determine positively the connotational value of the
roots. This may, however, be accomplished by con-
sulting the Dravidian languages, which are nearest
related to the ancient Sumerian.
The Dravidian languages have preserved the "bronze"
word in the sense of "iron, steel," a confusion which is
universal in India, due to the fact that the two metals
were interchangeable in common use. Thus Sanskrit
lohas means both "copper" and "iron." For "iron"
we have Tamil irumbu, Malayalam irimba, Telugu
inumu. But we also have Kannada urku, ukku, Tamil,
Malayalam urukku "wootz, steel," which cannot be
separated from Kannada urku, ukku "to rise, swell, be
greatly increased, boil excessively, be elated, pride,
power," that is, we have in urku "steel" a "strong"
word. Other derivatives, such as Kannada urbu, uxhu
2 AFRICA AND THE DISCOVERY OF AMERICA
"to be elated, become full of effort, energy, firmness or
joy," wnuvu "mass, excess, bigness, excellence," show
that the original root is uru " to fix, stick in the ground,
to settle one's abode, stop, reside, stay, penetrate,
exist, be," uru "to be, stay, stop," which are all de-
rived from the original meaning "to be strong" and,
if we go further back, "to enclose," hence both Tamil,
Malayalam, Kannada ura "exertion, great effort, to be
strong, to be violent as the wind" and ur " an inhabited
place, village, town" belong here. Similarly we have
Sumerian ur "enclosure, to surround," uru "to found,
plant, town, settlement, dwelling," uru "whirlwind,"
all derived from the same root. Even Sanskrit pos-
sesses this root in all these connotations. have We
uru "wide, spacious, extended, large, much, excessive,
precious, space, room," etc., and Sanskrit uru "the
thigh" the same as Sumerian ur "foundation, lower
is
part, leg." It is not my purpose here to trace the
enormous ramifications of this root in the Indo-
European, Semitic, and other families of languages,^
but only to show that Sumerian urudu is a derivative
of the root ur "to be strong."
This is also borne out by Assyrian eru, for we have
the corresponding Dravidian words, all of them de-
rived from the "strong" root. The Tamil irumbu
"iron," etc. is obviously derived from ir "to be, exist,
stay, hesitate," a variation of uxu. From the same
root come Kannada ira "the state of being compressed,
confined," iri "to beat, strike, butt, Mil." But we
have also Kannada eru "to become more or much,
mount, climb," e.r "the state of being fit, to meet in
battle, a wound," ere "master, fitness for being
poured, to cast metals," hence eraka "any metal in-
fusion. " Here again the Sanskrit has corresponding
forms, such as irya "active, powerful, destroying
' Such a work is in preparation.
"
THE HISTORY OF COPPER AND IRON 3
enemies, lord," ir "to move, rise, elevate," er "to

raise, obtain." The original metal to which the
"strong" root was applied was unquestionably "cop-
per" or "bronze" and not "iron, steel," as may
be seen from Tamil eruvve "blood, copper," while we
have Kannada ^rS "a dark-brown or dark-red color,
soil of a dark color." Here the notion "dark-red,
black" evolved from that of "copper, bronze," and
bronze is on the Egyptian monuments invariably rep-
resented red, while iron and steel implements are al-
ways painted blue. It would seem that Kannada
karbun, karbdn, kabbina, Tulu karba, Toda kabbun
"iron" are all derived from the Dravidian kar "black,
hence that iron is as old or even older than copper or
bronze, but this is at once dispelled by ancient Tamil
karumbon, for the modern irumbu, where obviously
kar corresponds to ir and is again a "strong" word.
We have Kannada kara, kadu "firmness, strength,
abundance," kadupu "firmness, great valor, pride,"
kadi "great, large, to bite, cut, hack," kare "to give
forth rain, blackness," and an enormous mass of de-
rivatives in all the other Dravidian languages.
Kar is the older form, from which ur, ir, er, already
discussed, are mere deteriorations. This also appears
from the Sumerian, where we have gar "fetters, to en-
close," kal "strong," gal "great," kar "wall," and
from the Sanskrit kar as "doing, hand, elephant's
trunk, " and a very large number of derivatives from it.
But that "doing, hand" is here identical with our
"strong, enclose" root appears from Sumerian gar
"to do," Kannada kay, key "to do, the hand, ele-
phant's trunk." In the latter case it is important to
observe that ar, ur, etc. have in Dravidian the tend-
ency to turn into ay, uy, etc. Thus by the side of Kan-
nada ere "lord" we have ayya "master, lord," and by
the side of Kannada ixi "to throw" we have ey "to
cast, throw, shoot," and to Sanskrit kara "pungency,
caustic" corresponds Kannada kdy "to grow hot."
Similarly Kannada dr "to become strong, be able,"
dru "power, daring," dr "to be fit" exist by the side of
dya "extent, measure, fitness." This brings us at once
to the mysterious Sanskrit ayas, Avestan ayo "ore,
iron, " which have puzzled the philologists and turn out
to be reduced Dravidian forms of the root ar, er, etc.,
which is also found in Latin aes "copper, bronze."
One of the important subgroups of Dravidian kar
is the one which develops the meaning "to cut," and
which is more commonly represented by kad or kat, as
in Kannada kati "to cut a stone with a chisel," kadi
"to cut, hack, a chip, piece, to bite off, to itch, gnash
the teeth, end, corner." That we really have here a
"strong" word follows from kadi, kadu "great, large,"
and we have already seen that this takes us back to
kar "strong." We have the interesting Kannada
karku, karaku, kadaku, kakku, garku "a jag, rough part
of a millstone, toothed part of a file, roughness, sharp-
ness, " Tamil karukku "roughness, sharpness, the edge
of a sword or knife," Kannada karadu "that is rough,
uneven, hard. " Hence Sanskrit kharas "hard, rough,"
karkaras "hard," karkagas "rough, hard," Greek
xdp-)âpot: "sharp," xap-^d^soz "rough, sharp," yapdaattv
"to whet" are all related to the Dravidian roots,

and a large number of Dravidian kar, ka\ "cut" roots
are found in Sanskrit and much less frequently in the
other Indo-European languages, but they do not in-
terest us here, except Persian kard "knife," which, no
doubt, is responsible for Ostyak karte, Votyak kort,
Zyrenian kort, Cheremissian kirtne "iron."
In the Dravidian languages the form kal is used with
the original connotation of "hard," as in Kannada kal
"stone." There can, therefore, be little doubt that
in Greek ;ijaAxoc "copper, bronze," OBulgarian zeUzo,
Lithuanian geleiis "iron" we have borrowings from a

Central Asiatic language corresponding to Kannada
karku, garku "sharpness."
So far the etymological history of the words discussed

shows that the use of copper and iron proceeded from a
non-Aryan race, either from the Sumerians or the re-
lated Dravidians. The Sumerian urudu leads to a
great number of Indo-European words that, by the
presence of the suffix du in the stem, betray their
borrowing from the original urudu. The Sanskrit
rudhiras, rohitas "red" and lohas "reddish, copper,
iron" are directly or indirectly derived from the Sum-
erian. In Chinese the latter oeciirs already in the Shu-
King as leu, old pronunciation lu, "steel," under the
date of 2200 B. C, and the same sound also means "to
cut, engrave." It would seem that Sanskrit lohas is
borrowed from the Chinese, but the date of the writing
of the Shu- King is not known. On the other hand San-
skrit lohas, in one form or another, is found so far spread
that at least a millennium before the Christian era India
must have been the home of metallurgy, from which
the information spread to the west and north. We
find it in Hebrew f^^^HJ nehoSet, Arabic tr'"**' nuhds,
Sjrriae nhdM "copper." The change from lohas to
Arabic nuhds is also found in another similar Hebrew
root l]],S, nh§ "to enchant." The usual Sanskrit name
for "copper" is tamra-loham, literally "red metal."
This led to Mongolian temur, Tatar temir, Turkish
demir "iron," which must be due to the ancient Hindu
colony in Turkestan. In the European languages the
Sumerian urudu did not pass through the Sanskrit
lohas. Basque urraida "copper" bears witness to the
antiquity of the word in Europe, possibly before the
Indo-European invasion. The "red" words, Greek
ipudpoc, Latin ruber, rufus, OBulgarian rudU, like
Sanskrit rudhiras, rohitas, cannot be separated from the
"ore" words, Latin rudus, OBulgarian ruda, and
ONorse raudi, Finnish rauta, Livonian roda, Lappish
ruovdde, in all of which the reference is to iron ore.
These are all borrowings from languages that ultimately
hark back to Sumerian urudu.
(b) Assyrian parzillu.
In Sumerian bar means 1) "to divide, open, a half;"

2) "to bind, surround, a side, outside, open field;"
3) "bright, to shine, the sun;" 4) "to be high, to rise;"
5) "to spread out." That all these connotations arise
from the same root appears from a study of the corre-
sponding Dravidian root. We have Kannada pare
"to spread, scatter, extension, stretch, fold," pari "to
run," that is, the fifth Sumerian connotation, which,
at the same time, explains Sanskrit pari "round,
around," and at once leads to the second Sumerian
connotation. Kannada piri, Tamil paru, peru "ex-
tensiveness, largeness, greatness, preeminence" show
how from "to spread" one comes to "large, great,"
and Sanskrit para "distant, beyond, subsequent,
other, very excellent, high" adds proof to the fact that
the fourth Sumerian connotation evolves from the other
two. Similarly the idea of "to spread, disperse" at
once leads to "to divide," hence Kannada pari "to
tear asunder, break off, cut," and the first Sumerian
connotation is accounted for. The third Sumerian
connotation is contained in Kannada pari "ornament,"
Tamil pari "gold," por "metal," also found in San-
skrit pari, which, by the side of "round," also means
"ornament, decoration."
Thus we arrive at Sumerian bar "metal" in za-bar,
older udka-bar, "copper," an-bar "iron." The first
means literally "stone-bright," and just as Assyrian
parasu "to divide," parsigu "band," §upparuru "to

spread out" are ultimately derived from Sumerian
bar, so is Assyrian siparru "copper" ultimately to be
referred to Sumerian za-bar, and Latin cuprum is, no
doubt, originally of a similar origin. Sumerian an-bar
"iron," literally "sky-bright," refers to the blue color
of the iron or steel, as generally represented on Egyp-
tian monuments, hence Egyptian bda-en-pet "metal of
heaven, iron" does not refer to meteoric iron, as some-
times assumed, but only to the color of the iron imple-
ments. The Assyrian term parzillu "iron" is obviously
a compound of Sumerian bar, and the second part is
presumably Sumerian sil "to cut," or gal "to divide."
This word is also found in the other Semitic languages,
Hebrew barzel "iron, " Arabic /arsaZa "he put on fetters,"
etc. This term is distributed over an enormous terri-
tory, for we have Malay besi, Javanese wesi. Cam pasei,
basei, sei,Carolinian wasei, Malagassy wy, Maori wi,
etc. There can be little doubt that Latin ferrum is de-
rived from the same source, through a form fers.
The Sumerian sil, which, as we see, led to the mean-
ing "iron" represented in a large number of languages.

is
We have Chinese t'ieh, old pronunciation fit, but, since
the ideogram is composed of two parts which mean
"foreign metal, " it follows that China is not the original
country of iron manufacture. This Chinese word is
found in Annamese as thiet or sat, in Tibeto-Burman as
Lalung sar, Dhima sher, Garo ser, sil, Tipura sor, Bara
shurr, Mech shoora, Andro sen, Sengmai sel, Chairel
thlr, Khougzai thi, Sairang thih, Siyin chi, khl, Lai tirh,
Shonshe tir, Taungtha shi, Sho n'thi. In Tibetan we
have ca, while Japanese tetsu is directly borrowed from
the Chinese. In the Philippine languages we have ap-
parently a borrowing from Chinese bak t'ieh "white
iron, "^ which is most likely a parallel to Assyrian
parzillu. Such are Igorot patachim, Bontok patatjim,
Bagobo puto, Moro putau, Bisaya pothao, Tagalog bakal.
By the side of Chinese tHeh "iron" we have ts'ieh,
old pronunciation ts'it, "to cut," and this makes it
possible once more to locate the origin of the word.
We have Sanskrit chedas " a cut, piece, " chidram "hole,"
chinatti "he cuts off," to which are related Avestan
saed, Greek ^X'^^^ Latin scindo "1 split." But these
are not specifically Indo-European words, for we find
them on a bewildering scale in the Dravidian languages.
The Kannada "little" words, kiri, giddu, cinna, cini,
citi, sidi, areablauts of the root kati "to cut a stone
i
with a chisel," kadi "a cut, piece, bit," already dis-
cussed. These at once show that Sumerian kid "to do,
to split, to enclose" are all derived from the same root
and are identical with kar, kad, which has the same
meanings in the Dravidian and Sumerian. This is
further shown by Assyrian gardSu "to split," kalu "to
enclose," by which Sumerian kid is translated, where
we get the original roots kar, kal.
But this Sumerian kid "to split" is also translated by
Assyrian gallabu, a development of the root kar, gal,
already discussed. We have Asssrrian galbu "cut,
torn," galabu "to cut, castrate," found also in Syriac
galdbd, Talmudic gelabd "a sharp tool, razor," which
show that Greek yXunrscv "to chisel," ylaûpbz
"smooth," Latin glaber "bald," etc. are all borrow-
ings from the east, where the sharp-cutting tool, that is,
steel, was first invented. But Assyrian galbu is a de-
rivative of Sumerian gal "to divide," that is, we once
more come back to our "cut" words.
Avestan sidara "rift, hole," sind "destruction" are
derived from the root saed, of which the participial
^ A reference to white iron is given in the Pi-tan. See L. Beck, Die
Geschiehie des Eisens, Braunschweig 1884, vol. I, p. 295.
form Qastram means "a cutting metal, iron," hence one

would expect somewhere in the Indo-European lan-
guages a form sidar a for "iron, " which is found in Greek
atdrjpoc. But this root is found in many languages.
Etruscan Sethlans "Vulcan" is apparently derived
from an Indo-European language. For Egypt we have
Plutarch's statement, from Manetho, that iron was
called "the bone of Seth, " and Set means not only " the
god Seth, " but also "to cut, pierce. " The metal tehaset,
which has been variously taken for "copper" or "iron"
is, in all probability, popularly derived from such a
combination and originally refers to iron.

The cumulative evidence of the preceding investiga-
tion confirms the assumption that the metals were at
first worked in Central Asia, whence, before the first
pre-Christian millennium, the knowledge of copper and
iron had spread in all directions.
(c) Assyrian haginnu.
It is now
well established that much of the Egyptian
religion,especially the sun-worship, is of Sumerian
origin.^ The sky-god of Edfu was surrounded by his
mesniu or smiths, but mesen originally means " the place
where metallic work is done," then "the adytum con-
secrated to Horus, " and only in the last instance
"smith." As a verb it means "to protect." This
shows that we are deahng here with an m derivative of a
verb, which should mean "to work in metals" and "to
protect," and this leads us to Assyrian haganu "to
protect," hugannu "sharp sword," haginnu "axe,"
which are derivatives of hagagu "to break, cut off."
The same connotations of "to protect" and "axe" are
found in Arabic u-^ hasn "he preserved or guarded
' L. King and H. Hall, Egypt and Western Asia in the Light of Recent Dis-
coveries, London 1907, p. 39 n.
it," ilrr^ hasln "axe." The relation of the two words

is madeclear from some Egyptian words, aghu "car-
penter's adze, axe" and aghau "axe-men, soldiers,"
which are from the root geh "a kind of stone, flint,"
gehgeh "to cut stone, carve, engrave." The Egyptian
root gefi is, no doubt, identical with the "cut" roots so
far discussed. It is certainly not an accident that Assy-
rian haginnu is found in Sanskrit as kathinas "hard,
sharp," and in Greek as dîv^ "axe." The latter are
borrowings from the source that first developed the
"axe" words. But this is certainly not a Semitic word,
since the regularity of the ending -in, attached to the
root has, as also in Talmudic **J"'¥Q hagina, ^^'^.^
hazina, Syriac hagind, points to a common borrowing.

The simpler Arabic ^^ had "sharp" produced
•^.^ hadld "iron," but the Assyrian haginnu has re-
mained an unrelated word in all the Semitic languages,
hence it is an intrusion from without.
The Sanskrit and Dravidian languages have deriva-
tives meaning "a cutting tool" from kati "to cut with a
chisel," such as Sanskrit kattdra "a dagger," kuthdra,
Kannada kodali "axe," hence, just as these are related
to Sanskrit kathora, Kannada katura "hard," so are the
other "axe" words related to Sanskrit kathinas, but
this is not of necessity originally Sanskrit. All these
words are more likely Sumerian derivatives, or com-
pounds, of gaz, ga§, âz "to break, crush." Ethiopic
hasin, hasin, hatin, Chamir agin more commonly mean
"iron, steel," and before the VII. century A. D. this
passed into Pehlevi as dsin "iron, " and is found also in
Kurd hasin, hesin, awsin, Persian dhen, Ossetic dfsdn,
Afghan ospanah, ospina "iron," while Gypsy ab6in,
apsin has preserved the meaning "steel. " In Egypt the
old masent "smith" has led to Coptic basnet, besnat
(that is, vasnet, vesnat) "blacksmith."
(d) Arabic âtr.
Vincent of Beauvais, quoting Avicenna, says: "There

are four kinds of iron. The first is of Spain, ît is hard; —
from it hammers, wedges, and such like things are made,
but it is not good for cutting, and does not enter the
science of alchemy. Another kind is alidena, ît is —
coarse and is not good to work with. The third kind is
steel {acerium), which can sharpen Spanish iron. The
—
fourth is Hindu iron, it cuts better than any of these."
"There is another kind of iron in the Orient, it is —
commonly called alidena. It is good for incisions, and is
malleable like copper or silver, but is not ductile."^
Roger Bacon, also quoting Avicenna, says: "Iron, ac-
cording to the universal opinion, is of the nature of
Mars The kinds and operations of iron are ex-
pounded in the Fifth Book of Avicenna 's De Anima.
There are three plainly different kinds. One kind of
iron is good to sustain and give blows and forgings, and
to be cast into any form desired by strong blows and fire,
as in the case of hammers and anvils. This kind of iron
is good for striking and warding off blows, not for cut-
ting. Another is commonly called 'iron' (ferrum),
from which are made iron tools that have to sustain per-
cussions. Another kind is better for sharpening and
cutting, such as steel (chalybs), and is purer than com-
mon iron, and it has more of heat and so is better adapted
for cutting and sharpening, according to Avicenna, be-
cause it is not so ductile nor so malleable, nor fit for
striking or sustaining blows. And a third kind is called
andena, which is less common among the Latins. It
differs from common iron, in that common iron cannot
be drawn out or beaten except when greatly heated,
whereas andena needs be heated only like silver, and it
is not so fit for cutting as steel. But it is better for
' Speculum naiurale, viii. 52 ff.; Speculum doctrinale, xi. 114.
sustaining and giving blows than steel, while common
"^
iron surpasses both in this.
Avieenna unquestionably referred to "male" and
"female" iron, in addition to the common iron. The
poet Ibn Errumi says, "Which weapon is best? Only
a sharp sword with male edge and female blade, "^ and
the alchemical writings make it clear that Avieenna 's
definition, as given by Roger Bacon, is correct, even as
Ibn Haukal, in the middle of the X. century, referred to
the two kinds of iron.^ The hard " male " iron was called
_H J dzakar, while the soft was known as '^' 'anit
"female." Dkar is still the expression for "steel"
among the Berbers. In the Spanish-Arabic dictionary
of Alcala azero is translated by daquir and hind, that is,
by "male" and "female," but the latter has been con-
fused with iJ-^ hindl "Indian," because the best steel
actually came from India.
For "male" another Arabic word gained
J^'^ dzakar
recognition, namely J^ 'atr, 'itr, 'utr, originally "the
diversified wavy marks of a sword and its luster and
glitter," hence J/ ^ ma'tur "a sword having in its

blade diversified wavy marks, or luster, or glitter, or
having its blade of female, or soft iron, and its edge of
male iron, or steel, or that is said to be of the fabric of
the jinn, or genii." Thus atr came to mean "steel"
par excellence. This word was in Spain confused with
j^^ 'asr "strength of make or form," v-"' 'asara "he
bound, braced, or tied," through a series of formal
blunders.
' Fr. Bogeri Bacon opera inedita, ed. by J. S. Brewer, London 1859, vol. I,
p. 382 f.
" F. W.
Schwarzlose, Die Waffen der alien Araber aus ihren Dichtern
dargestellt, Leipzig 1886, p. 142.
' E. O. von Lippmann, Entstehung und Ausbreitung der Alchemie, Berlin
1919, p. 403.
In the Affatim glosses^ we have "calips ferrum uel

fornax, " where "fornax" a contamination with the
is
succeeding gloss "caminus fornax." In the Codex
Vaticanus 1468^ we read "calips fornax ferri uel furca
penalis, " where to the blunder in the Affatim gloss is
added an unusual explanation, and where steel is made
equal to "gallows. " The earliest reference to this is in
Codex Sangallensis 912} The addition can be under-
stood only through the gloss in the Latin- Arabic vocab-
ulary,* where we have "calips ferrum," and, immedi-
"calibum" J^, Jt^, that is, "a ring, or

ately preceding it,
which is put upon the neck, a shackle
collar, of iron,
with which the Arabs used to confine a captive when
they took him, made of thongs." This is precisely a
"furca poenalis" of the Latin glosses, but the reading
calibum at once indicates that we have here a gloss to
Lucan's Pharsalia,YI. 797, " aeternis Chalybum nodis,"
which has obviously the meaning "vinculum, fetters."
Five years ago I wrote: "If the Gothic Bible is based
on a Greek text, then eisarn never means 'iron,' but
only 'chains, irons,' for it is the translation of Greek
iXuaeti; and Tzedat. As the corresponding passages
in the Vulgate have each time 'catenae' for eisarn,
there cannot be the slightest doubt that eisarn did not
mean 'iron' in Gothic. It is generally supposed that
Goth, eisarn means 'iron' and is derived from the
Celtic, because of the specific statement in the Vita
Eugendi (+510), that in the Gallic language Ysarno-
dorus means f errei ostii.
'
But as it has been conclu-
'
sively shown that the Vita was written after 800, the
explanation is valueless, as are similar other attempts
of the mediaeval author.
^ G. Goetz, Corpus glossariorum latinorum, Lipsiae 1894, vol. IV, p. 491.

2 Ibid., vol. V, p. 493.
5 Ibid., vol. IV, p. 252.
* C. Seybold, Glossarium latino-arabicum, Berolini 1900.

"As Goth, eisarn means 'catena, S-lumz,' it isab-
sxird to begin with the meaning 'iron.' Now Arab.
J^' isdr-un means which one binds, a
'a thing with
thong of hide, a rope or cord, with which a
untanned
captive is bound, a pair of shackles.' Obviously the
other Germanic languages borrowed this Arabic word
through the Gothic, where it had a leaning toward the
meaning 'iron chain.' In OHG., Isarn means 'iron,'
and very early we get here the corruption Isen, isin, lead-
ing to Ger. Eisen. Similarly AS. Isern deteriorates
early to Isen, Iren, producing Eng. 'iron,' while in ONorse
Isarn occurs only in poetry and popularly changes to
earn, jdrn. From the ASaxon the word passed into
Olrish as tarn, hiarn, and spread to the other Celtic
languages."^
The discovery I have now made confirms my former
assumption to a nicety rarely to be hoped for in
philological investigations, and the lateness of the
Gothic Bible is once more vindicated beyond any
possibility of cavil, except by the mentally bUnd.
Again, in a Hebrew translation of Ibn-al-Gezzar 's work
on stones, there is a quotation from the Lapidary of
Aristotle, a IX. century forgery, where it says, among
other things, that the magnet attracts TDVi aslr.
Ruska'' thinks that this is a miswritten v-r-' or _;•.',
but he is certainly mistaken. It can be only J^ 'atr or

J-' 'asr "steel" or "iron."
The Germanic "iron" words could not have arisen
before the VIII. century. But the Romance languages,
too, have them. The Arabic word for "steel" is found
in the Glossae nominum, which is of Anglo-Saxon origin.
1 Contributions toward a History of Ardbico-Gothic Culture, New York 1917,

vol. I, p. xxvi f.
* J. Ruska, Untersuehungen iUber das Steinbuch des Aristoteles, Heidelberg
1911, p. 36.
Here we read "ferumen acer."^ As the Graeco- Latin

glosses give "ferrumen ffTdfzwfia,"^ and for aTo/itofjia
also stands "ferrum durum, "^ it follows that the un-
usual gloss in the Codex Sangallensis 912, namely "acer
durus, "* is due to an abbreviation for "ferrum durum,"
the glossator having naturally taken acer to mean
"hard." The early Anglo-Saxon glosses give "acce-
arium steli," which concludes the proof that we are
dealing here with an Arabic word. Outside the Anglo-
Saxon glosses, the word is found only in Graeco-Latin
vocabularies and in the Liber glossarum where it is given
as aciare and is glossed "ferrum durum, "^ whence it
also found its way into the Codex Vaticanus 3321.^
Thus we find that French acier, Spanish acero "steel"
are of Arabic origin.''
' Goetz, op. cit., vol. II, p. 580.

» Ihid., pp. 71, 438.
»
lUd., vol. Ill, pp. 204, 368.
* Ibid., vol. IV, p. 202.
6 Ihid., vol. V, p. 162.
• Ihid., vol. IV, p. 6.
' As K. Hofmann (Archiv fiir lateinische Lexikographie, vol. II,
p. 275)
assumed that the Romance word for "steel" came from the Festus gloss for
necessary to point out the fact that acieris is a ghost word.
acieris, it is In
Festus there is an entry "acerra ara, quae ante mortuum poni solebat, in qua
odores incendebant. Alii dicunt arculam esse turariam, scilicet ubi tus
reponebant." Here ara is unquestionably a miswritten area, which Paulus
found in a bad copy of Festus, and so doubled up the lemma, for it does not
appear from any source that altars called "acerra" were placed before the
dead. In the glosses we have only "area turaria" or "uas (far) quod sacri-
ficiis adhibetur." Acerra is generally missjêlled accersa, acersa. Some
gloss must have had "accersa area quae sacrificiis adhibetur," which was
read "accersa aerea," etc., and gave rise to an interpolation in Festus,
"acceris securis aerea, qua in sacrificiis utebantur sacerdotes." In the AA
glosses (G. Goetz, vol. V, p. 436) the two glosses follow each other:
Acerra alcolatoria uel turibulu.
Acersu securis quam fiamines aut pontifices habent.
The two glosses are also in Codex Sangallensis 912 (G. Goetz, vol. IV,
p. 202):
Acerlis securis quam flaminei subpontificis habebant
Acersa arculatoreania,
and the gloss agrees with Festus' lemma for acceris, and the Graeco-Latin
gloss "d|/>Tj Upo<t>i,vTov, in IIXoCtos" {ihid., vol. II, p. 13), where the latter,
most likely, should have been "is HaOXos." When the glossator found in a
We have Arabic >^'' 'anlt, cr*' 'unta, with the

nominative ending 'anltun, for the soft iron of the
sword blade. Although not so good for the edge as the
"male" iron, it was better than iron and became more
popular in the East, hence we have Votyak andan, Os-
setic andun, Beduye enti "iron, steel. " In LLatin it be-
came andanicum, the ondanique of Marco Polo. In
Beauvais* alidena, Bacon's andena we have still earlier
forms. We have already seen that in Spain hindl be-
came substituted for •^'^ 'anlt, hence Spanish alinde,
alfinde, alhinde is used for "steel."
Wehave seen that the Anglo-Saxon gloss for "ac-
cearium" is steli. The OHGerman glosses have for
this stahal, stahel, stahl, and that these are borrowings
follows from the fact that we have also Coptic stall
"steel. " To trace the history of this word we shall be-
gin with its Sumerian prototype. We
have Sumerian
za-gin "shining, lapis lazuli," literally "shining stone."
We have also the uncompounded gin "shining."
That gin is an original root follows from its Dravidian
correspondent, where kan has the fundamental mean-
ing of "shining," hence Kannada kani "glow," kan
"the eye," kannadi "mirror," Tamil kanja "mirror,
a pane of glass. " The Assyrian borrowed the root gin
in its ugnu, uknu "shining, precious stone, crystal,
copy of Festus "acerra aerea," he hunted through the vocabularies for an
explanation of the word, and found it in the Placidus glosses, where the
etymology of "heros" is given, namely:
"Heroes, dicuntur aerii uel celo digni. id uel fortes uel sapientes.
ab area, id iunonem. quam aerem dicunt esse ubi regnum. et sedes. ani-
marum est. ut aeris in campis latis. et cicero. in sommio. scipionis. ergo hie
heros huius herois. huic. hero. hunc. heroem. ab hac heroem. mulier uero.
heroin, e. uel heroadas. aut eroas. ut. lemnias
"Hec securis dicimus. huius securis. huic securi. hanc securem. o securis.
ab hac securi. et pluraliter. secures, o secures, ab his securibus. secur. nus-
quam legimus," {ibid., vol. V, p. 108, and again, p. 24 f .).
The sequence "hero" and "haec securis" made the stupid glossator assume
that the axe had something to do with "aeree," which he took to be "made
of brass," and thus arose the interpolated gloss in Festus, where accersa,
aeeersu, aecerlis of the glosses at last become acceris and accieris.
glass, "^ and we find it in Chinese as king "metallic

mirror, to sMne. " The Dravidian word led to San-
skrit kdric, kac "to shine," kacas "crystal, glass."
The Sumerian word za-gin must also have existed
as za-gagin or za-gigin, for it is a peculiarity of
Sumerian "bright" words that they appear redupli-
cated, such as babara, dadaga. The Sumerian word
entered early into the Semitic languages, for we have
Hebrew 2T zag "the transparent grape-skin," nST
zkh "to be pure," ^31 zakku "pure, shining." The
Sumerian iag, sig, dag "shining" words may them-
selves be apocopated za-gin; the Hebrew words and the
Aramaic KDT zka, KDT dka may be derived from the
apocopated forms rather than from za-gin. But Arabic
^WJ zigag, zagdg, zuâ^ "glass," J^^^- saganâl,
(^y^j zaânâl "mirror" are obviously derived from

the Sumerian. As may be expected these Arabic
words are found in Gothic and OHGerman. In Gothic
we have skuggwa, in OHGerman scuchar "mirror."
It has been suggested that the Arabic sagangal may
have been a corruption of Latin speculum. While the
above discussion shows that we have to go back to the
Sumerian for the origin of the word, a contamination
with the Latin word is not excluded. We have the
Arabic J^ "he furbished, polished," which
âqala
may well have arisen from J*^ sfaqala, misread
§qaqala, since the diacritical marks were often confused,
and reduced to qaqala. This word is also found in
Syriac and in the Talmud as vj?''? slqll "a furbisher,"
but here it is a S.na^ hyofisvov and not reliable. That
there was also an Arabic form staqla or stala ap-
pears from the languages that have borrowed from the
1 H. V. Hilprecht, Zur Lapislaeuli-Frage im Babylonischen, in Zeiischrift
fiir Assyriologie und verwandte Gebiete, Berlin 1893, vol. VIII, p. 188.
Arabic. We
have not only Hindustani sakeld "a kind
of steel," but also Coptic stelli "to shine," by the side
of stall "steel," and neither is explicable from its own
language. But we have also Gothic stikls "beaker,"
OHGerman stahel, OPrussian stakla "steel," OBul-
garian stiklo "glass," Russian stakan "beaker."
(e) Greek nXdvrjz.
Arabic 'jA* filizz, filazz, fuluzz means "white

copper, whereof are made cooking pots of large size,
and mortars in which substances are pounded," and
the Persian dictionary adds "ore, metal in general,
dross, scoria, a stone." The plural '^^'j^ filizzatu
signifies "the seven metals, gold, silver, copper, iron,
lead, tin, and tutenag. "^ The relation of "white
copper" to the seven elementary metals would not ap-
pear clearly, if we did not have the explanation in the
earliest Arabic lapidary. This copper alloy is called in
Persian haftâuS, that is, "prepared from the seven
metals."^ This establishes the relation of the two,
and at the same time makes it possible to identify the
Arabic word flzz. In the Greek alchemy the seven
metals were identified with the seven planets, the signs
of which are the alchemist's expressions for these
metals, hence Greek nXdvrjz, nkdvr^Tsi; must have come
to mean "metal, ore, stone." In Arabic it would ap-
pear as ltA bins, or j^. blnz, or j^ flnz. The early
transcriptions into Arabic are so bad that almost any-
thing may be expected there.^ The Arabic J^ has
' Ruska (op. cit., p. 61) identifies this metal called ^^w" j\>- h,ar Îniy as
a kind of red copper alloy imported from China.

2 Ihid., p. 60 f.
' Ihid., p. 55.

THE HISTORY OP COPPER AND IRON 19
degenerated to 3 jl* , written 'j^ fizz, which., pronounced

felizz, etc., became the appellation par excellence of
"ore, metal in general."
In the Hrabanian-Keronian glosses we have "berillus
genus saxi candidi (pisleht) pilent chunni felises scin-
andi. " This corresponds exactly to the Arabic defi-
nition of balaur, balur, etc., which, however, is not
Jj^.
"the beryl," but "crystal" or "flint." Taifasi says

that the balur is a kind of white borax, which has be-
come a white transparent stone.^ In the OHGerman
gloss the "stone" or "ore" is felis, a precise rendering
of Arabic 'j^ felizz. Felis is almost entirely absent

from the other Germanic languages, except as ONorse
fjall "mountain," where the very apoeopation shows
that it is not a native word.
OHGerman pilent is the same Arabic word but in a
different form. We have Arabic -^. balant "a stone
resembling marble, but inferior to it in softness."
This is, of course, again our Greek nXdvTjrez. The
reason for applying this name to crystal or flint is
obvious. According to Taifasi the crystal could be
melted like glass. Taifasi here quotes Theophrastus,
who says that the best glass was produced from crystal
or flint mixed with copper,^ and it is well known that
the famous green glass was produced by a mixture with
copper compounds.^ But crystal or glass was fre-
quently indentifled with the moon, which usually stood
for silver, or with Jupiter, which usually stood for
bronze or tin. It was the shining "metal" par excel-
lence, the "fire-stone," which originally did not mean
"flint, " but a shining substance, that is, crystal. That
' M. C16ment-Mullet, Essai sur la mineralogie arabe, in Journal Asiatiqw,
Paris 1868, series VI, vol. XI, p. 231.

a IMd.,
p. 235.
' Von Lippmann, op. ciL, p. 102.
—
the Arabic ^ balant is only a variation of a form -^

flnt or y^ shown by the early ASaxon gloss
flnz, is
" petraf ocaria flinta," and the OHGerman gloss
"silexîns. " Hence OHG. pilent is only a variation of
Arabic -^. balant, and has the meaning of "crystal"
or "flint."
The natural crystal was supposed to be defective, and
already Pliny called it a "blind" stone/ because it did
not reflect the sun's rays, and he said that the eyes of
certain animals, such as the panther, emitted a bright
light not unlike that of the beryl. According to him
also the beryl, that is, the crystal or flint, was good for
diseases of the eye. This conception is contained in a
highly interesting series of Germanic glosses. The
Keronian glosses read "hyena bestia cuius pupille
"^
lapideae sunt staraplint des seha augono stani sint.
This is a slight change of Isidore's "hyaenia lapis in
oculishyaenae bestiae invenitur,"^ which is also in
Pliny: "hyaeniae, ex oculis hyaenae lapides. "* The
Latin soiirce is based on a blunder. In Egyptian
medicine the eye-salve hetem is frequently referred to.
This hetem is identical with Greek xadfiia, the eye-salve
par excellence,^ but Egyptian hetem also means
"hyaena," hence the assumption of the hyaena stone
found in its eye. In the Cassel glosses of the IX.
century we read "albios oculos staraplinter," where
staraplint obviously means some kind of blindness
"white eyes." The Latin albios oculos produced the
French aveugle, and staraplint led to German Staar
"cataract of the eyes," but it can be shown that this is
1 See my discussion of the beryl in CoMnbulions, vol. IV, p. 114 ff.
^ E. Steinmeyer und E. Sievers, Die alihochdeutsehen Glossen, Berlin
1879, vol. I, p. 170.
'XVI. 15. 25.
* XXXVII. 168.
' H. Hiring, Die &ber die medieinischen Kenntnisse der alten Agypter
berichtenden Papyri, Leipzig 1888, p. 91.
a mere misunderstanding of the particular disease which

produces blindness, for in the ASaxon Corpus glosses
we have " scotomaticus staerblind," and Greek axorofxa
is a blindness produced by vertigo, hence ASaxon
staerblind, OHGerman starplint, which means "hyaenia
the vertigo stone. " We have already
lapis, " is really "
seen that the white stone is pilent, from the Arabic.
Similarly OHG. Arabic J-^ sadar "vertigo."
star is
Just as albios oculos has led to French aveugle, the blint of
OHGerman staraplint, ASaxon staerblind has produced
Germanic blind "blind," while OHG. starblind assumed
the meaning "glaucoma" instead of "scotoma,"
leading to German Staar "cataract of the eye."
Just as the beryl was the "shining stone," so bronze
was the "shining metal" par excellence, hence Arabic
'j^ filizz referred more particularly to "white copper."
But by the this we have found Arabic -^
side of
balant, and Armenian we actually have plindz
in
"copper, brass." In Persian we have, side by side,
birindz, piring "brass, copper."
beredz, In reality,
however, this not the origin of the name for "bronze."
is
Copper was in Greek alchemy called by the planet
Venus, namely iippodirrj. This became in the Syriac
alchemy afrodltl,frodltl, which also referred to "bronze,"
but for this more usually the Syriac name bilatl was
used. The accidental resemblance of the triliterals
&Zf, frd with the derivatives from Greek nXdvrjz led to
the identification of "bronze" as the "chief" planet,
hence we have Arabic 'j^ fizz by the side of Armenian
pirindz, etc. At the same time Coptic barot, Ethiopic
bert "copper, brass" bear witness to the "Venus"
origin of the metal.
After discussing the silver mirrors, Pliny says:
"However, to finish our description of mirrors on the
—
present occasion the best, in the times of our ances-
tors, were those of Brundisium, composed of a mixture
of stannum and copper."^ That is clearly an inter-
polation, since there is no reference before to mirrors
made of tin and copper. Further on, while speaking of
tin, the following passage occurs: "Stagnum inUtum
aereis vasis saporem facit gratiorem ac compescit virus
aeruginis, mirumque, pondus non auget. specula etiam
ex eo laudatissima, ut diximus, Brundisi temperabantur,
donee argenteis uti coepere et anciUae. "^ This runs
in Isidore as follows: "Stagnum inlitum aereis vasis
saporem facit gratiorem et conpescit virus aeruginis.
Specula etiam ex eo temperantur. "' Here again the
interpolation is seen from the fact that Isidore, who
quoted just before Pliny's account of lead in full, has
nothing whatsoever about the Brundisian mirrors
which were manufactured, but speaks only of mirrors
which are manufactured. Not a word is found about
such a composition until the end of the VIII. century
when "compositio Brandisii''^ or " Brindisii" is found
in two metallurgical recipes.* As this alloy is com-
posed of copper, tin, lead, and glass, in somewhat the
manner mentioned in PHny, there can be little doubt
that we have here a haftâuS, hence a confusion with
Greek TiXdvYjc;; hence we get, originally through the
Arabic, Greek ppovn^mov, English bronze, etc.
Forstemann* has observed that derivatives of
berhta in names hardly occur among Goths, Vandals,
Frisians and Normans, are rare among Saxons, but very
common among Anglo-Saxons, Langobards, Franks and
Bavarians. He also notices that the name does not seem
to occtir at all before the VI. century. When we now
1 "Atque ut omnia de speculis peragantur in hoc loco, optima
aput maiores
fuerant Brundisina, stagno et aere mixtis," XXXIII. 130.
2 XXXIV. 160.
' XVI. 23. 2.
* Von Lippmann, op. cit, p. 561.
' AUdeutsches Namenbuch, Bonn 1900, vol. I, col. 277.
THE HISTORY OP COPPER AND IRON 23
further observe that Goth, bairhts "bright, shining"

is used only very rarely as an adjective in OHGerman,
namely peraht "clarus, " and has survived only in
Ger. pracht "splendor," while ONorse bjartr "clarus,"
birta "to illuminate," like Goth, bairhtjan "to reveal,"
are only derivatives of the Goth, bairhts or OHG.
peraht, and ASaxon beorht does not even occur in the
early vocabularies, the suspicion is at once roused that
berht is not a common Germanic word, but due to some
borrowing. It is useless to adduce a Sanskrit bhrag
"to shine" as the prototype of this word, because
Semitic braq is much nearer, and the same root may be
discovered elsewhere, but particularly because the word
appears in the Germanic languages exclusively in the
form berht, and not berh.
was a thin gold lamina,
Brattea, originally bractea,^
which was variously used as an ornament. Pliny tells
of jewels made of such laminae stuffed with a lighter
material,^ and of a cheap imitation of such bratteae,
made of brass or bronze, generally used for actors'
crowns.^ The vast number of bracteates in Scandi-
navia and on the continent, the rich garments adorned
with gold and bronze laminae, found in the south of
Russia, bear witness to the fact that bractea was identi-
cal with "rich adornment, splendor." This at once
shows that the Latin word is derived from the Semitic,
most probably from Phoenician. We have Hebrew
pis bdraq, Syriae s>s braq, Arabic i3a baraqa "to
' See Rheinisches Museum, vol. XLV, p. 495.
* "Alii bratteas infercire leviore materia propter casum tutius gemmarum
soUicitudini putant," XXXIII. 25.
' "Nunc praevertemur ad differentias aeris et mixturas. in Cyprio
[coronarium et regulare est utrumque ductile] coronarium tenuatur in lamnas
taurorumque felle tinctum speciem auri in coronis histrionum praebet,
idemque in uncias additis auri scripulis senis praetenui pyropi brattea
ignescit. regulare et in aliis fit metallis, itemque caldarium. differentia
quod caldarium funditur tantum, malleis fragile, quibus regulare obsequitur
ab aliis ductile appellatum, quale omne Cyprium est," XXXIV. 94.
shine, Talmudic '^CRI^ barqta, Syriac pja
glitter,"
bdrqd "emerald," and the Romans were acquainted

with the word at least through Hamilcar's name,
Barcus "the lightning."
The Romans already had identified bractea with
"splendor," more especially "tinsel splendor," hence
while brattealis for "gilt," others
Prudentius used
employed bratteatus for "beautiful, superb," but also
"tinsel, simulated." The glossaries^ wrote brattea,
bratthea, brattanea, brantia, branzia and glossed these
words with "lamina aurea. " The forms brantia,
branzia arose from a confusion with brando "torch,
flashing sword."
There is a common gloss "vibrantia iacula ful-
gentia,"^ which in the Corpus, Vatican 3321, Abavus,
and Asbestos glosses reads bibrantia, while the Abactor
Anglonianus II glosses read for it brantia. Here the
"flashing weapon" was once more confused with brando
"torch, flashing sword." The earliest reference
to brand is found in Gregory the Great, who says
that the Latins were not in the habit of using parts of a
saint's body as a relic, but the shroud, brandeum, in
which he was buried,^ and from the Merovingian
writers we learn that the saint's shroud was red.*
At a later time brandeum also referred to a belt or other
garment, and from the Edict of Leo the Wise we learn
that the npavdwnpdzcu were Syrian cloth merchants who
^ Goetz, op. ciL, sub brattea.

^ Ibid., sub vibrantia.
'"Romania consuetude non est, quando sanctorum reliquias dant, ut
quicquam tangere praesumant de corpore. Sed tantummodo in buxide
brandeum mittitur, atque ad sacratissima corpora sanctorum ponitur,"
Epistolae, IV. 30.
' "Involutum est corpus illius integrum, sed exsiccatum, de brandeo
rubeo," Scriptores rerum merovingicarum, vol. Ill, p. 321; "ipsum corpus
sanctissimum, sicut et in anteriori translatione, ab episcopis Remorum
dioceseos integrum inventum est et brandeo rubeo involutum," ibid., p. 326.
.
imported garments from Syria, silk stuffs from Selencia,

and Saracenic and Bagdad cloth.. ^
We are, therefore, sure that brandeum was some cloth
from the east. Indeed, we have Persian par and "the
glittering surface of a polished sword, a kind of fine
figured, painted silk, plain silk," but also "a bird,"
which makes it sure that we have here a popular
etymology, a participial derivative from par "a wing,"
from which we also get pari magas, literally "fiy's
wing," but now "the brilliancy of the sword, a delicate
kind of silk stuff." It is clear that Greek npdvdcov,
LLatin brandeum are derived from the Persian, but the
Persian parand is itself a corruption of parniyan "a
kind of fine painted China silk, also garments made of
it, shroud for a royal corpse," and this is from Chinese
pin-lien " to bury in a shroud, " which brings us back to

the beginning of the Christian era, when, according to
Pliny, Chinese silks were introduced into Rome.
The silk, to judge from the connotation this brandeum
took, was "watered" or "shot, " and therefore led to the
designation of "undulating surface," as in a sword,
besides "torch" and "flame," on account of the pre-
vailing red color of this silk. Hence the Persian
parand, barand, read in Arabic as firind, barand, be-
came the name of Damaskeen steel, the best of which
came from India, but more especially from China.^
In Anglo-Saxon we find early brand "sword," which
is obviously the Persian and Arabic word. But in
Persian and Arabic it means "the glittering, flashing
sword," hence both in Anglo-Saxon and Old High
German brand means "torch, flame," even as we have
side by side Italian brando "sword," OFrench brandir
"to brandish," that is, "to flash." very much How
the Arabic or, possibly, somewhat earlier, the Persian
' J. Nicole, L'edit de Verrvpereur Lion le Sage, in Mimoires de I'lnstitut
national genevois, vol. XVIII, p. 29 f
* Von Lippmann, op. cit, p. 399.
word influenced the connotation of the word in the
Germanic languages, is seen from the ASaxon gloss
"rubigo brond om," where "rubigo" means either
"rust" or "erysipelas," for we have also the ASaxon
glosses " oman, homan erysipelas," omig "rusty, in-
flammatory," omiht "full of inflammation," ONorse
dmr "loathsome, black," dma "erysipelas," all of which
are from Arabic 'j-^ humrah "erysipelas, a certain dis-
ease which attacks human beings, in consequence of
which the place thereof becomes red," j-^ humar
"redness, anthrax, erysipelas." Similarly the Arabic
bar and "flashing sword" has led to ASaxon brand
"erysipelas." And thus was evolved the idea of
"burning" in connection with brand, which entered all
the Germanic languages. By the side of Arabic •*'j'.
bar andwe have also *' J firind, hence there must also
have existed a form birind, and this is preserved in
Gothic brinno "fever," from which the verb brinnan
"to burn" is formed. The variant ASaxon beornan,
byrnan, ONorse brenna " to burn" fm-ther show that the
verb is not originally Germanic.
(f) OHGerman aruz.
The Mahabharata, Herodotus, and the Mongols and

Tibetans tell of gold which was mined by ants,^ and
Lauf er has shown that the reference is to gold mined in
the Altai region by Tatar or Mongol tribes, bearing
the name of "ants" in the Tatar language. It is not
necessary to subscribe to Laufer's philological identi-
fication, but there cannot be the slightest doubt that the
Indians and Greeks distinctly referred to a kind of gold
' B. Laufer, Die Sage von den goldgrdbenden Ameisen, in T'oung Poo,
series II, vol. IX, p. 429 ff.
.
whieli came from the mountains of the interior of Asia.

Now the Assyrians have a -word hurdsu "gold," while
harigu means "ditch," haragu "to dig." Hurctgu is
not the usual word for "gold" in the Semitic languages,
but the Hebrew has pT? harug "dug up, gold," and
there is also recorded Phoenician harag "gold."
Greek xf"^'^°'^ refers to the same
It is obvious that
mined gold and of Semitic origin.
is Strabo quotes
Polybius to the effect that "mined gold," ;fjOUffoc dpuxvo^,
was found in mines, dpuyfiara, and that this differed
from wash gold in being partly impure and containing
useful mineral ingredients.^ It has long been sus-
pected that Greek dpuaam is no other than Assyrian
hardgu,^ and this makes the expression ;f/>u<Toc dpoxToz
merely a tautology. The Assyrian haragu "to dig"
is merely an extension of the root har, for we find also
hurru "hole, ravine, cave," harm "canal," haru, Mru
"to dig," haraku "to engrave, cut," hiritu "ditch,
canal."
Ball has pointed out that the Sumerian ideogram for
"gold" is identical with that of the Chinese, which was
pronounced kin.^ De Lacouperie thinks that the
identity of the two ideograms is, among others, a proof
of the derivation of the Chinese civilization from the
Sumerian.* But in Sumerian we have a compound
guskin, whereas in Chinese kin merely means "metal,"
so that a denominative, apparently meaning "yellow,"
must have dropped out in Chinese. In Japanese we
have ko-gane, literally "yellow metal," and Sumerian
gus-kin stands for ku-kin, literally "shining metal."
1 IV. 6. 12.
" J. An Hebrew and English Lexicon, London 1813, p. 242; L.
Parkhurst,
Meyer, Handbuch der grieehischen Etymologie, Leipzig 1901, vol. Ill, p. 323;
Zeitschrift der deutschen morgenliindischen Gesellschaft, vol. XXXIII, p. 327;
The Annual of the British School at Athens, vol. VIII, p. 144.
3 Proceedings of the Society of Biblical Archaeology, vol. XIII, p. 84 f
* Western Origin of the Early Chinese Civilisation, London 1894, p. 85.
Chinese hwang "yellow" was originally kung or gung,
and this, with Sumerian ku, gu§, indicates an original
kur, that is, we come back to a form resembling Assyrian
hur "to dig." That hwang "yellow" in Chinese had
something to do with "to dig," follows from the com-
position of its ideogram, one part of which was "a
field."
From the Assyrian or another Central Asiatic lan-
guage the word for gold spread into all directions. We
have Sanskrit hiranyam "gold," haris "yellow," etc.,
and similarly Avestan zaranya, zarona "gold," zaray
"golden, yellow," Persian zer, zerin "gold," zerlr, zirlr
" yellow dye wood, " zerd "yellow." The derivations
in the other Indo-European languages are well under-
stood. In Tibetan "gold" is gser, possibly borrowed
from the Persian. In the Turkish languages we have
forms like altun, alcin, but al means "yellow" and is
a variation oi jal, zil, kil, 6il "yellow," also represented
by jar, sar, zar,^ hence we have here once more re-
lationships to Assyrian har.
Pliny has the following account of the mining of
gold: "Gold is found in our own part of the world;
not to mention the gold extracted from the earth in
India by the ants, and in Scythia by the Griffins.
Among us it is procured in three different ways; the
first of which is, in the shape of dust, found in running
streams, the Tagus in Spain, for instance, the Padus in
Italy, the Hebrus in Thracia, the Pactolus in Asia, and
the Ganges in India; indeed, there is no gold found in a
more perfect state than this, thoroughly polished as it is
by the continual attrition of the current.
"A second mode of obtaining gold is by sinking shafts
or seeking it among the debris of mountains; both of
which methods it will be as well to describe. The
' H. Vdmbfiry, Etymologisches Worterbuch der turko-tatarischen
Sprachen,
Leipzig 1878, pp. 11, 114, 117.
'
persons in search of gold in the first place remove the

'
segutilum, such being the name of the earth which
'
gives indication of the presence of gold. This done, a

bed is made, the sand of which is washed, and, according
to the residue found after washing, a conjecture is
formed as to the richness of the vein. Sometimes,
indeed, gold is found at once in the surface earth, a
success, however, but rarely experienced. Recently,
for instance, in the reign of Nero, a vein was discovered
in Dalmatia, which yielded daily as much as fifty
pounds weight of gold.
' The gold that is thus found
in the surface crust is known as 'talutium, in cases '
where there is auriferous earth beneath. The moun-

tains of Spain, in other respects arid and sterile, and
productive of nothing whatever, are thus constrained
by man to be fertile, in supplying him with this precious
commodity.
"The extracted from shafts is known by
gold that is
some persons as canalicium, and by others as can-
'
'
'
aliense;' it is found adhering to the gritty crust of

marble, and, altogether different from the form in
which it sparkles in the sapphirus of the East, and in the
stone of Thebais and other gems, it is seen interlaced
with the molecules of the marble. The channels of
these veins are found running in various directions
along the sides of the shafts, and hence the name of the
gold they yield
—
canalicium.
' In these shafts, too,
'
the superincumbent earth is kept from falling in by

means of wooden pillars. The substance that is ex-
t^cted is first broken up, and then washed; after
which it is subjected to the action of fire, and ground to
a fine powder. This powder is known as apitascudes,
'
while the silver which becomes disengaged in the fur-

nace has the name of sudor given to it. The impurities
'
'
that escape by the chimney, as in the case of all other

metals, are known by the name of scoria. In the case
'
'
of gold, this scoria is broken up a second time, and
melted over again. The crucibles used for this purpose
are made of 'tasconium, 'a white earth similar to
potter's clay in appearance; there being no other sub-
stance capable of withstanding the strong current of
air, the action of the fire, and the intense heat of the
melted metal.
"The third method of obtaining gold surpasses the
labours of the Giants even: by the aid of galleries driven
to a long distance, mountains are excavated by the light
of torches, the duration of which forms the set times for
work, the workmen never seeing the light of day for
many months together. These mines are known as
arrugiae and not unfrequently clefts are formed on a
;
'
'
sudden, the earth sinks in, and the workmen are crushed
beneath; so that it would really appear less rash to go
in search of pearls and purples at the bottom of the sea,
so much more dangerous to ourselves have we made
the earth than the water! Hence it is, that in this
kind of mining, arches are left at frequent intervals for
the purpose of supporting the weight of the mountain
above. In mining either by shaft or by gallery,
barriers of silex are met with, which have to be driven
asunder by the aid of fire and vinegar; or more fre-
quently, as this method fills the galleries with suffocat-
ing vapours and smoke, to be broken to pieces with
bruising-machines shod with pieces of iron weighing one
hundred and fifty pounds: which done, the fragments
are carried out on the workmen's shoulders, night and
day, each man passing them on to his neighbour in the
dark, it being only those at the pit's mouth that ever
see the Ught. In cases where the bed of silex appears
too thick to admit of being penetrated, the miner traces
along the sides of it, and so turns it. And yet, after all,
the labour entailed by this silex is looked upon as
—
comparatively easy, there being an earth a kind of
potter's clay mixed with gravel, 'gangadia' by name,

which, it is almost impossible to overcome. This earth
has to be attacked with iron wedges and hammers like
those previously mentioned, and it is generally con-
sidered that there is nothing more stubborn in existence
— except indeed the greed for gold, which is the most
stubborn of all things.
"When these operations are all completed, beginning
at the last, they cut away the wooden pillars at the
point where they support the roof: the coming downfall
gives warning, which is instantly perceived by the
sentinel, and by him only, who is set to watch upon a
peak of the same mountain. By voice as well as by
signals, he orders the workmen to be immediately
summoned from their labours, and at the same moment
takes to flight himself. The mountain, rent to pieces,
is cleft asunder, hurling its debris to a distance with a
crash which it is impossible for the human imagination
to conceive; and from the midst of a cloud of dust, of a
density quite incredible, the victorious miners gaze
upon this downfall of Nature. Nor yet even then are
they sure of gold, nor indeed were they by any means
certain that there was any to be found where they first
began to excavate, it being quite sufficient, as an in-
ducement to undergo such perils and to incur such vast
expense, to entertain the hope that they shall obtain
what they so eagerly desire.
"Another labour, too, quite equal to this, and one
which entails even greater expense, is that of bringing
rivers from the more elevated mountain heights, a
distance in many instances of one hundred miles
perhaps, for the purpose of washing these debris. The
channels thus formed are called corrugi, from our word
'
'
'corrivatio, I suppose; and even when these are once

'
made, they entail a thousand fresh labours. The fall,

for instance, must be steep, that the water may be pre-
—

cipitated, so to say, ratherthan flow; and it is in this
manner that it is brought from the most elevated points.
Then, too, vallies and crevasses have to be united by
the aid of aqueducts, and in another place impassable
rocks have to be hewn away, and forced to make room
for hollowed troughs of wood; the person hewing them
hanging suspended all the time with ropes, so that to a
spectator who views the operations from a distance, the
workmen have all the appearance, not so much of wild
beasts, as of birds upon the wing. Hanging thus
suspended in most instances, they take the levels, and
trace with lines the course the water is to take and thus
;
where there is no room even for man to plant a foot-

step, are rivers traced out by the hand of man. The
water, too, is considered in an unfit state for washing, if
the current of the river carries any mud along with it.
The kind of earth that yields this mud is known as
'urium;' and hence it is that in tracing out these
channels, they carry the water over beds of silex or
pebbles, and carefully avoid this urium. When they
have reached the head of the fall, at the very brow of
the mountain, reservoirs are hollowed out, a couple of
hundred feet in length and breadth, and some ten feet
in depth. In these reservoirs there are generally five
sluices left, about three feet square; so that, the moment
the reservoir is filled, the floodgates are struck away,
and the torrent bursts forth with such a degree of
violence as to roll onwards any fragments of rock which
may obstruct its passage.
" When they have reached the level ground, too, there
is stillanother labour that awaits them. Trenches
—
known as 'agogae' have to be dug for the passage of
the water; and these, at regular intervals, have a layer
of ulex placed at the bottom. This ulex is a plant like
rosemary in appearance, rough and prickly, and well-
adapted for arresting any pieces of gold that may be
'
carried along. The sides, too, are closed in with

planks, and are supported by arches when carried over
steep and precipitous spots. The earth, carried on-
wards in the stream, arrives at the sea at last, and thus
is the shattered mountain washed away; causes which
have greatly tended to extend the shores of Spain by

these encroachments upon the deep. It is also by the
agency of canals of this description that the material,
excavated at the cost of such immense labour by the
process previously described, is washed and carried
away; for otherwise the shafts would soon be choked up
by it.
"The gold found by excavating with galleries does

not require to be melted, but is pure gold at once. In
these excavations, too, it is found in lumps, as also' in the
shafts which are sunk, sometimes exceeding ten pounds
even. The names given to these lumps are palagae, '
and 'palacurnae, while the gold found in small grains

'
is known as 'baluce. The ulex that is used for the

'
above purpose is dried and burnt, after which the ashes

of it are Washed upon a bed of grassy turf, in order that
the gold may be deposited thereupon.
"Asturia, Gallaecia, and Lusitania furnish in this
manner, yearly, according to some authorities, twenty
thousand pounds weight of gold, the produce of Asturia
'
forming the major part. Indeed, there is no part of the

world that for centuries has maintained such a con-
tinuous fertility in gold. "^
Here at least the majority of the technical terms are
interpolated. In Spanish mining the terms segullo,
alutacion (for alutium or talutium), sudor, tasconio,
arrugia, ganga, palaca are used just as in Pliny, but they
1 XXXIII. 66-78.
are unquestionably late borrowings and so are of no
avail for the determination of the corresponding terms.
Similarly baluce "gold in small grains" is of no avail,
since it had long been in use among the Romans, having
originally been borrowed from a Semitic language.
The Greek gloss " xPâafi[io(; halluca" shows that the
original meaning was "gold-bearing sand." We have
Sanskrit vdlukd "sand, gravel," which cannot be ex-
plained from the Sanskrit. But Arabic "^j^. haluqah
"sandy desert" the same as Assyrian balaqu "to lay
is
waste," and this is but an extension of the Semitic bal

"destroy" root. From the Latin text of Pliny "pala-
gas, alii palacurnas, iidem quod minutum est balucem
vocant, " it does not clearly appear whether the first
two are the same as the third or refer to larger nuggets,
but in any case they seem to be mere variations of
balux or baluce.
Segutilum appears in some manuscripts as segullum.
This is unquestionably Latin sigillum, which in the
Arabic form in the Koran assumed the meaning of
"book," and means "judgment, transcript."
also As
the verb J^ sagala also means "to verify," it is clear
that the noun also meant "verification," hence, as we
find in Pliny, "indicium."
Gold found on the surface is in Pliny called talutium,
forwhich some manuscripts read alutatium, alutationem,
alutiatum, which produced Spanish alutacion. In
XXXIV. 157 the gold mines in which lead is also found
are called alutiae. Here we fortunately can see that
the word is an interpolation, because the sentence in
Pliny is distinctly an elaboration of the corresponding
passage in Isidore, where neither alutiae, nor eluente, to
explain the word, are to be found:
"Lavant eas harenas, "Lavant eas arenas, et

metallici et, quod subsedit, quod subsidit, quoquunt
cocunt in fornacibus. in- in fornacibus. Inveniun-
venitur et in aurariis metal- tur et in aurariis metallis
lis, quae alutias voeant, aqua missa calculi nigri et
aqua inmissa eluente cal- graves, et dum aurum col-
culos nigros pauUum can- ligitur, cum
eo remanent;
dore variatos, quibus postea separati conflantur
eadem gravitas quae auro, et in plumbum album re-
et ideo in calathis, quibus solvuntur, " Isidore, XVI.
aurum ooUigitur, cum eo 22.1.
remanent; postea caminis
separantur conflatique in
plumbum album resolvun-
tur," Pliny, XXXIV.157.
The Spanish term alutacidn "a bed of gold nuggets"
shows that alutiae, talutium mean "pure gold, as it
comes from the gold bed," that is, that we have here
Arabic u^^ halis "pure, refined (gold)," w'^' tahlis
"to refine (gold)," u^j^ halus "genuineness," etc.

The term seems to occur in a X. century Portuguese
document as the denomination of a pall, which is
mentioned by the side of "Greek" garments,^ which
are distinctly mentioned as being made with gold
thread.^ However, this particular aluz may be a
different spelling of alguexi, alvexi, alveice, from Arabic
cT'J wasl "a delicate silk fabric, generally embroidered

with gold thread."
' "Palae greciscas, duas palas de aluz. omnique casula piscina, glisissas,
tres de aluz," (959), Portugaliae monumenia diplomatica, vol. I, p. 47.
" "Dono etiam frontales, pallas, acitaras auro textas, grecirias varias,"
(812), Espana sagrada, vol. XXXVII, p. 317; "pallas lineas greciscas II,
cum auro porto paratas. item palla linea I polimita, et ipsa cum auro porto
parata," (IX. c), Zeitschrift fiir die Geschiehte des Oberrheins, vol. IV, p. 250;
"mantos duos auri frissos, alio alquexi auro texto, cum alio grieisco in
dimisso cardeno: casulla aurifrissa, cum dalmaticis duabus aurofrissis: et alia
alvexi auro texta," (1063), Espafia sagrada, vol. XXXVI, p. clxxxix.
" "

Pliny's gangadia "gravelly earth, a kind of potter's
clay mixed with gravel" is written in some manuscripts
gandadia, gandeda, Candida, and Arabic "^-^ qaddatis
"land in which are pebbles. " It has been adopted into

Spanish as ganga, whence it passed into French and
English as gangue "the earthy or stony matter in a
mineral deposit.
Urium is given as "the land which carries mud."
This Arabic J^ haur, plural jj''j>- hu'wur, "low or
is
depressed ground, the part in which the water flows

from the two sides of a valley."
Arrugia is "mine," while corrugium is given as "arti-
ficial rivers. " They are clearly the same word and un-
questionably from Arabic oj- fiarq "aqueduct,"
'^.J- harlqah, plural oj- huruq, "a channel of water,
a low or depressed tract of land," words in use in the
Arabic documents in Sicily. In the Spanish documents,
beginning with the VIII. century, we frequently find
arrogium for "water-course, brook," which leads to
Spanish array o "brook, " array ar "to water, inundate.
No doubt, Provengal arrazar, French arroser "to
water" are formed from the same Spanish word which
in Proven5al was popularly etymologized as from ros
"dew."
Iso Magister, who died in 871 at St. Gall, glossing
Prudentius' Apatheasis, 1. 724 f., "non sicut sculptor ab
aeris rudere decoctam consuescit viuere massam,
"^
says " rudere = mina a terra, vbi aes toUitur, " that is,
he identifies "rudus, ore" with mina, which he etymol-
ogizes as being derived from minare "to lead away,"
because the ore is taken away from the earth. This
etymology is as useless as Isidore's "aurum ab aura,"
' F. Arevalo, M. Avreli Clemenlis Prvdenti V. C. Carmina, Romae 1788,
vol. I, p. 465.
"
"ferrum dictum quod farra, id est semina frugum terrae

condat, " but it shows that mina "mine," hence
miner ale "mineral" were already known in 871. The
origin of this mina takes us back to dim antiquity. In
the old Egyptian texts we often find mend, mendt,
menit "an amulet worn to give physical happiness, an
ornament worn on ceremonial occasions," which was
made of turquoise, malachite, emeralds, or other similar
gems, and was in the form of a necklace, obviously be-
cause it was worn about the neck, especially by nurses.^
This Egyptian amulet, in the form of a necklace, was
early in use in the east and the west. To this bear
witness Latin monlle, Olrish muinde, muince, OWelsh
mind, Greek [j.dvvo<;, fiowoi;, ONorse men, ASaxon
mene, OHGerman menni, OBulgarian monisto, Avestan
minuS "necklace. " In Persian, mind has assumed the
meaning of the gems from which the necklace was com-
posed, "a glass globule or bead, a false gem of blue
color, bluestone, caustic, a stone resembling lapis
lazuli with which silver is tinged, vitriol, enamel, concha
veneris, a goblet, glass, " and has coincided with mind
"heaven, sky," which is from Avestan mendl "spirit,
heaven." In Arabic it has chiefly the meaning of
"enamel, alchemy," hence we get in the IX. century
LLatin mina "mine, shaft," as in Iso Magister, whence
we get later miner ale "ore, mineral."
Iso Magister confuses, as others do, "mine" with
"mineral," hence we should expect arrugia "mine,"
in the sense of "ore, " to get into Iso's gloss to Pruden-
tius' Peristephanon, II. 190, "effossa gignunt rudere,
where we get the equation "rudere = arizzi, vel stercora,
rudera dicunt maceriam ruinae. "^ The two words,
mina and arizzi, run through all the Germanic glosses of
' See H. Brugsch, Hieroglyphischêmotisches Worterbuch, Leipzig 1867,
p. 645, where a representation of it is given.
2 Arevalo, op. cit., vol. II, p. 905.
—"

Prudentius.^ In reality Iso, who obviously quoted
from an Arabic source, correctly rendered "stercora"
by Arabic <^JJ raut, plural ^hj^ 'arwdt, "dung."
Aruz later produced German Erz "ore."
This Arabic root had before that entered into the
Germanic languages. Arabic ^'-jj rautah means not
only "dung," but also "tip of the human nose,"
unquestionably from its discharge of mucus, while
Ibn-al-Awam regularly used '^jj 'rawata in the sense
of "to manure," so that it is clear that "rot" was one of
the connotations of the word in Arabic. This is well
brought out in one of the Prudentius glosses, "muccu-
"
lentis i. sordidis a mucca i. rotz, "^ where " mucculentus
is given as "filthy," hence related to the Arabic word.
But the lemma "a mucca" produced the OHGerman
roz "snot. " We have a still closer Arabic form in the
Keronian glosses, where "aerugo," that is, "rust, rot,"
isgiven as rost, in the Corpus glosses as rust. In the
Gothic Bible this word does not occur, but, instead, we
have nidwa, which is from the Arabic °-?-^' nadwah
"dampness, " in Alcala, "liento por unidad, rocio nedve.
Now we come to the most interesting word in Pliny
tasconium "a white earth from which pots are made."
In Greek medicine S^jpiou was a desiccating powder
used for the cure of wounds and ulcers. In Greek
alchemy it very early designated a powder used in the
1 Steinmeyer und Sievers, op. cit., vol. II, "rudere. stercore. al mina .i.
aruz," p. 383; "stercore al: nama aruz," p. 385; "mina .i. aruz," p. 386;
"arize," p. 396; "ris .i. molt," p. 402; "aruzi," p. 405; "ris .i. molta, ari
zin," p. 432; "arize 1 molto," p. 458; "ruris .i. molt," p. 480; "arvzze," p. 482;
"aruz I. stercora," p. 491; "mina. aruze," p. 493; "aruzz, arutz," p. 496;
"aerizze, arid stercora 1 molta," p. 505; "aerizze, arizce," p. 513; "aruzae," p.
526; "arice," p. 535; "arizzae," p. 555; "arizze," p. 560; "mina dicitur, arize,"
p. 565; "ant," p. 572; "mina dicitur. arize," p. 574; "aruze," p. 578; "arutos.
rudus mist," p. 586; "aruz," vol. IV, p. 93; "metalli aruzzes," vol. II, p. 420;
"aruzzin," p. 432; "ariz," p. 505; and "massam .i. mina. ariz," p. 578.
2 Ibid., vol. II, p. 382.
.
transmutation of metals, hence it became the philos-

opher's stone, in Arabic ^-j VI al-iksir, the elixir of life.
In Arabic medicine and alchemy the chief ingredient of

such a cure-all was o^ talq "a sort of medicament
which, when one is anointed therewith, prevents the
burning of fire," which was also taken to mean "a
plant that is used in dyes." The word is unquestion-
ably taken from the Coptic where taUo, also pronounced
talgo, means "medicine, cure. " This talq referred to a
number of substances, gypsum, mica, but especially
talc, which was supposed to be well adapted for pots,
because it withstood fire, as, indeed, such vessels were
often made of soapstone, a variety of talc, so called
from its greasy feeling. Pliny's tasconium is clearly
this Arabic talq.
Talq was readily etymologized as of Arabic origin,
since the root 6^ talaqa means "to become cheerful,
be freed from slavery or impediment, " in which sense it
is already found in the Koran. The Arabic word was
early adopted by the Germanic peoples. We have in
the Corpus glossary "fucus faex taelg," where taelg has
the meaning of "dye" and "sediment." The first
meaning was retained in ASaxon, while the second is
due to the fact that "the art employed in dissolving it
consists in putting it into a piece of rag with some peb-
bles and immersing it in tepid water, then moving it
about gently until it becomes dissolved and comes forth
from the piece of rag into the water, whereupon the
water is strained from it, and it is put in the sun to dry."^
We also find in the Corpus glossary "rediua aet-taelg,"
where "rediua" stands for "rediviva," and aet-taelg
literally means "again made cheerful, " and " propensior
tylg," that is, "more cheerful, more readily inclined."
» E. W. Lane, An Arabic-English Lexicon, London 1874, sub 3^

In the Gothic Bible we have similarly tulgjan "to make
safe, fortify," which corresponds to Arabic J^ tuluq
"not shackled."!
From the above investigation it follows that with the
appearance of the Arabs in Europe the metals assumed
a far greater importance in life than they had held here-
tofore. The intensive working of the gold mines, the
improved methods in the alloys of the metals, especially
of bronze, and the universal use of welded steel accord-
ing to the Indian method revolutionized metallurgy in
Europe. It cannot be said that the Arabs discovered
any new paths in this direction. They were rather the
popularizers of the Greek alchemy which heretofore had
been restricted to Byzantium and Egypt and to a small
number of men initiated in the mysteries of the craft.
The most surprising thing in the dissemination of the
Arabic metallurgy is the enormous Arabic technical
vocabulary which has left its impress upon the Romance
and Germanic peoples, even upon the Goths, whose
contact with the Arabs in the VIII. century was of the
closest. It behooves us now to investigate the type of
smiths that the Arabic invasion brought in its wake
into Europe and Africa, but, before doing so, we shall
study the history of the smith in Europe shortly before
the arrival of the Arabs, as it is revealed in language.
The Greek fjXexTpoz, OH German smaidar, was the
classical name of the Egyptian asem, a natural mixture
of gold and silver.^ This word, like Greek ijXsxrcjp "the
sun," is related to Sanskrit arkas "the sun," arcati
"to shine," which are taken from the Dravidian, Kan-
nada arka, akka, etc. "the sun," which is from the root
"to shine," preserved in Kannada ulku "to shine,"
Tamil ul "sunshine," etc. In Sumerian we have the
' See my Contributions, vol. IV, p. 233.
2 See C. R. Lepsius and W. Berend, Lesmitaux dans les inscriptions egypl-
iennes, in BihliotMque de l'6cole des hautes tltudes, part XXX, p. 12 flf.
root el "clear, pure," which, in Ass3rrian is found as ellu

" to shine. " We are, therefore, once more in the basic
group which produced Tatar alcin "gold," hence it
must be assumed that the electrum proceeded from
somewhere in the center of Asia.
The
electrum was more brilliant than gold, and re-
sembled brass, ^ hence it led to an artificial alloy in
which the alchemists did not necessarily employ the
native electrum} They manufactured it from two
parts of silver, one part of copper and one of gold,^
and produced it of even inferior quality.* This com-
position is as frequently called elidrium^ and lato,^
and it hardly needs any proof that these words are cor-
ruptions of electrum. We
may, however, follow the
degeneration of this group through the vocabularies.
The Greekalchemists used iXMpiov as an equivalent
for the plant chelidonium' and also for a certain yellow
mineral, and Berthelot has shown that among the alche-
mists writing in Latin elidrium has similarly this double
meaning; but in this case the yellow metal is definitely
the electrum^ Apparently there is here a confusion
of YjXexrpoz and ;^df5ov«a, which has produced this
1 Ibid., p. 15.
^ "Quod ex urina pueri et auricalco fit aurum optimum: quod intelligen-
dum est in colore, nonin substantia; hoc auricalcum frequentis scripturae
vocatur electrum," Vincent of Beauvais, Speculum naturale, lib. VIII. 36,
and M. Berthelot, La chimie au moyen Age, Paris 1893, vol. I, p. 83.
^ Ihid.,
pp. 78 and 218; also in Mappae Clavicula, in Archaeologia, vol.
XXXII, p. 215.
* "Eris partes, iij. argenti pars. i. simul confla et adicies auripigmenti non
usti partes, iij. Et cum valde calefeceris, sinito ut refrigeret et mitte in

patina, et obline argilla, et assa donee fiat cerussa; tolle et confla, et invenies
argentum. Si autem multum assaveris, fiet electrum, cui, si pars. i. auri
addideris, fiet aurum optimum," Berthelot, op. cit., p. 218.
^ Ibid., in Index, sub elidrium.
6 Ibid., p. 83.
' "XeXidovla larl rb iXidpioy," Berthelot, Collection des anciens alchimistes
grecs, vol. I, Paris 1877, p. 16. See also vol. Ill, pp. 306 and 310.
* Berthelot, La chimie au moyen Age, vol. I, p. 31. See also pp. 213, 218, 220,
and Archaeologia, vol. XXXII, pp. 193, 194, 195, 196, 200, 203, 227. Also
J. M. Burnam, Recipes from Codex Matritensis A 16, in University of Cirv-
cinnati Studies, vol. VIII, p. 16.
elidrium. In the Anglo-Saxon vocabularies we find the
forms elotr, elothr,^ electre, elehtre, which show inter-
mediate stages between electrum and elidrium, and also
betray a form elotrium, which, however, is not recorded.
But we findthe form lato in the Romance
early
languages, which is apparently a corrupt form, in which
chelidonium affected electrum and elidrium. Indeed,
we possess intermediate stages which show how electrum
developed into lato. In the Dissertationes printed by
Muratori, we have the two forms eletarum and letarum,^
from which there is but a step to letanum and latonum,
recorded by Ducange. It is, however, not unlikely that
we have here a confusion with x'^hcbz kXardz,^ the ductile
brass of the Greek alchemists.
The ASaxon vocabularies have for Latin electrum
also the glosses smylting,^ smelting.^ This is identical
with the OH
German gloss for electrum, which is
gismelze,^ but this word is also employed for pure gold,
obryzum.'' In the tenth century the Germans well knew
that this was of Italian origin,^ and, indeed, the earliest
reference to it is to be found in Anastasius Life of Leo '
IV, of the end of the IX. century, where we learn that

the enamel, for this is meant here by smaltum, was
carried out on a background of pure gold.^
' J. H. Hessels, A LatinÂnglo-Saxon Glossary, Cambridge 1906, p. 102.

^ "Ad cluttan auream de Letarum. Eletarum comodo fiet? Pones duas
partes argenti, et eramenti tres, et auri tres," Antiquitates italicae medii aevi,
Mediolani 1739, vol. II, col. 384.
5 Berthelot, Collection, vol. II, p. 313.
* T. Wright and R. Wiilcker, Anglo-Saxon and Old English Vocabularies,
London 1884, vol. I, cols. 141 and 334.

' Ibid., col. 148.
' Steinmeyer und Sievers, op. cit., vol. I, p. 641, vol. II, pp. 399, 463, 523,
533, 546, 593, vol. Ill, pp. 121, 192, etc.
' "Gismelcit," ibid., vol. I, p. 509; "smehigolt, smalzgolt," ibid., vol. II,
p. 14; "smehigold," p. 18.

' "Electrum heizet in uualescun smaldum," E. G. Graff, Althochdeutscher
Sprachschaiz, Berlin 1842, vol. VI, sub smelzi.

' "Fecit denique tabulam de smalto, opus ducentas sexdecim auri obrizi
pensan. libras," Acta sanctorum, July IV, p. 314.

As the foundation was more frequently of composi-

tion than of gold, electrum came to have not only the
meaning of a substitute for gold, but also of the whole
enamel work as well, hence "to adorn with electra"
was tantamount to "to adorn with enamel work. "^
Theophilus has devoted a chapter to the manner in
which the electra were to be laid on,^ and from this it is
again evident that the enamel was laid on by means of
a composition called electrum. In the Liber pontificalis
we have mention, under Benedict III (855-858), of a
reticule made of gems and golden pendants, which had
in it pieces of gold enamel, where, for electra, we have
"petias exmaltitas."^ Toward the end of the century
we hear, in the same Liber pontificalis, of enamel work
under the name of smaltum}
Theophilus gives, in the chapter on enamel, a careful
description of the manner in which the electrum has to
be melted down evenly and smoothly.^ The electrum
itself, that is, the gold composition, was, besides, very
carefully polished until it shone brilliantly.^ It is,
therefore, clear that the confusion between electrum and
smaltum is due to the employment of the alloy, or of
pure gold, in the enamel work, and that smaltum meant
' "Ornare lapidibus et electris atque margaritis," Theophilus, Schedula

diversarum artium, lib. Ill, cap. 50, in Quellenschriften fur Kunstgeschiehte
und Kunsttechnik des Mittelalters und der Renaissance, vol. VII, p. 223.
235 flf.
2 Ibid., p.
'"Benedictus praesul pulcherrimi decoris retem factam miro opera totam

gemmis ex alva veris et bullulis aureis, conclusas etiam auri petias in se
habentem exmaltitas praecepit fieri," L. Duchesne, Le liber •pontifi-
calis, Paris 1892, vol. II, p. 147.
^ "Posuit cantram auream I, cum pretiosis margaritis et gemmis ac
smalto," (885-891), ibid., p. 194; "obtulit in ea crucem auream super altare
cum gemmis et smalto," ibid., p. 195.
' "Aperiens vero tolles electrum et lavabis, rursumque implebis et fundes
sicut prius, sicque facies donee liquefactum aequaliter per omnia plenum
sit," Theophilus, op. cit., p. 239.
* "Super quod polies ipsum electrum donee omnino fulgeat, ita ut si dimidia
pars ejus humida fiat et dimidia sicca sit, nuUus possit considerare, quae
pars sicca, quae humida sit," ibid., p. 241.
—

" enamel work made with molten electrum carefully and
smoothly distributed over the plate."
We can now proceed to trace the history of smaltum.
The Semitic languages have the root mlt "to smear
over," from which they derive their word for "cement,
paste." Thus we have Syriac 4:^^ mlat "he smeared
over, put on, meditated," -^-^^ milt "smeared over,
wise," hence ;^^>^ ^4^ mlat, mud,"
mlata "cement,
Hebrew ^7.^ melet "mortar, cement," Arabic •^
malita "smooth," -^ malata "he smeared over,"
^"^ milat "cement," Ethiopic mald6 "to rub off,
make bald." But there are also the forms, Syriac

4^^ mrat "to make bald," and Hebrew derivatives of
the root IS"1D mrt which mean "to pull out the hair,
smooth down, make bald, polish," Arabic marata "he
plucked out the hair," which are related to this group.
The enormous mass of words in all the languages of
Europe and Asia that are related to this group cannot
be discussed here. All that is necessary is to keep in
mind the specific meaning of "cement, mortar," which
the Semitic nouns have evolved. Related to these, or
directly derived from one of them, are Latin mortarium,
"mortar," that is, "cement" and "pounding instru-
ment, pan in which cement is made," Greek /z<£Aâ
"calking pitch, wax smeared on writing-tablets."
The Latin mortarium shows at once that we have here
the same root as in Sanskrit marda "crushing, grinding,
rubbing."
The word which directly concerns us is Latin maltha,
the very writing of which shows its borrowing from
Greek. Pliny calls by this name a natural cement,
found in the Syrian city of Samosata,^ and an artificial
' "In urbe Commagenes Samosata stagnum est emittens llmum maltham
—
vocant flagrantem. cum quid attigit solidi, adhaeret," II. 235.
hard cement made chiefly from chalk. ^ He also used

the verb malthare "to lay on a cement, enamel,"^ and
an old gloss to Juvenal reads "Sulphure solent vitrum
solidare, id est malthare,"^ but solidare means not only
" to solder, " but also " to cement, enamel. " In Italian
we have side by side malta "cement, mortar," maltare
"to cement," and smalto "cement, mortar, enamel,"
smaltare "to enamel," smaltire "to digest," smaltito
"digested, clear, intelligible, easy."
Thus we see that the Semitic word, still noticeable in
Pliny's maltha, as referred to the natural asphaltum, and
the Greek word, which gives Latin maltha "cement,"
have in Italy produced the two forms malt- and smalt-,
both of them of late origin, although the Latin mor-
tarium, Sanskrit mar da, and the endless number of re-
lated words are widely diffused and go back to pre-
historic times. Now the Greek has already a verb
fisXdecv "to melt," but it is an extremely rare word, and
has left no trace behind in either Latin or the other
Indo-European languages. In Modern Greek we have,
indeed, a number of similar verbs, but it can be shown
that they all are newly formed from ^/Aroc, afidlroz,
that is, that they owe their origin to the employment of
li6.l-â in the enamel work of the Middle Ages.
In Modern Greek we have aiidX-d-vj, aiidXroz, afidydoz

"enamel," a[ia).T(i)vu), afiayhmvui "I enamel," and iiilroz
"enamel," imXtocd "I enamel." Here there is a confu-
sion between Italian smalto and Greek fiiXroi; "rubric."
Modern Greek afidydoi; is found at an earlier time as
aiidpdoz,^ where there is apparently a confusion with
a(idpay8oc. "emerald," for we find a(idpay8oc, mentioned
among the metals.
' "Maltha e calce fit recenti. glaeba vino restinguitur, mox tunditur cum
adipe suillo et fico, duplici lenimento. quae res omnium tenacissima et duri-
tiam lapidis antecedens," XXXVI. 181.
2 "Quod malthatur, oleo perfricatur ante," ibid.
' In Forcellini.
* Berthelot, Collection, vol. Ill, pp. 323, 329.
The alchemists of the early Middle Ages, according
to Berthelot/ confused a large number of red sub-
stances of mineral origin under the name of fj-t^roz,
namely oxides of iron, lead, and mercury, and protoxide
of copper, so that rubric {/liXzoi;), minium, cinnabar,
vermilion are frequently synonyms. But the metallic
oxides are also known under the name of ioc virus. The
Latin glosses give rubrica for fiihoz and aerugo for ioc.
Another word for the oxide is rubigo, and, as Toe indi-
cates, the spoiled or poisonous product of the metal is
meant by this word. The Germanic glosses have
rubigo^ and erugo,^ miltow, milcdov, milidou, etc., and
distinctly identify rubigo with erugo'^ and also with
metallaJ' This is also the case with the ASaxon
glosses, where we read "erugo, i. uitium frumenti, uel
ferri. "^ OHGerman miltou has, however, lost the
meaning of "metallic rust," and refers only to "plant
rust," while ASaxon mildeaw, unrecorded before the
XI. century, is popularly connected with Latin mel
"honey," and at first meant "nectar."^ The ASaxon
glosses give for rubigo and erugo the words brand and
om, the latter of Arabic origin, as has already been
discussed.
In a similar, if not identical, way has arisen the con-
fusion in Syriac for the words denoting "metal" and
"mine." The Talmud has already i^,?^? metalyah,
p^telD metalon "mine, quarry," which is borrowed
directly from the Latin, since the reference is invariably
to the Latin "damnare in metallum. " But the He-
1 Ibid., p. 261.
2 Steinmeyer und Sievers, op. cit., vol. I, pp. 290, 430, 438, 669, 819, vol.
II, pp. 473, 683, vol. Ill, pp. 307, 343.
> Ibid., vol. I,
pp. 443, 445, 519, 524, 669, vol. Ill, pp. 234, 384.
' Ibid., vol. I, pp. 242, 669.
»
Ibid., p. 710.
' Wright and Wiilcker, op. cit., vol. I, col. 229.
' Ibid., col. 455.
brew ''''P? metll, Talmudic ^to? metal "iron rod" are

older, and with Arabic matala "he forged, extended"
seem to point to a root with the original meaning "to
forge." The Semitic word is not universal and so is
itself a borrowing, together with the Greek, from some
other source. The Syriao has vo-sîo metalon, obviously-

Greek fiizaUov"mine," and, side by side with it,
ô.i.i^ melaton "mine," which is supposed to be a cor-
ruption of the first. It is more likely a confusion with

Greek [xiXroz.
The Arabic has o->«* ma din "mine, mineral, metal,"
but in some places, as in Morocco, it generally signifies
"brass" or "bronze." This is referred to the root
ij"^ âdana "to stay," but that is mere popular etymol-
ogy, for not only is "metal, mine" only forcedly de-
duced from but we have also derivatives from o-^

it,
ma din, namely o-^ mddana "to mine," o-^—

mumddan "metalUc, " where any ultimate origin from
<j-^ âdana not suspected. The word is utterly un-
is
related to any Semitic word, and is unquestionably a

borrowing. As the Syriac shows a confusion with
Greek //«7roc, due to the Greek alchemists vocabulary, '
Arabic o-*" mddin is without doubt due to the same

confusion which in Modern Greek produces afidltoz,
aiidydoz, [i'dTo<; for "fusible metal." Indeed, Arabic
mddin must be due to a popular borrowing, from Greek
afidySov or fiiXrov, from which a form [liydov, although
unrecorded, is to be presupposed from the form an6.y8oz
for andXToz. Such a popular borrowing would have to
take place in a territory where Greek and Arabic met on
intimate terms. Such a place would be Sicily, and it is
here, indeed, that we learn of the intermediate forms
of the unrecorded fiAydov, fiiydov. Here we have in
Sicilian mauta, smauto for Italian malta, smalto.
Similarly, we have Neapolitan smardire, smantire,
smautire for Italian smaltire, and smardare for Italian
smaltare. In like manner, the Provengal dialects have
esmaut, esmalt, esmart for "enamel, " and even the LLatin
records esmaudus by the side of smaltus. As Italian
smalto and malta have the same meaning, "cement,
mortar," the Arabic ma din cannot be separated from
Syriac melaton or malton (for the vocahzation is not
given), and both must be considered as contaminations
of the same type as MGr. afiaydchvu) and fidzdo). In-
deed, the most likely direct derivation of Arabic o-*"
is from the Syriac ôîi«i* mlton, read as î^^ m'ton and

popularly derived from âdana "to stay."
o-**
The Prudentius glosses several times gloss "rudera"

by "stercora, " and this by molta, that is, molta is equal
to "waste, dirt. "^ In the Keronian and Hrabanian
glosses molta is the translation of " humus "^ and
"telus. "^ Similarly the ASaxon glosses give molde
for "sablo, sabulum, humus, rus, aruum, " and the
Gothic has mulda "dust," muldeins "earthly." All
these words are borrowed directly from LLatin molta, a
variation of malta. "Cementum est molta ex quo
conjunguntur lapides, " says the Liber sacerdotum,^
and from the Ordo Romanus we learn that in the dedica-
tion of a Catholic church the relics of the saint were de-
posited at the altar in mortar, and that this mortar,
malta, was made of chalk and holy water.^ But in the
beginning of the XII. century this deposition of the
1 See p. 38.
^ Steinmeyer und Sievers, op. cit., vol. I, p. 168.
' Ibid., p. 8.
* Berthelot, La
chimie au moyen Age, vol. I, p. 217.
' "Tunc maltam de calce et tegula cum ipsa aqua benedicta ad oc-
faciat
cludendas sanctorum reliquias loco altaris," in Ducange, sub malta.
relics had become purely symbolical, for the mortar was

made of holy water mixed with salt and ashes, and ^
this mortar, called molta or mola, was, after the cere-

mony on the ground.^ Thus
of dedication, scattered
molta synonymous with "dust, earthly
or mola are
relics," and thus the Germanic words are directly de-
rived from the Latin.^
The Greek fidkay/ia "poultice, plaster" was accepted
by the Romans as a medical term, but the mediaeval
alchemy extended the meaning of malagma, with its
corruptions magma, mulmus, so as to include the mean-
ing "any finely-ground substance," more especially
"a wet, macerated mineral compound,"* that is, we
get here a synonym for molta. Most of the recipes in
which malagma, magma occur, refer to the making of
liquid gold, or, rather, a composition that should pass
for gold, hence fiakayfiaziêm came in Greek alchemy to
mean "to amalgamate with mercury,"^ and iidXayfia,
in Modern Greek very early was synonymous
fidlafia
with "gold." At a later time the alchemists used
amalgama for this reduction of gold or silver by mer-
cury. This amalgama is Arabic <«^?J-J' almalgamah
1 "Post haec benedicitur aqua, cui admiscentur sal et cinis, et ad ultimum
vinum. .cinis vero significat reliquias," J. P. Migne, Patrologia latina,

.
vol. CLXXVII, col. 386.

^ rru)lae ad basim altaris condendo.
"De residue moltae seu .Et quod
. .
remanet de molta qua conditae sunt sanctorum reliquiae, fundit ad basim

altaris," ibid., col. 387.
' Mola is the mola salsa of the Romans, salted meal used, just as in the
Catholic church, for sacrificial purposes; but molta is our malta, apparently
influenced by mMita "ground (grain)," so that, after many vicissitudes,
nuiUa has returned to the original root form, represented in Latin molere.
*"Conteres limaturas, adiciens aceti acerimi salisque modicum, donee
argentum combibat limaturam, et fiet malagma," Liber sacerdotum, in
Berthelot, La chimie, vol. I, p. 193; Mappae Clavicula, in Archaeologia, vol.
XXXII, p. 195; "et commisce guatum cum coctione magmatis; et tere
diligenter donee pulvis fiat," ibid., p. 219; also Muratori, Antiquitates, vol. II,
col. 378; "fungus est rotundus, pagani vocant amanita. desiccatum
ubicunque percusseris pulverem levat mulmum," Mappae Clavicula, in
Archaeologia, vol. XXXII, p. 239.
' Berthelot, Collection, vol. II, p. 164.
"

"amalgam,"^ of which the popular form was r*-^' i^-^
almaraham, marham "poultice." The amalgamation
of the metals consisted in a calcination of the metals by
a drying process after treatment with mercury, hence
d(ppoasXyjvo(;, the usual term for "chalky substance,
fine powder,"^ became in Syriac the usual word for
"amalgam." Thus malagma in alchemy meant both
"amalgam" and "fine, chalky substance, powder."
This is represented in the Germanic languages in the
shorter forms, Gothic malma "sand," ASaxon mealms-
tan "sandstone," ASaxon, OHGerman melm "dust,"
but is also found in Italian as melma "fine, pasty mud
at the bottom and pits.
of rivers
The Arabic word "metal" is found in all the lan-
for
guages that have borrowed directly from the Arabic.
In the Slavic languages we have OBulgarian med'
"brass," meden "of brass," but Slovenian medina "ore,
metal," by the side of med "ore, metal, copper, brass,"
and medinar, medar "coppersmith," Serbian mjedenica
"brass kettle, bell" make it certain that we have in
OBulgarian med' a backformation of meden, that is,
that we have borrowings from the Arabic. In the
German Messing we have a totally different origin.
ASaxon maestling is given as the translation of "elect-
rum" and " aurichalcum. " The first means, as we
have seen, "a metal mixed from gold and silver, " and is
generally given in the vocabularies as ^'aurum et ar-
gentum mixtum, " hence ASaxon maestling is an ASaxon
transformation of Latin mixtum. Indeed, in Ducange
there is recorded a form mestallum, as though trans-
formed from "metallum, " for "copper" or "brass,"
even as a mixture of grains is found in the form mestal-
lium, which leads to OFrench m4tail. When "aurich-
alcum" came to mean "a mixture of metals known as
' R. P. Smith, Thesaurus syriacus, p. 351, sub afroselinon.
^ Berthelot, La chimie an moyen dge, vol. I, pp. 31, 203.
"
bronze or brass," the ASaxon maestling assumed the

meaning of "brass," and a similar form is found in
German Messing "brass."
The Germanic languages have borrowed either the
Latin malta or the Italian smalto for words to denote the
art of enamelling. In the Gothic Bible we have once
gamalteinais, as a side gloss to " diswissais, " which is
the translation of Greek dvakuasax;. But ivdluaci; is an
alchemical term for dissolution of a body by melt-
ing, hence the Gothic glossator has chosen the root
malt- with which to explain the less common "diswis-
sais. " There is no way of determining whether Gothic
possessed any derivatives from smalto. We have al-
ready seen that in OH
German the usual translation for
"electrum" is gismelze. The earliest references to
smelzan show that the enamel was understood to be
produced by liquefaction of the metal, hence we get in
the Keronian glosses the meanings "Uquefieri, mulcere"
for it,^ and we have even the unmutated form smultar
"liquidus, serenus. "^ The electrum was generally
confused with the purest gold, the obryzum, hence we
find electrum glossed ubarguldi, "^ where
by "obrizum
ubar is unquestionably the abbreviation GBR,* for
obryzum; but more frequently electrum was identified
with quicksilver,^ or even with glass, ^ while the Ker-
onian gloss translated electrum by " uueralttiurida,
that is, "natural glory."
' "Mulcet smilcit, mulcendus smelzendi," Steinmeyer und Sievers, op. cit,
vol. I, p. 207; "liquore smelzi," ibid., vol. II, p. 450; "liquido smehindimo,"
ibid., p. 516; "liquitur smalz," ibid., p. 555; "smalt," ibid., p. 584; "liquatur
smilcit," ibid., p. 680; "sagimen smalz," ibid., vol. Ill, p. 259.
" Graff, op. cit., vol. VI, col. 830.
' Steinmeyer und Sievers, op. cit., vol. II, p. 499; see also vol. I, p. 508.
* See my Commentary to the Germanic Laws and Mediaeval Documents,
Cambridge 1915, p. 185.
= Steinmeyer und Sievers, op. cit., vols. I, p. 134, II, pp. 384, 886, 390, 411,
499, 530.
«
Ibid., vol. I, p. 653.
"

In the ASaxon glosses we have smelting, smilting
" electrum, " where there is possible a confusion between
" electrum " and "amber, " but the transferred meanings
survived in smolt "serene, quiet, peaceful," smylte
" quiet, tranquil, calm, serene; " but we also have meltan
"to melt, become liquid," and, like Italian smaltire,
"to digest." Curiously, while the OHGerman glosses
several times translate "obryzum" by smelzigold, the
ASaxon glosses have ymaeti gold and smaete gold}
The assumption that smaete comes from smiton "to
beat," is inadmissible, because "obryzum" is not leaf-
gold in particular, but any pure gold. We
have here
simply a form without the Z, such as we have already
met in the languages discussed above. Indeed,
smaete did not survive in Anglo-Saxon, except in the
connection smaete-gold, and that, too, not for any
length of time. By the side of smolt, smylte "serene,"
we have, in the early vocabularies, also the ASaxon
smoedum "politis, " smod "smooth," smede "lenis,
smedie "polio." Thus we have here as large a variety
of derivatives from the original, which means "to en-
amel, make a smooth surface," as we have in the
Romance languages or in the Greek. There cannot be
the slightest doubt as to the philological identity of
Gothic smipa "smith," gesmipon "to forge," OHGer-
man smid, smitha "smith," smidon, smithon "to forge,"
and AS. smidian "to forge" with this group, because
OHGerman smeidar "a worker in metal" and smida
"metal," smidaziereda "necklace" show that we proceed
to "smith" from the idea "metal, precious metal."
The peculiar vocalization and the whole form of
OHGerman smid, Gothic smipa is apparently due to a
confusion with Arabic -V- smid " finely ground, smooth,
white flour." We have also Syriac smlda, Hebrew
1 Hessels, op. cit., p. 156, sub obrizum.
'^'J'^? smidd, Coptic samit, related to or derived from

Greek from which also comes Latin simila,
ae/itdaXci;,
similago. This word is, in all likelihood, derived from

Egyptian jfma, Coptic §ma "to grind fine," even as
the other Coptic word for "fine wheat flour," noeit,
noit, is derived from nout "to grind," and Greek
dXeopov "flour" is related to d^-&ecv "to grind." Such
a confusion is the more probable, since in ASaxon
we have direct derivatives from the Arabic samld or
samldun in ASaxon smeduma^
We can now show how smidan has really arisen from
smaltarius "metal-worker par excellence." The Ker-
onian glosses, text Pa., has "Dedalus smaidar," and,
under "opifex, " in texts Gl.K. and Ra., "artifex
smeidar." Text Gl.K. has also here "fabricat smid,"
but it is clear that the regular VIII. century word for
"smith" was smaidar, smeidar, which is a Germanized
form of smagdarius for smaltarius, even as maistar for
"opifex" is a popular, Germanized form of magister.
1 "Polenta .i. subtilissima farina .i. sineduma; uel gisistit melo," Steinmeyer
und Sievers, op. cit., vol. I, p. 375; "simila smliuma," ibid., vol. II, p. 341;
and in the Leiden Glossary, and in Bosworth.
CHAPTER II.
The Gypsies in Eueope.^
In 789 Charlemagne issued at Aix-la-Chapelle a

general admonition to the clerical and secular authori-
ties of his realm, in which the following decree is con-
tained: "Item ut isti mangones et cotiones qui sine
omni lege vagabundi vadunt per istam terram, non
sinantur vagare et deceptiones hominibus agere, nee
isti nudi cum ferro, qui dicunt se data sibi poenitentia
ire vagantes: melius videtur, ut si aliquid inconsuetum
et capitale crimen conmiserint, ut in uno loco permane-
ant laborantes et servientes et paenitentiam agentes
secundum quod canonice inpositum sit. "^ As it
sibi
stands, this decree says: "Let those cheats who travel
about this land without any law be stopped from run-
ning around and cheating people, and those naked ones
with iron, who say that they are roving on account of
a vow of penance. It seems best that, if they have
committed any unusual and capital crime, they should
stay in one place working and serving and doing penance
according to ecclesiastic law."
Hefele' thinks that this law was directed against a
class of cheats called "mangones" and "cotiones,"
who were no longer to wander about freely, and the
naked people with chains, who pretended to wander
about for penance sake, were to stay in one place and
'
there do penance, if they had committed a grievous
1 See mycontributions in the Journal of the Gypsy Lore Society for July,
1909, April, 1910, and October, 1910.
^ Monumenta Germaniae historica, Capitularia, vol. I, p. 60 f.
' Conciliengeschichte, Freiburg im Breisgau 1877, vol. Ill, p. 670.
THE GYPSIES IN EUROPE 55
crime. Mabillon^ tried to explain this penancing

with iron by adducing two or three cases from the VII.
century where men, having committed murder on their
relatives, took the vow of penance by carrying bonds
made of the iron with which the crime had been commit-
ted upon their necks and arms until, by a miracle, they
burst open. Even assuming that these cases of pen-
ancing for parricide are typical, they cannot have oc-
curred in such large numbers as to demand special and
oft-repeated legislation. Besides, if this penance were
sincere, why should those who did penance wander
about and deceive people? There is nowhere any
mention of "naked men" doing penance with iron, and
Canciani^ uses this passage as a proof of such a prac-
tice, just as Mabillon and others had adduced other
cases of penancing with iron, in order to prove it.
The mistake made by the authors who have analyzed
thispassage is due to the fact that they have misunder-
stood the phrase "nudi cum ferro" and thus have
created two or three classes of people to whom the de-
cree is supposed to refer. But "nudi cum ferro" is a
mere lemma taken by Charlemagne out of Vergil and
Servius. To Georgica 1.58 "At Chalybesnudiferrum,"
Servius writes: "Chalybes populi sunt, apud quos
nascitur ferrum, unde abusive dicitur chalybs ipsa
materies, ut vulniflcusque chalybs. 'Nudi' autem
aut apud quos arbores non sunt: aut vere nudi propter
ferri caedendi studium; nam legimus Brontesque Sterop-
esque et nudus membra Pyracmon. " The Brevis Ex-
positio writes: "Chalybes gens in Ponto inventrix
ferri, dicti a Chalybio, Euboiae vico, quod hinc coloni
sunt. Nudi' ad hoe expediti, vel quod ita operantur,
'
cum fodiunt ferrum. " In the Commentary of Probius

' D. Joannis Mabillonii Praefationes, Venetiis 1740, p. 69.
' F. P. Canciani, Barbarorum leges antiquae, Venetiis 1785, vol. Ill, p. 209.
"

we read: "Chalybes natio Pontioa, in qua sunt ferri-
fodinae: ibi autem a nudis ferrum quaeritur.
It follows from these glosses that Charlemagne, in
using "nudi cum ferro, " had in mind the Chalybes, that
is, iron- workers, who were in his day roving over the
country, on the plea that they were doing penance.

It was they that were the cheats, the "mangones" and
"cotiones, " against whom he was legislating. In other
words we are dealing here with Gypsies, although we
are not yet in a position to ascertain whether they be-
longed to the same linguistic family as the modern
Gypsies. All we know so far is that they worked in
iron and at the same time cheated the people. Here-
tofore we heard only of smiths who were respected for
their art and who often were supposed to be endowed
with supernatural powers, while here w& are informed
that they were cheats and vagabonds. What was the
cause of this sudden change in attitude towards the
very important class of iron-workers?
We
are able to ascertain this from the identical re-
lation that the Arabs bear to the blacksmiths.^ The
best proof of the contempt they have for the Gypsies
appears from the study of a series of words applied to
them.
Aristophanes several times uses the term x6âko<;,
which obviously means "trickster, cheat." The word
is otherwise totally unknown in Greek literature, ex-
cept in annotations of the scholiasts, as in Harpocra-
tion, where the word is also quoted as found in the
second 'Azd-lc: of Philochorus, where Bacchus is men-
tioned as being a xoâXoz, a cheat. Yet the large
Greek dictionaries, from Stephanus to Liddell and Scott,
and Pope, aver that xoâkoi "were also a set of mis-
chievous goblins, invoked by rogues. " This erroneous
statement is due to an interpolation in the scholia to
' A. Robert, L'Arabe tel qu'il est, Alger 1900, p. 99.
Harpocration/ which Maussacus unfortunately at-

tached to the word xoâXeia, although it is absent from
the oldest scholia of the Codex Venetus and Ravennates,^
and is not found earlier than the XII., possibly not
earlier than the XIV., century.^ This interpolation
reads: "xSâXoi, daifiovs^ dai Tevsz axhjpoi Ttepi zbv JtSmaov/'
"the cobaloi are certain malicious spirits about
Bacchus." The scholiast confused the reference to
Bacchus as a cheat with the mediaeval covali, cobalt
"goblins," and thus created for Aristophanes the im-
possible imps. Neither Hesychius, nor Suidas, nor any
other mediaeval Greek lexicon knows anything about
this reference, although the rest of Harpocration's
note is quoted in full. It appears clearly that xoâXoz
was an Attic slang expression, otherwise unknown to
the Greeks, hence apparently not of Greek origin. It
can easily be shown that it was borrowed from a Semit-
ic tongue, most likely from Phoenician. The root hhl
is found in all the Semitic languages. We have Assyrian
habalu "injury, ruin," habbilu "wicked rascal," Syriac
habdld "corrupter, pederast," hubdld "corruption, de-
struction," Ethiopic Mbul "tricky, perverse," Arabic
tijr^ Kabul, ^'y- huwwal "tricky, agile," J;*- habl
"corruptness," Hebrew hdbal "to hurt, destroy, cor-
rupt. " Among the Hebrews the evil angels were called
angels of **/?CI fiabbald "destruction," and from this
generic term there developed the specific Arabic evil
jinn. We have Arabic tM- habal, plural <J-e- hubbal,
"the This specific meaning does not oc-
jinn, devil."
cur in the other Semitic languages.
' G. Dindorf, Harpocrationis lexicon in decern oratores atticos, Oxonii 1853,

vol. I, pp. 180, 183.
* F. Blaydes, Aristophanis comoediae, Halis Saxonum 1886, vol. VI, p. 192.
' F. Dtibner, Scholia graeca in Aristophanem, Parisiis 1843, p. iii ff.
The Semitic root is an amplification of a root repre-
sented in Egyptian heb " to grieve, mourn, " hebau "mis-
erable man, wretched," heben "one who is dejected,
cast down," which in their turn are related to kheb "to
diminish, pilfer, destroy, deceive, defraud, " khebt, khebent
"moral obliquity, deceit, fraud, sin, wickedness,"
khebenti "offender, sinner, criminal," all developing
from an original meaning "to lay low," Coptic hobe
"to humble."
The Arabs also possess these words in oW=r âban
"coward," •r*- hab, hib, hub "a great deceiver, wicked,
deceitful, mischief-maker," o>- haun "diminution,
weakness," Cr^ gabn "to cheat, defraud, be unmindful
or inadvertent, to pass by inclining, to elude observa-
tion. " These Arabic words became immensely popu-
lar throughout Europe, and are found in most of its
languages. Arabic oj^ haun, i>'^ hd'in "treacherous,

unfaithful, looking treacherously or clandestinely at a
thing at which it is not allowable to look" is found in
Gothic hauns "low, humble," ASaxon Man "low, con-
temptible, miserable," OHGerm.an honen "to scorn,"
hence French honnir "to scold," honte "shame,"
Italian, Provengal, Catalan onta, OSpanish fonta
"shame." Arabic v^ hab "a great deceiver, mischief-
maker" has similarly entered into the Germanic and
Romance languages, producing ONorse gabb "scorn,"
ASaxon gabban "to mock, delude, jest," Italian gabbo
"jest," OFrench gab "mocker, deceit," Portuguese
gabar "to praise." Arabic oW^ gaban "coward" is
derived from the verb a^ gabn "to cheat, to pass by

inclining so as to elude observation." The great
variety of phonetic changes in the Arabic at once indi-
"
cates a foreign origin. Egyptian khebent has also left
behind the Arabic J^'j=~ hawwar, jj-'J>- hawwur ^îeehle,
cowardly," of which the plurals are ojj'j^ huwwarun,

'Jj=- huwwarat, of which the first is preserved in Vene-
tian cabalone "trickster, cheat," while the second is
found over an enormous territory. We have French
couard, Spanish coharde, Italian codardo "coward,"
while in the Balkan Peninsula we get Albanian gabarde,
kabarde "barbarian," OBulgarian kovannH "tricky,"
Turkish hor "vile, abject," horiat "barbarian, rustic."
The Turkish is, however, borrowed from the Persian,
where the Arabic word was popularly related to the
native Jiwar "light, easy," which only in Modern Per-
sian assumes the meaning "vile, abject."
The derivatives from the Arabic in the Albanian and
Slavic languages are instructive. We have Albanian
gabel "Gypsy, horse dealer, " gabe "lie," gabim "decep-
tion," gaboj "I deceive, cheat," kobe "theft," kobon
"1 steal secretly," kobim "deception," kohe "I tell
fortunes." The same group is found in the Slavic
languages. We have OBulgarian kovali "blacksmith,
kovalinya "anvil," kovii "deception," and similarly in
the other Slavic languages. The Albanian kobe "I tell
fortunes" by the side of kobe "theft," etc. shows that
OBulgarian kobi "augury," Rumanian kobi "to tell
fortunes" belong to the same group. The Celtic
languages use the Arabic u^ gabn to designate a

smith, namely Olrish goba, Armorican, Welsh gof, but
we also have Olrish gau, Cornish gow, Welsh gau "a
falsehood, a lie."
In the XII. century German kobolt, OFrench gobelin
make their appearance, the first obviously in the sense
of "puppet of the pantomime,"^ for the sources speak
'
J. Grimm, Teutonic Mythology, London 1883, vol. II, p. 5003.
"

of being struck dumb like a kobolt, speaking like a
kobolt, in silence or whispering.
that is, The kobolt was
carried by the juggler or showman under his mantle,
and he is often associated with the taterman, another
kind of puppet. The latter makes it possible to show
that the particular puppet-show mentioned by the
mediaeval German authors was derived from Spain and
ultimately from the Arabs. Taterman is derived from
Spanish titere "puppet," and this from Arabic J J
tartar "to shake, tremble." This Spanish word pro-
duced ONorse titra, OHGerman zitaron "to tremble,"
while in taterman the phonetic change did not take
place. German kobolt, OFrench gobelin are from Arabic
S-^ hubbal "mischievous spirit, devil."
From Charlemagne's decree it follows that he had
before him an Arabic law directed against the Gypsies,
where they were called Je- habal, a name, as we shall
later see, by which they have been known throughout
Africa. It may be that this Arabic word is a mere
transformation of the classical Chalybes, but in any
case it served to indicate the cheating diviners, who
also were blacksmiths, as which the Gypsies have ever
been known. Charlemagne divided the lemma of
habal into "mangones et cotiones, " that is, "cheating
traders, " and " nudi cum ferro, " that is, "blacksmiths,
and thus gave us the first definite reference to the des-
pised smiths found in literature.
Charlemagne 's reference to the penance of the Gypsies
is not spontaneous, but is based on a legend long cvirrent
in the Christian world. In the pre-Christian apocry-
phal book of Jannes and Jambres,^ unfortunately lost
to science, the two wizards who withstood Moses, accord-
ing to Midrash Yelammedenu, Ki Tissa (Ex. XXXII),
^ See The Jewish Encyclopedia.

were "among the mixed multitude that went up with

Israel from Egypt" and aided in the making of the
golden calf. A Greek tale on the subject is mentioned
in Gelasius' Decretum under the name of Poenitentia
Jannis et Mambre, and the title shows that the Egyptian
magicians were supposed to have done some penance.
Jambres, Mambre has survived in the Koran as
Samirl: "And we spake by revelation unto Moses,
saying, Go forth with my servants out of Egypt by
night; and smite the waters with thy rod, and make
them a dry path through the sea: be not apprehensive of
Pharaoh's overtaking thee; neither be thou afraid.
And when Moses had done so, Pharaoh followed them
with his forces and the waters of the sea overwhelmed
;
them. And Pharaoh caused his people to err, neither

did he direct them aright. Thus, O children of Israel,
we delivered you from your enemy; and we appointed
you the right side of Mount Sinai to discourse with
Moses and to give him the law; and we caused manna
and quails to descend upon you, saying, Eat of the good
things which we have given you for food; and transgress
not therein, lest my indignation fall on you; and on
whomsoever my indignation shall fall, he shall go down
headlong into perdition. But I will be gracious unto
him who shall repent and believe, and shall do that
which is right, and who shall be rightly directed.
What hath caused thee to hasten from thy people, O
Moses, to receive the law? He answered. These follow
close on my footsteps; but I have hastened unto thee,
O Lord, that thou mightest be well pleased with me.
God said. We have ah-eady made a trial of thy people,
since thy departure; and al Samiri hath seduced them
to idolatry. Wherefore Moses returned unto his
people in great wrath, and exceedingly afflicted. And
he said, O my people, had not your Lord promised you
a most excellent promise? Did the time of my absence
seem long unto you? Or did ye desire that indignation
from your Lord should fall on you, and therefore failed
to keep the promise which ye made me? They answer-
ed, We have not failed in what we promised thee of our
own authority; but we were made to carry in several
loads of gold and silver, of the ornaments of the people,
and we cast them into the fire; and in like manner al
Samiri also cast in what he had collected, and he pro-
duced unto them a corporeal calf, which lowed. And
al Samiri and his companions said. This is your god,
and the god of Moses; but he hath forgotten him, and is
gone to seek some other. Did they not therefore see
that their idol returned them no answer, and was not
able to cause them either hurt or profit? And Aaron
had said unto them before, O my people, verily ye are
only proved by this calf; for your Lord is the Merciful,
wherefore follow me, and obey my command. They
answered. We will by no means cease to be devoted to
its worship, until Moses return unto us. And when
Moses was returned, he said, O Aaron, what hindered
thee, when thou sawest that they went astray, that
thou didst not follow me? Hast thou therefore been
disobedient to my command? Aaron answered, O son
of my mother, drag me not by the beard, nor by the
hair of my head. Verily I feared lest thou shouldest
say. Thou hast made a division among the children of
Israel,and thou hast not observed my saying. Moses
said unto al Samiri, What was thy design, O Samiri?
He answered, I saw that which they saw not; wherefore
I took a handful of dust from the footsteps of the
messenger of God, and I cast it into the molten calf; for
so did my mind direct me. Moses said. Get thee gone;
for thy punishment in this life shall be, that thou shalt
say unto those who shall meet thee, Touch me not, and
a threat is denounced against thee of more terrible
pains, in the life to come, which thou shalt by no
means escape.And behold now thy god, to whose

worship thou hast continued assiduously devoted;
verily we will burn it; and we will reduce it to powder,
and scatter it in the sea. "^ The accidental resem-
blance of Samiri with Samaritan has led to the identifi-
cation of the Samaritans by the Arabs with the
"Touch-me-nots," which was the more easy, since the
Samaritans actually washed themselves after having
touched a stranger.^ However, while the Samaritans
purified themselves, the "Touch-me-not" of al-Samiri
was included as a curse.
The real identification of the Gypsy fortune-tellers
with the Egyptian magicians is found in the Syriac
versions of the Cave of Treasures, of not earlier than the
VI. century. Noah, upon awakening from his intoxi-
cation, cursed Ham and his race: "He was very angry
at his son, Ham, and said Cursed be Canaan, let him
:
'
be the slave of the slaves of his brothers!' And why

was Canaan cursed on account of Ham's guilt? When
he was a tall youth and had reached discretion, Satan
flew into him and became his teacher in sinning. And
he renewed the work of the house of Cain and manu-
factured flutes and zithers; and the demons and devils
flew in and settled in them, and every time the wind
blew through them, the demons sang from them and
uttered a mighty voice. And when Noah heard that
Canaan had done so, he was grieved that the work of
error had been renewed, which had brought about the
fall of the children of Seth, for with the singing and
playing and raging of the children of Cain Satan had
overthrown the giants, the children of God, and with
'
'
the playing of the flutes and zithers sin had grown in

the older generation, until God in his anger brought
about the flood. And because Canaan had dared to
1 Chapter XX, 79-97.
2 J. A. Montgomery, The Samaritans, Philadelphia 1907, p. 319.
. '

do so, lie his seed became the slaves of
was cursed, and
slaves, whoare the Egyptians, and the Cushites, (and
the Indians), and the Mysians. And because had Ham
been bold and had scorned his father, he has been called
'the unclean one' until the present day. But Noah,
by his sleep of intoxication, indicates the cross of the
Messiah, as the pious David has sung of him and said,
'The Master awoke as a sleeper and a man whom
his wine had shaken. The heretics are mad who say,
'
'
God was crucified, for he calls him here Master,
'
'
as the apostle Peter says, 'God has made him for a

Master and Messiah, namely this Jesus whom you have
'
crucified. And he
does not say 'God' but 'Master,'
for he means the oneness of the two hypostases, that
are united in one sonhood. And when Noah awoke
from his sleep, he cursed Canaan and humbled his seed
unto an enslavement and scattered his seed among the
nations. But the seed of Canaan, as I said, were the
Egyptians, and, behold, they were scattered over the
whole earth, and served as slaves of slaves. And what
is 'the slavery of slavery?' Behold, these Egyptians
are driven about in the whole land and carry burdens
on their backs. But those who were brought under
the yoke of subjection travel, when they are sent out
upon journeys by their masters, not on foot and do not
carry burdens, but ride in honor upon beasts, like their
masters. But the seed of Ham are the Egyptians, who
carry burdens and travel on foot, while their backs are
bent under their burdens, and who wander about at
the doors of their brothers' children. This punish-
ment was sent down upon them on account of Canaan 's
"^
foolishness, so that they became the slaves of slaves.
Another Arabic version of the story runs as follows:
"When Noah awoke from his drunken sleep, he cursed
1 C. Bezold, Die Schatzhohle, aus dem syrischen Texte dreier unedierten
Handschriften in's Deutsche ubersetzt, Leipzig 1883, p. 25 f
.
Canaan and made his posterity slaves. Likewise when

the Christ arose from the grave He cursed the Devil and
destroyed those who had crucified Him and scattered
them among the nations. The sons of Canaan became
slaves for ever, carrying burdens upon their necks.
Every proprietor negotiates riding about on his business,
but the children of Canaan negotiate about the affairs
of their masters, as poor men on foot, and they are
called the slaves of slaves."^
In these apocryphal works we learn that the Gypsies
were wandering about on account of the crime of magic,
and that they were called "the unclean ones," even as
al-Samiri was a "Touch-me-not," but we still have to
show that the appellation Egyptians actually refers to
our Hindu Gypsies, as they are known to us since the
XV. century. That the Gypsies actually passed
through Egypt, before scattering over Europe, follows
from the presence of at least one Coptic word in the
Gypsy vocabulary and one Gypsy word in the Coptic
vocabulary. We have Greek Gypsy biland "melted,"
bilanov "to melt," etc. Miklosich was unable to find
a Sanskrit root for it, but it is found in Hindustani
bilana, pighldna " to melt, " and is from Sanskrit pragal-
anlyam "melting, fusion." This is found in Coptic
bel, bol "to melt," for which there is no Egyptian
antecedent. Here we have a direct proof of the intro-
duction of Indian metallurgy into Egypt, which has
been established historically from the Arabian manu-
facture from Indian steel. But the Copts have supplied
the Gypsies with their very word for "Gypsy." We
have Gypsy rom "man, husband, Gypsy," which is
Coptic rome "man," rem "man, native," hence
rm-n-keme "Egyptian." As the Gypsy language in-
dicates no presence of Arabic words in its vocabulary,
the Gypsies must have started from Egypt on their
'
M. Gibson, Apocrypha arabiea, in Studia sinaitica, vol. VIII, p. 30 f
European and Asiatic pilgrimages before the arrival of
the Arabs in Egypt, that is, before the middle of the
VII. century. This agrees with the account in the
Cave of Treasures, where they are first mentioned.
In the Cave of Treasures we have a distinct reference
to Melchizedek, the high priest of God, whose altar was
at the center of the earth, where Adam was buried, and
where was the tree that was the precursor of holy rood.^
To those who considered the sect of the Melchizedekites
to be an abomination, it was therefore natural to iden-
tify the "unclean" Egyptians with Melchizedekites.
We find the earliest mention of these in Timotheus
Presbyter, of whom we only know that the latest date
mentioned by him is the year 622,^ who, therefore,
must have written in that year or later. He says:
"The Melchizedekites are those who now are called
Athinganoi. They worship Melchizedek, whence they
received their name. They live in Phrygia, and are
neither Jews nor pagans. They seem to keep the
Sabbath, and do not circumcise. They do not allow
anyone to touch them, and, if someone gives them bread
or water or some other article, they do not dare to take
it from their hands, but call out to the giver to put it on
the ground, and then they walk up and take it, and they
do similarly when they give something to others. For
this reason they are called Athinganoi, because they do.
not allow anyone to touch them. "^
It is clear that we have here a vague recollection of
the Jannes and Mambre story and the Cave of Treasures,
where the important point is that the cursed race of
magicians were called "Touch-me-nots." The very
fact thatnow the Melchizedekites, now the Samaritans
' Bezold, op. cit., p. 35.
2 J. P. Migne, Patrologia graeca, vol. LXXXVI. 1, p. 9.
For the later references to Athinganoi, see P. Miklosich, Vber die Mun-
'
darten und die Wanderungen der Zigeuner Europa's, Wien 1876, part VI,
p. 58 ff.
are called "Touch-me-nots" indicates that the impor-

tance lies in the name, which had to be explained. There
can be no doubt that the appearance of the Gypsies in
the Byzantine Empire under the Pehlevi Persian name
of dtsingar, aslngdr, literally "iron-workers," immedi-
ately suggested the Greek etymology 'Adii-yavot
"Touch-me-nots," for which the name "Egyptians"
attached to the Gypsies gave no explanation, whereas
the hated sectarian Melchizedekites, or the Samaritans,
at once suggested such an explanation. Gelasius'
Decretum is, to say the least, full of interpolations,
hence it cannot be adduced for the settlement of the
date when the "Touch-me-not" story arose. We are
left for this only with Timotheus Presbyter and
Mohammed, both of whom lived in the first quarter of
the VII. century. We are, therefore, not far from lo-
cating the arrival of the Gypsy dtsingar from a Persian
region into Byzantine territory, which gave rise to the
'Adifyavoi, the " Touch-me-nots, " whence they passed
into the Arabic and Syriac versions of the Jannes and
Mambre story and into the Cave of Treasures.
But the Gypsies, from their association with the
Egyptians and their scant borrowing of one word from
the Coptic vocabulary, must have passed through
Egypt and out of it immediately before the arrival of
the Arabs. We have, therefore, narrowed down their
Egyptian sojourn to a very short period, and the history
of their origin can be written with definite dates. The
Jannes and Mambre story and the Koran refer to the
"Touch-me-nots," but, although the reference is to the
Egyptian magicians, the Egyptians are not yet singled
out as "slaves of slaves." Therefore, the Greek
'Adiyyavoi arose either in the time of Mohammed, who
died in 632, or earlier. In 616 the Persian, Chosroes,
conquered Egypt. Unquestionably he brought with
his army Gypsy smiths, for we shall soon see that the
'

Greeks later also carried Gypsies in the army, a custom
which they inherited from the Persians. In 626 Egypt
was restored to Rome, and in 636 the Arabs swooped
down upon the country. Some of the Gypsies were
carried back to Greece by the retreating Greeks, who
settled them in various parts of the Balkan Peninsula
in localities that became known as runroxdarpa or
"Little Egypt." Hence the versions of the Cave of
Treasures which contain definite references to the
"Egyptians" who became "slaves of slaves" must have
arisen after 626, when the Egyptian magicians upon
whom there rested a curse were definitely identified with
the Indian Gypsies, the smiths, fortune-tellers, and
vagabonds, as which they have been known ever since.
The Gypsy name for "a Greek" in Greece is halamo,
while in Egyptian Gypsy balamu means "Christian."
The word does not seem to occur elsewhere and Mik-
losich was unable to explain it. It obviously cannot
come from a Greek source, and must have arisen in
Egypt, or before the Gypsy settlement in Egypt, to
have been preserved in the two countries. We have
the rare Hebrew 072. Mm "to bind, tame," belim
"dumb," Syriac hlam "to check a heresy," ballm "one
whose mouth has been tied up, mute, " expressions used
in regard to the silencing of a heretic. In Arabic, r-^-
'iblam "to be silent" occurs and is unquestionably a

borrowing from the Syriac. The application of
Syriac blam, belim to "a Greek," as in the Gypsy
language, can have arisen only in a Christian country,
among the non-orthodox people, that is, in or near
Syria itself, and must have gotten into the Gypsy
language from the Pehlevi, which also supplied
aslngar " iron- worker. " We are, therefore, once more
in the presence of the Persian conquest of Egypt, when
balamo was by the Gypsies applied to the Greeks in
Egypt and, later, to tlie Greeks in Greece. Had the

Gypsies received the word later, they would have been
obliged to apply the word to other nations as well, which
is not the case.
We have already seen that the Syriac Cave of
Treasures forced upon the Gypsies the appellation of
"Egyptian," hence of the Coptic rom "man." But
the Coptic word for " Egypt, " Kemi, must equally have
been current among the Gypsies, hence their full name,
in Gypsy, would have been Kemeno-rom, or some such
form. Indeed both rom and Kemeno-rom, as applied to
the Gypsies, are recorded. Kemeno-rom is found
Hellenized as xofiodpdfxoc: or xoDfiodpojioz, as though
it meant
"village-rover. " The word xcafiodpofietv occurs
already in Pollux, but the work of this author is so
badly interpolated that no conclusion can be drawn
as to the age of the word. We next find it in Malalas,
who tells of an Italian wizard, who came to Greece,
showing his magical art, xcofiodpofimv, in presenting
a blind dog that could return property to their owners
and do other wonderful things.^ Malalas places the
event early in the VI. century, but, of course, this does
not prove that the word was already in use then, al-
though such may have been the case. Theophanes
places this event in 536 and calls the wizard xwpo8p6fjio(;
{xopto3p6fio<;)^ Constantinus Porphyrogenitus, who
wrote in the X. century, says that each bandum, or
cohort, contained one xonodpofioz, that is, blacksmith,*
and there was a "vicariatus" called "that of the
xopodp6po(;."^ We have a form of xofiodpo/ioc in the XI.
century, when we read in a glossary, " gallodromi sunt
mangones discurrentes et fraude decipientes, "^ where
' Corpus scriptorum historiae Byeantinae, vol. XIV, p. 453.
2Ibid., vol. XXXVIII, p. 347. The same story Is also told by Glycas,
ibid., vol. XXVI, p. 501.
' Ibid., vol. VIII, p. 494.
* Ibid., vol. X, p. 225.
' Goetz, op. cit., vol. V, p. 620.

. '

we for the first time have the Gypsies mentioned as
cheats. In Modern Greek, xo/ôdpo/ioc: is given as "black-
smith, while xofjiodpofjteua) is "I forge."
"
The statement made in Constantinus Porphyrogeni-
tus that a Gypsy was attached to each cohort is of ex-
traordinary importance in determining the early settle-
ments of the Gypsies in the Balkan Peninsula, because
we find them later in military or administrative divi-
sions in connection with blacksmi thing. In 1204 the
Greek possessions were divided among the Emperor, the
Venetians, and the Crusaders. To the share of the Ven-
etians fell, among other provinces, the east coast of the
Adriatic, from Lacedaemon to Venice. In the enum-
eration of the lands ceded to them, we find "prouintia
^
Dirachii et Arbani, cum chartolaratis de Olauiniza,

de Bagenetia." Arbani is the modern Albania, and
Bagenetia is thus indicated as being in its neighborhood.
A footnote gives the variants Bagenatia, Vagnetia,
and quotes from Anna Comnena, 5, 4 (ed. Bonn,
T. I, p. 236), "xazala/LtjSdvs! 8td r^c Baysvrjrlaz ra 'Imdwcva,''
from Eustathius (Op. p. 282, 20) "ixno&sv Bayei^rla^," from
Chronicon Moreae (ed. Buchon, v. 7819), "rd fisprj t^(:
Ba^susvia^, vd /jaav Ttpbi; tyjZ d-akdaarji;,''^ and several other
passages, one from Le livre de la conqueste (ed. Buchon,
pp. 314, 324), where the form Vagenetie is given. From
all this we conclude that Vagenetia was the strip of land
opposite Corfu. In the Greek text of the convention,
the passage runs as follows: ^ —
"To Mpa Aup[)a-îou xal
'AXfidvou, ahv rdiz jâprouXapdzocc; t^c 'Z"£ rXa^cvirl^yjt; xal Bafev&riaz.
The next year the Podest^ of the Venetians in Romania
clearly defined the territorial division thus acquired:^
"In nomine domini Dei et saluatoris nostri Jhesu
Cristi. Anno domini millesimo ducentesimo quinto,
' G. L. Fr. Tafel und G. M. Thomas, Urkunden zur alteren Handels-
und Staatsgeschichte der Republik Venedig, Wien 1856, vol. I, p. 472.
Îbid., p. 491.
' Ibid., p. 569 f
'
mense Octobris, indictione nona. Constantinopoli.

Cum aliquid a prineipibus terre communiter ordinatur,
oportet, ut scripture uinculo anodetur, qua possit ordo
reioportuno tempore manifestius recognosei. Igitur
nos Marinus Geno, Venetorum in Romania Potestas
eiusdemque Imperij quarte partis et dimidie dominator,
cum judicibus et sapientibus conscilij et populi con-
laudacione, decreuimus, in scripturis publicis hoc esse
corroboratum: videlicet quod in diuisione iam dicte
nostre quarte partis et dimidie eiusdemque Imperij
Romanic, que nobis nostroque comuni habere contin-
gebat tempore, quo diuidebamus inter nos Venetos et
alios homines, qui uenerant in fldelitate et seruitio
domini Venecie Ducis, iam dictam quartam partem et
—
dimidiam tocius dicti Imperij dimisimus comuni
Venecie prouinciam Dirrachij cum chartolarato Glauen-
izi, et prouinciam Vagenecie, et Corfu cum tota eius
insula. Hec sunt enim sub Dirrachio, uidelicet
Sfinarsa, [cum] chartolarato de Glauenitis, quod chart-
olaratum potest esse cum tribus uel quatuor casalibus,
et Ablona: est catepanikium Vagenetie, et habet unum
char tola[ra] turn, de Gliki cum alijs duabus uillis et
duobus agridijs, idem (id est?) paruis casalibus. Hec
autem omnia suprascripta comuni Venecie dimissimus,
ut superius est denotatum, et uolumus, quod supra-
scriptum comune Venecie plenissimam imperpetuum
habeat potestatem ad faciendum de his suprascriptis
omnibus, quicquid sibi placuerit; et hec conflrmatio
'
cartemaneat in sua firmitate.
What is a chartolaratumf The editors of the text are
not sure about it. Proceeding from the assumption^
that Macedonia, Thessaly, and the Epirus were well
fitted for the raising of horses {seminaria equina), they
assume that the chartolarato were territories set aside
1 Ibid., p. 267.
for such a purpose. In another place^ they quote a
gloss "quae Constantinopolitani Imperii strategiae er-
ant, " and again assume that strategiae equorum were
meant. It looks as though they had merely been
guessing at the context, but they guessed well. In
Ducange we find xapzouXdpioi;, among other things,
with the meaning of "attendant upon horses." The
ps-j-a(; xaprouXdpcoz was an important dignitary in the
immediate service of the Emperor, and xaprouXdpyjz
is given as equivalent to "equiso, groom. " In the long
list of geographical names mentioned in the above
quoted documents chartolarata are given but three
—
times once on the Adriatic coast, once in Thessaly,
once in Macedonia. Now, we do know that Gypsies in
the thirteenth or fourteenth centuries migrated from
Vagenetia to Corfu and were there called Vageniti
homines. These apparently came from the chartular-
atum de Gliki in Vagenetia, where, if horses were raised
there, they would have naturally been employed. We
are fortunate to be able to show that there were also
Gypsies in the chartularatum de Glauenitis, near
Dyrrhachium, the modern Durazzo, on the Adriatic
shore, and that they were known there as Egyptians.
There is a Life of St. Barbaras the Egyptian,^ in
Greek and in Bulgarian. St. Barbaros was an Egyp-
tian of black color. At twenty years of age he lost his
parents and joined a piratical band, by which he was
chosen leader on account of his bodily strength. At
one time he set out to Durazzo on a piratical expedition.
A storm broke out, and Barbaros, who was a Christian,
began to pray to God, and he vowed that in case of
being saved he would devote his life to the service of the
1 Ibid., p. 472.
În the account of his life, and in the conclusions drawn from it, I follow
K. Radfienko, Einige Bemerkungen zur neugefundenen Absckrift des Lebens
des heil. Barbar in bulgarischer Uebersetzung, in Archiv fiir slavische Philologie,
vol. XXII (1900), p. 575 fi.
Lord. The ship with all its men was lost, but St.
Barbaros was saved. With the fantastic episodes in
his life we are here not concerned. What is interesting
to us is the fact that the Bulgarian author of the Life
says that there were many Egyptians near Durazzo,
and that by means of them St. Barbaros made himself
understood to others. In the Greek version St. Bar-
baros was called an African, but the Bulgarian author
transferred the scene to Durazzo, where he knew of the
existence of Gypsies who, as Egyptians, were to him
real Africans.
In some parts of Greece fidvTcz means both "fortune-
teller" and "blacksmith."^ Another Greek word for
this is [lavTcnokoi;, and this was, between 1364 and
1379, used by a Cologne clerical, presumably Joannes of
Hildesheim, in regard to the Gypsies in the Orient:
"In the Orient and in all the parts across the sea there
are especial men,
called Mandapolos (MandopoU).
They keep no especial rite nor heresy, nor have they any
priests among them. They travel about in large
crowds with their wives and children and asses, and
they do not sow nor reap, nor do they sleep in houses,
neither in winter, nor in summer, in the rain, or cold, or
heat of the sun, in daytime or at night, nor do their
wives bear children in a house, but they roam the whole
year from place to place, from one town to another; and
when they stay in one place, they manufacture sieves or
similar household utensils. They cannot stay in one
place more than three days, and it has frequently been
1 "MapTih /ivpljâ Cypern, wie ein Schmied riechen, da daselbst lidvris
sowohl den x'^^''^' (= Schmied) als auch den /xdrrts (= Wahrsager) bedeu-
tet," G. N. Hatzidakis, Zur Wortbildungslehre des Mittel- undNeugriechischen,
in Byzantinisehe Zeitschrift, vol. II, p. 266. XoXkci/s is not merely smith, ît
—
generally means "Gypsy:" "xaXmAs, xap*"**. <"Sep4s, dTf(7Koyos, fabbro,
ferraro," A. da Somavera, Tesoro della lingua greca-volgare ed italiana,
Parigi 1709; in _Chios the Gypsy blacksmith is called xop^J^s: "'Ek«
XO-pT^s Toi/s diravr^, xoprfi&s fi^ ra iratdui tou, x^P^^^ A*^ "^^^ yvvatKd tov Kt ij
liavpo<pafir]\id tov," A. T. ILaa-triTTjs, ri XiaKbv VXatraipiov, iv 'Affiiraij 1888, p.

383 f.
found that if they remained three days in one place, or
slept in a house or under a roof, they immediately died.
And they have among themselves a special language
which no one but themselves can understand or learn,
whereas they understand many tongues. They never
have any discord among themselves, whether in words
or deeds, but if anyone steals from another, as is the
custom with them, or if anyone finds another man with
his wife or daughters in adultery, he does not become
angry, but retaliates as well as he can. And these men,
coming to a place of the Christians or heretics or Sar-
acens, so long as they are among them, they live accord-
ing to their rites, manners and customs, and wherever
people fast, celebrate, eat, drink, work, lament, weep,
or rejoice, these people do likewise. Nor have they
any priest, nor any especial rite, nor law, but in
whatsoever place of the Christians or heretics their
wives bear children, they baptize the children according
to their rites, and in whatsoever place of the Christians
or heretics they become ill, they confess and receive
communion according to their rite, and the dead are
consigned to earth according to the rite of their church.
And in whatsoever place of the Christians or heretics
they happen to be on a Sunday, there they go in the
morning to church with trumpets and musical instru-
ments, with their wives and their children, from the
youngest to the oldest, and attend church while still at
fast, and similarly adore God in great humility, and
there cause a Mass to be sung of the Three Kings, that
God for their deserts may lead them safely all the week
through the deserts, fields and mountains, and preserve
them from all the danger of worms and animals."^
This famous book is found in innumerable versions,
where the name of the Gypsies is written mandopoli,
* C. Horstmann, The Three Kings of Cologne, London 1886, p. 287 flf.

mandopolijn, mandeopolos, mandropolos.^ The pass-

age on the MandopoU occurs in Sudheim's De itinere
Terre Sancte, written about the year 1340, but in a
somewhat abbreviated form: "The MandopoUni or
Mandindes Kve according to no law, but they call them-
selves Egyptians and say that they are of the race of
Pharaoh. They are first-class thieves, travel with their
wives from place to place, making sieves and similar
things, and paying no attention to the heat of the sun.
With the Greeks they are Greeks, with the Saracens,
Saracens, and thus with others. And if one is caught
with his wife, he does not get angry, but, if he can, he
retaliates in a similar way."^
In 1322 Symon Simeonis saw Gypsies in Crete:
"We saw here a people that was living outside of the
town according to the Greek rite and that professed to
be of the race of Chaym, who rarely or never remained
more than thirty days in one place but, as though cursed
by God, roved after thirty days from field to field with
their small oblong black tents, in the manner of the
Arabs, and from cavern to cavern, because the place
which has been occupied by them for the above men-
tioned time becomes full of worms and other uncleanli-
ness, with which it is impossible to stay."^
To this identification of the Gypsies with Ham, as in
the Cave of Treasures, must be added a quotation from
the XII. century German Bible paraphrase, in which
the Gypsy blacksmiths are identified with the Ismael-
ites: "Hagar bore a son from which came the black-
smiths (chaltsmide) When Hagar bore the child, she
.
called him Ismael, whence came the Ismaelites who

travel far over the land, and whom we call chaltsmide.
1 T. Scheepstra, Van den heilighen drien coninghen, Groningen 1914, p.
278.
" L. de Sudheim, De itinere Terre Sancte, in Archives de I'Orient latin, Pans
1884, vol. II, part II, p. 375.
' J. Nasmith, Itinsrarium Symonis Simeonis, ei Hugonis Jllummatorts ad
Terrain Sanetam, Cambridge 1778, p. 17.
Woe Everything which they have for sale
to them!
has always some fault. Let him buy anything well or ill,
he always wants something to boot. They deceive the
people with whom they have dealings. They have
neither house nor home, and they are everywhere
equally at home. They roam over the land and
oppress the people. They deceive the people, and
never rob them openly."^ These Gypsies are here
called chaltsmide from their occupation as smiths. It
is generally assumed that the word means "cold-
smiths," that is, "hammerers of cold metal," but this is
not very certain. It is far more likely that the first
part of the word is identical with German Kdlte "a
tumbler" or "dish" of some kind, or Italian caldaia "a
vase" or "pot," and that thus the whole means "tin-
ker." This is the more likely since we not only have
Italian calderaio, French chaudronnier, who are some-
times identified with the Gypsies, but Modern Greek
xaTîj3eko<: "Gypsy" is similarly derived from xaTQi^sXa
"dishes," and this from Italian cazzuola "ladle, trowel."

From the above discussion it follows that, with the
exception of the last reference, every other mention of
the Gypsies, from Charlemagne in 789 to The Three
Kings of Cologne, at the end of the XIV. century, goes
back to the magicians of Jannes and Mambre and to
Ham and the Egyptians in the Cave of Treasures. It
does not appear that the Gypsies were found in any
considerable numbers in the west, since all the accounts
locate them in the Byzantine Empire. They were, how-
ever, settled in the northern part of the Balkan Penin-
sula, and in the XIV. century sedentary Gypsies are
frequently recorded in Hungary,^ whence they began
their westward movement in 1417, unquestionably
' J. Diemer, Genesis undExodusnach der Milstater Handschrift, Wien 1862,
vol. I, p. 36.
2 See my
article Ismaeliies, in Journal of the Gyvsy Lore Society, new series,
vol. IV, p. 83 fl.
under the pressure of the Turkish invasions, which had

been begun in the previous century. The later history
of the Gypsies^ is of no interest to us here, except that
it shows that in a thousand years the Gypsies have
changed httle from their original state. Their in-
fluence upon social conditions in Europe has upon the
whole been small, since they had nothing to add to the
store of European civilization.
^ For a full bibliography on the subject, see G. F. Black, A Gypsy Bibli-
ography, London 1914.
CHAPTER III.
The Gypsies in Africa.
The Gypsies of Persia are known by several names,

among -wliicli are tlie dhingar "blacksmiths, tinkers,"
and gerhal-hand} "sieve-makers," with whom we have
become acquainted in the European sources. Ahingar
is the Modern Persian form of Pehlevi asingdr, itself for
atsingdr, hence in the Modern Syriac jinganih "Gypsy"
we see a corrupted Pehlevi dslngar, even as the European
Gypsy words, such as German Zigeuner, are similar
corruptions of the original Pehlevi word. The Syrians
also have the designation rumeli and kurbdt for the
Gypsies. The first is unquestionably derived from the
Gypsy romani "Gypsy."
The most common word in the early Arabic authors
of Egypt for Gypsy is -^J >i j^^ da'r, dza'r, za'r, also
. >
J^j dzu'dr.^ The first means "vile," but the great

variety of forms indicates that this is merely a popular
etymology and that the original form was more nearly
dzigar, that is, a word derived from dtslngdr. A
Modern Arabic form is j^ gagar, no doubt another
transformation of the same word. Al-Maqrizi fre-
quently refers to a class of people with Gypsy charac-
teristics as \-^J ramadlah or rumaiiah, and Quatrem&re
is unquestionably right when he identifies them with
' Capt. Newbold, The Gypsies of Egypt, in The Journal of the Royal Asiatic
Society of Great Britain and Ireland, vol. XVI, part II, p. 310.
* R. Dozy, Dictionnaire dUailU des noms des vitements chez les Arabes,
Amsterdam 1845, p. 259.
.
THE GYPSIES IN AFRICA 79
our Gypsies.^ The word, like Syriac rumeli, is again

our Gypsy romani.
The Egyptian gaârs are conscious of their relation-
ship to the Hungarian Gypsies. They are tinkers and
blacksmiths, vend earrings, amulets, bracelets and
instruments of iron and brass, and, above all, are
thieves.^ There is also a class of Gypsies known as
Nuris or Ndwars, which Quatrem^re takes to be from
Arabic Jy nur "light," since among the XIII. century
writers onEgypt we frequently hear of 't^'^ maâllah
"executioners," whose name was derived from Arabic
tK-* mas'al "a lantern," which they carried before
them.^ The scanty vocabulary of the Nawars, as
recorded by Newbold,* shows Gagar affinities.
The greater number of the Egyptian Gypsies belong to
the cyr^ lielebl, who, like the rest, "are looked down
upon with almost the same horror as the Pariahs of
India by the Brahmans."^ "The male Helebis are
chiefly ostensible dealers in donkeys, horses, camels,
cattle, &c., and pretend to great skill in the veterinary
art; but their character for common honesty does not
stand very high in the estimation of those who know
them best. With their women, they lead a vagabond
but return to the towns at stated periods.
life, Their
wanderings are confined to the Rif or valley of the Nile, ,
and to the Delta, rarely extending far into the desert,

except when they go forth to meet the Hdjj, on its re-
turn from Mecca, in order to cheat the way-sick pil-
grims out of their jaded beasts, or to sell cattle-medi-
cines. Some few accompany the Hdjj all the way to
1
M. Quatremere, Histoire des sultans mamlouks, Paris 1837, vol. I, part
II, p. 5 f
2 Newbold, op. cU., p. 292 ff.
s
Op. cit., p. 4 f.
4 Op. cit., p. 295 ff.
5 IWd., p. 285.
Mecca; and, having performed the pilgrimage, are
proud of prefixing Hdjji to their names a title, how- —
ever, which among the more experienced Cairenes is
supposed to add but little to a man's credit in the
ordinary dealings of life. 'If your neighbour,' say
they, 'has performed one hdjj, be suspicious of him; if
two, avoid him; but if three, then by all means give up
your house immediately, and seek another in some re-
mote quarter.'
"The Helebis usually live in tents or kheish (portable
huts), which they pitch on the outskirts of some large
village or town. Near Cairo they are to be found at
certain seasons (chiefly during the winter and spring),
near a village on the right of the road from Cairo to
ShUbra. They are expert in disguises, and hardly
yield the palm to their brethren in Europe in cunning
and deception."^ "The female Helebis (the Fehemis),
as before stated, practice palmistry and divination.
During their halts on the outskirts of towns and
villages, and in roaming about the streets, bazars,
and coffee-houses, in different disguises, they contrive
to pick up, with wonderful tact and accuracy, the
information necessary to their vocation, regarding the
private history and prospects of persons with whom
they are thrown in contact. In this secret intelligence
department they are also assisted by their male rela-
tions, who, it is said, are to be found in every of&cial de-
partment in Egypt, though not known to be gypsies;
and, at all events, mingle much both with residents on
the spot, and with strangers in the coffee-houses and
caravanserais.
"Practice of Palmistry and Divination. —
In practis-
ing the art of palmistry, the Fehemi takes the right
hand of the inquirer into the book of destiny into her
own, holding it by the tips of the fingers, which she
1 IMd., p. 286 f.
often bends gently back, so as to render the lines on the

palm more distinct. Muttering some spell, she looks
gravely and earnestly into these lines for a moment or
two; and then raising her penetrating eyes, fixes them
steadily on those of the fortune seeker, gazing into them
as if reading his destiny, written in large characters, at
the bottom. She then unfolds to him the result with
much decision and emphasis. The tale she tells is very
much like what the gypsy women impart to the nursery
maids and young lads on Blackheath. There are the
different dangers and felicities awaiting them at
different epochs of life— the dark or light lady, or gentle-
—
man, who is to love and be loved the jealous enemy of
—
whom they are to beware the number of children they
are to have, &c. It is almost unnecessary to add that
in most cases the weight of the silver coin, with which
the sybil's hand must be crossed, exerts a corresponding
influence over the future (silvery or coppery, as the case
may be) aspect of the aspirant's fortunes.
"In divination, the Fehemi seats herself on a mat or
carpet at the foot of the divan, or on the floor, and
empties her gazelle-skin bag of a portion of its contents,
viz., small shells, broken bits of glass, small coloured
stones of agate, jasper, basalt, &c.; coloured bits of wax,
&c. She throws the shells repeatedly on the carpet,
after much jugglery, grimace, repeating spells, &c. and;
from the position they chance to lie in she draws her

inferences, much in the same way as the servant girls in
England tell their fortunes from the arrangement of
the grounds of tea at the bottom of their cups.
"On one occasion the shell, which is supposed to rep-
resent the person whose fortune is being told, happened
to fall in the centre of a circle formed by the other shells
being accidently ranged round it. This answer to the
question, which was: 'Will his friends prove faithful in
the hour of need?' was interpreted as highly favourable.
"Thus the Fehemi goes on casting the shells and di-
vining from them. Money is required at various
stages of the operation, and the farce usually concludes
with the gypsy's presenting a few bits of coloured stone
or wax to her employer as charms.
"I witnessed a curious trick played by one of the
Fehemi women near Cairo in this sort of divination.
— —
She put one of the shells a small cowry into a basin of
clear water, which was placed on the carpet of the floor,
at the foot of the divan where a friend and myself were
seated, enjoying our chibouqes and coffee. She then
covered the basin with a cloth, and directed me to repeat
after her an invocation in Arabic, and, while doing so,
retired a few feet from the basin, after taking off the
cloth, to the edge of the carpet. The shell was seen
lying under the water, at the bottom of the basin as be-
fore; but no sooner was the invocation finished than
the water bubbled up, and the shell was shot out to the
distance of several feet, with some of the water, with a
slight explosion, like that of a percussion-cap thrown
into the fire.
"This, doubtless, was the effect of some chemical sub-
stance, placed probably in the shell itself; but whether
the secret of its preparation be a remnant of the art of
ancient Egypt, or vended to the gypsies by some itiner-
ant charlatan from Europe, is doubtful. The last
appears the more reasonable hypothesis.
"The Fehemi women, as well as the men, have a
family resemblance to the Kurbats of Syria. They are
noted for their chastity, in contradistinction to the
Ghagar women. Intrigues, however, have happened,
but, if discovered, they are punished with death; the
woman being usually thrown into the Nile, with a bag of
stones tied to her neck.
"Until their marriage the young Fehemi females wear
a cincture of silk or cotton thread round their loins, in
token of virginity. They never intermarry with the

Arabs, Copts, or other inhabitants of Egypt. In this
respect they are as rigid as the Hindus. They are not
remarkable for cleanliness either of person or apparel;
in this respect, and their passion for trinkets of brass,
silver, and ivory, they remind one of the Brinjari
women of India.
"They are remarkably intelligent, quick in gaining
information, and would make capital spies in an enemy's
camp. An instance of their shrewdness in this respect
fell under my
own observation. Passing their en-
campment one day, I persuaded my
companion to stop
and have his fortune told; to which, after some demur,
he at last consented. While the gypsy woman was
looking at the Lines of his hand, I took the opportunity
of inspecting the interior of their tents. They re-
sembled those of the common Bedouin of the desert,
and contained little beyond some wretched horse and
donkey furniture, pots, pans, &c. Everything ex-
ternally denoted the most squalid poverty, excepting
only an enormous mess of fowls, mutton, and savoury
vegetables, seething in a large iron cauldron over a wood
fire; and which proved, to more senses than one, that
the care of the flesh-pots of ancient Egypt had not de-
volved on a race insensible to their charms. On re-
turn, I found my companion still in the hands of the
gypsy, now listening to her tale with as much serious-
ness in his face as there was merriment and mockery
before.
"When she had finished, he told me that he had been
perfectly astounded in hearing from her lips a circum-
stance which, to the best of his recollection, he had
never divulged to any person; but which, no doubt,
must have on some occasion inadvertently escaped
him."^ "The Helebis do not give their daughters in
1 Ihid., p. 287 ff.
marriage to the Ghagars, though they occasionally
marry Ghagar dam.sels. The hug or dilk (zone of
chastity) is often made of plaited leather, like the
waist-covering of the women of Soudan, and is cut off
on the wedding night.
"The Helebi females, though chaste themselves, oc-
casionally do not scruple to act as procuresses of Gentile
or Husno women, and will even sometimes expose their
own persons for a reward. The Arabs and Copts
charge them with kidnapping children; but this they
strenuously deny, as well as the common accusation of
their eating cats and dogs, and other animals held in
abhorrence by Moslems.
"They bury their dead, but have no fixed places of
interment."^
The collective name of the helebi is u^M^^ mahlibaS,
which at once leads to Greek XdXu^sz, the word which
we found glossed in Charlemagne's law of 789 as "nudi
cum ferro," hence the Gypsies in Egypt go back at
least to the Greek occupation, after the Persians, as we
have already found from other considerations. The
lengthening of t^r^ to u^^^ is universal in Arabic and
shows its un-Arabic origin. We have Arabic ts^
hilllba "deceit, guile," v^^ halib, ^j^ halub "deceitful,"
"^.^ hallabah "enchanter," and, at the same time, u^.}^
halbu^ "servant of the almeh, buffoon," that is, we
come back to our Gypsies. "The Helebis pretend to
derive their origin from Yemen or the Hadramat; and
assert that the early history of their race is chronicled
in a written record, called the Tarih az-Zir, which, as
far as I can glean, is an obscure and unsatisfactory docu-
ment. From Yemen, they say, their tribes were ex-
pelled by the persecutions of Zir, a king of the Tuba
1 IHd., p. 293.
race; and wandered over Syria, Egypt, Persia, and

Europe. The seven brother chiefs of the tribes which
migrated into Egypt obtained from its sovereign the
privilege of exemption from taxes, and of wandering
about the country without molestation. The tombs of
these seven chiefs are regarded by the Helebis as holy
places to this day. Two of them are said to be in the
Bahriyeh district, one in the Kelyubiyeh, and the rest in
the Syud."i
The Helebis, Gaârs, and Nawars, no doubt, repre-
sent different migrations of the Gypsies, and there is no
reason to cast any doubt on the assertion of the Helebis
that they came from Arabia. Indeed, the Helebis are
found in Arabia under the name of v^ ^leb, a slightly
changed form of the Egyptian name. Oppenheim saw
them in northern Arabia: "Not far from the Hufne
there were encamped about twenty Sleb, who belong to a
remarkable tribe, who in their appearance and manner
of life in many ways differ from the Arab Bedouins.
Generally less tall than these, they subsist chiefly on
the chase of gazelles, the skins of which they often wear
as clothing. The women, to begin with, are remark-
able for their wonderftd beauty. I saw in Ruhbe a
Slebiye of moderate height, with a tawny complexion,
small face, melancholy, but flashing eyes beneath long
lashes, superb, straight, black hair, and magnificent
teeth. The Sleb seldom own horses or camels, and
generally ride on small donkeys of great endurance, and
occasionally keep a few sheep and goats. The Sleb
live in peace with nearly all the desert tribes, and, in
exchange for the hospitality which they receive on all
sides,and for the immunity to their scanty property,
they everywhere offer their services as guides, for no
one knows the desert so well as the Slebi. At certain
1 Ibid., p. 291.
occasions the Sleb men and women dance for the Be-
douins, for which they receive presents. The ethno-
graphical origin of Sleb has not yet been fully estab-
lished, but in all probability they are not Semites, but
of Hindu origin. It is said that the later Khalifs
had Indian musicians sent to their court at Bardad,
who, upon the arrival of Tamurlan, wandered into the
desert. I have heard the opinion that the modern
Sleb are descendants of those musicians. At the pres-
ent time the Sleb are to be found from Aleppo to the
Persian Gulf, south of the Euphrates. Whereas the
Bedouins wander about by tribes, the Sleb roam only by
families, sometimes only by twos or threes, and occasion-
ally even without tents, but they bury their dead, if
possible, only in large common cemeteries. They to a
certain extent may rightfully be called 'the Gypsies of
the desert.' "^
Pelly^ gives other current notions about the ^leb,
from which we only learn that they were held in con-
tempt by the Bedouins: "A few miscellaneous re-
marks on the Selabah or Selaib tribe, based on informa-
tion collected among themselves, may be interesting.
"The called Seleb or Selaib, because on cer-
caste is
tain festivals, and particularly on occasions of marriage
and circumcision, they fix a wooden cross, dressed in red
cloth and adorned at the top with feathers, at the door
of the person married or circumcised. At this signal
the people collect and dance round the cross. They
have a particular dance. The young men stand op-
posite their female partners, each advances, and the
youth slightly kisses the shoulder of the maiden: any-
thing like touch of the hand or waist is out of etiquette.
' M. Oppenheim, Vom Mitielmeer zum Persischen Golf, Berlin 1899, vol. I,
p. 220 f.
^ L. Pelly, AVisit to the Wahabee Capital, Central Arabia, in The Journal

of the Royal Geographical Society, vol. XXXV, p. 189 ff.
"The word Seleeb means a cross. But some of the

caste derive their name from As-Solb-Al-Arab, i.e., from
—
the back of the Arabs meaning to assert that they are
pure descendants of aboriginal Arabs. The Moham-
medans, on the other hand, stigmatise them as out-
castes. The tradition is that, when Nimrod was about
to castAbraham into the fire, some angels appeared and
protected him. Eblis or Satan then made his appear-
ance and pointed out to the bystanders that if someone
would only commit a shameful crime, the angels would
be obliged to depart, and thus Abraham would be left
unprotected. Upon this one of the Arabs lay with his
own mother, and forthwith the angels fled. Upon this
the angel Gabriel came to the rescue, and changed the
spot where the fire was kindled into a garden. The des-
cendants of the man who lay with his mother were,
thenceforward, called As-Selaib.
"The Selaib who have emigrated
into Nejd and other
Mohammedan settlements conform outwardly to the
religious rites and ceremonies of the dominant creed.
But in their own tents, or when alone, they do not so
conform.
"No intermarriage takes place between the Selaib and
the Arabs. Even a Bedouin will not stop to plunder a
Selaib, nor to revenge a blood feud against him. The
Selaib are capital sportsmen. They live largely on
deer's flesh, and wear a long shirt of deerskin coming
down to the feet. Their common diet is locusts, and
dates when procurable; but they will eat anything.
They tend their sheep and camels, wander for pasturage
during eight months of the year, and for the remainder
seek some town or village where to exchange their pro-
duce for necessaries of life. Their tents are black, of
goat's hair, and are pitched separate from those of the
Arabs. The Selaib are filthy in appearance; but the
Arabs confess that, in point of features, the Selaib
women are the most beautiful among them.
"Forty days after birth a child must be washed, being
dipped seven times in water.
"Marriage is contracted by mutual consent of the
parties. The assent of the father, or failing him, of the
nearest of kin, must also be obtained. The father of
the girl receives some sort of payment, according to the
ability of the bridegroom. The parties go before a
moUah, or an elder of the tribe, who asks them three
several times if they freely consent to the union.
The parties replying in the affirmative, the moUah takes
his fee, and they cohabit. The neighbours then collect
at the tent, sheep are killed for them, and they dance.
The only invitation is the sign of the cross fixed outside
the tent.
"The Selaib wash their dead, cover the body with a
white shroud, and inter it with a prayer. Failing a
white shroud, they use a new shirt of deer-skin.
"They profess to reverence Mecca, but state that
their own proper place of pilgrimage is Haran, in Irak
or Mesopotamia. They say also that their principal
people have some psalms and other books written in
Chaldean or Assyrian. They respect the Polar star,
which they call Jah, as the one immovable point which
directs all travellers by sea and land. They reverence
also a star in the constellation, called Jeddy, cor-
responding with Aries. In adoring either of these
heavenly bodies the Selaib stands with his face towards
it, and stretches out his arms so as to represent a cross
with his own body. They believe in one God. Some

of them pretend to believe in Mohammed. Others
deny the prophet, but trust in certain intermediate
beings, who are called the confidants of God. They
pray three times a day: first, as the sun rises, so as to
finish the prayer just when the entire disc is above the
horizon; secondly, before the sun begins to decline from

the meridian; and thirdly, so as to finish the prayer as
the sun sets. It is asserted, however, that the Selaib of
Haran have pure forms of prayer, in the Assyrian or
Chaldean. They fast three times a year: for thirty
days in Ramadan; for four or seven days in Shaban, and
for five or nine days in a summer month. They are
peaceful, and are undisturbed by the Arabs, who hold
them below injury. They are markedly hospitable,
like all people who have nothing to give. They assert
themselves to be a tribe of the Sabians emigrated to
Nejd. The Mohammedans deny this. The Selaib
eat carrion and profess themselves to be the chosen
people of God, who pay no tribute or tax, since no one
will deign to receive it from them."
The fullest account of these Arabic Gypsies is given
by Doughty:^ "As we went by to the mejlis, 'Yonder
(said Zeyd) I shall show thee some of a people of antiq-
uity.' This was a family which then arrived of poor
wanderers, Solubba. I admired the full-faced shining
flesh-beauty of their ragged childrien, and have always
remarked the like as well of the Heteym nomads.
These alien and outcast kindreds are of fairer looks than
the hunger-bitten Beduw. The Heteym, rich in small
cattle, have food enough in the desert, and the Solubba
of their hunting and gipsy labour for they are tinkers of
:
kettles and menders of arms, in the Beduin menzils.

They batter out upon the anvil hatchets, jedum, (with
which shepherds lop down the sweet acacia boughs, to
feed their flocks,) and grass-hooks for cutting forage,
and steels for striking fire with the flint, and the Uke.
They are besides woodworkers, in the desert acacia
timber, of rude saddle-trees for the burden-camels, and
of the theim saddle-frames, of pulley reels, {mdhal) for
drawing at any deeper wells of the desert, also of rude
' C. Doughty, Travels in Arabia Deseria, Cambridge 1888, vol. I, p. 280 ff.
milk vessels, and other such husbandry: besides, they
are cattle surgeons, and in all their trade (only ruder
of skill) like the smiths' caste or Sunna. The Solubba
obey the precept of their patriarch, who forbade them to
be cattle-keepers, and bade them live of their hunting in
the wilderness, and alight before the Beduin booths,
that they might become their guests, and to labour as
smiths in the tribes for their living. Having no milch
beasts, whereso they ask it at a Beduin tent, the house-
wife will pour out 16ban from her semlla, but it is in their
own bowl, to the poor Solubba: for Beduins, otherwise
little nice, will not willingly drink after Solubbies, that
might have eaten of some futts, or the thing that is dead
of itself. Also the Beduw say of them, 'they eat of vile
insects and worms:' the last is fable, they eat no such
vermin. Rashly the evil tongue of the Beduin rates
them as 'kuff^r', because only few Solubbies can say
the formal prayers, the Beduins are themselves not
better esteemed in the towns. The Solubba show a
good humble zeal for the country religion in which they
were born, and have no notice of any other; they are
tolerant and, in their wretched manner, humane, as
they themselves are despised and oppressed persons.
"In summer, when the Beduw have no more milk,
loading their light tents and household stuff, with what
they have gained, upon asses, which are their only cattle,
they forsake the Aarab encampment, and hold on their
journey through the wide kh&la. The Solubby house-
hold go then to settle themselves remotely, upon a good
well of water, in some unfrequented wilderness, where
there is game. They only (of all men) are free of the
Arabian deserts to travel whithersoever they would;
paying to all men a petty tribute, they are molested by
none of them. Home-born, yet have they no citizen-
ship in the Peninsula. No Beduwy, they say, will rob
a Solubby, although he met him alone, in the deep of the
wilderness, and with the skin of an ostrich in his hand,

that isworth a theltil. But the wayfaring Beduwy
would be well content to espy, pitched upon some lone
watering, the booth of a Solubby, and hope to eat there
of his hunter's pot; and the poor Solubby will make the
man good cheer of his venison. They ride even hunt-
ing upon ass-back. It is also on these weak brutes,
which must drink every second day, (but otherwise the
ass is hardly less than the camel a beast of the desert,)
that they journey with their families through great
waterless regions, where the Beduwy upon his swift and
puissant thehil, three days patient of thirst, may not
lightly pass. This dispersed kindred of desert men in
Arabia, outgo the herdsmen Beduw in all land-craft, as
much as these go before the tardy oases villagers. The
Solubba (in all else ignorant wretches,) have inherited a
land-lore from sire to son, of the least finding-places of
water. They wander upon the immense face of Arabia,
from the height of Syria to el-Y6men, beyond et-Tdif,
—
and I know not how much further! and for things
within their rat-like understanding, Arabians tell me, it
were of them that a man may best enquire.
"They must be masters in hunting, that can nourish
themselves in a dead land; and where other men may
hardly see a footprint of venison, there oftentimes, the
poor Solubbies are seething sweet flesh of gazelles and
bedfin, and, in certain sand districts, of the antelope;
everywhere they know their quarries' paths and flight.
It is the Beduw who tell these wonders of them; they
say, 'the S'lubba are like herdsmen of the wild game,
for when they see a troop they can break them and
choose of them as it were a flock, and say, "These will
we have to-day, as for those other heads there, we
can take them after to-morrow".
'
—
It is human to
magnify, and find a pleasant wonder, this kind of large
speaking is a magnanimity of the Arabs; but out of
doubt, tlie Solubba are admirable wayfarers and hardy
men, keen, as living of their two hands, and the best
sighted of them are very excellent hunters. The Sol-
ubba or SUyb, besides this proper name of their nation,
have some others which are epithets. West of H^yil
they are more often called el-Khlua or Kheluiy, 'the
desolate,' because they dwell apart from the Kabdil,
—
having no cattle nor fellowship; a word which the
Beduw say of themselves, when in a journey, finding no
menzil of the Aarab, they must lie down to sleep
'solitaries' in the empty khala. They are called as well
in the despiteful tongue of this country, Kiiab el-KhS,la,
'hounds of the wilderness. ' El-Ghrunemy is the name
of another kindred of the Sl^yb in East Nejd; and it is
said, they marry not with the former. The Arabians
commonly suppose them all to be come of some old
kafir kind, or Nas^ra.
"
—
Neither are the Sherarat and Heteym nomads
(which are of one blood) reckoned to the Beduin tribes.
The dispersed kindreds of Sunna are other home-born
aliens living amongst the Aarab, and there is no marry-
ing between any of them. Ma li-hum asl, say the
Beduw, 'They are not of lineage, which can be under-
'
stood to signify that 'not descended of Kaht^n, neither

of the stock of Ishmael, they are not of the Arabs.'
And if any Arabians be asked, What then are they?
they answer: 'Wellah, we cannot tell but they come
of evil kin, be it Yahtid or Nas§,ra' (this is, of the
Ancients which were in the land before Mohammed, and
of whom they have hardly any confused tradition).
As often as I met with any Solubba I have asked of
their lineage: but they commonly said again, wonder-
ing, What is this to enquire of us mesquins dwelling in
'
these deserts? we have no books nor memory of

things past: but read thou, and if anything of this be
written, tell us.' Some said the name of their an-
cestor M'aibt; the Beduw also tell of them, that

is
which read in Arabic authors, how they were the
is
Aarab Jessds, once Beduins: being destroyed in their
controversy with the Aarab K'leyb and bereaved of all
their cattle, they for their livelihood took up this trade
of the hammer, and became Solubba. Later in the
summer I found some Solubba families pitched under
the kella at el-H6jr, who were come over the Harra and
the Tehama from Wejh, their own station. At that
season they make a circuit; last year they had wandered
very far to the south, and I saw their women grinding a
minute wheaten grain, which they had brought from a
wady near Mecca! They (as coast and Hejfi,z dwellers)
were of more civil understanding than the uplandish
Solubba. To my questions the best of them answered,
'We are Aarab K'f^, of old time possessors of camels and
flocks, as the Beduw: those were our villages, now ruins,
in the mountains southward of el-Ally, as Skeirdt in
Wady S6dr; but at last our people became too weak to
maintain themselves in an open country, and for their
more quietness, they fell to this trade of the Solubba.
Said one of them, 'We are all Beny Murra, and fellow-
ship of Sdlim Ibn ez-Zir, from the hill Jemla, a day on
the east side of Medina; we are called Motullij and
Derruby.' Haj Nejm laughed as I came again, at
'this strange fantasy of Khalll, always to be enquiring
somewhat, even of such poor folk. Khalll! these are
the Beny Morr, they are dogs, and what is there besides
to say of them?'
"When Beduins asked me if I could not tell them by
book-craft what were the Solubba, it displeased them
when I answered, 'A remnant, I suppose, of some ancient
Aarab they would not grant that Solubbies might be
;
'
of the right Arabian kindred. All who are born in the

Arabs' tongue are curious etymologers; a negro, hear-
ing our discourse, exclaimed, 'Well, this is likely that
KhalU says; not Solubba to say Sulb el- Arab, the
is
Arab's stock?' The poor soul (who had spoken a

little in malice, out of his black skin, for which he was
dispraised among the white Arabs) was cried down by
the other etymologers, which were all the rest of the
company, and with great reason, for they would not
have it so. 'The Solubba are rich (say the Arabs),
for they take our money, and little or nothing comes
forth again; they need spend for no victuals. They
have corn and dates enough, besides samn and mereesy,
for their smith's labour.' The Solubby has need of a
little silver in his metal craft, to buy him solder and
iron; the rest, increased to a bundle of money, he will,
they say, bury in the desert sooner than carry it along
with him, and return perhaps after years to take it up
again, having occasion it may be to buy him an ass.
Yet there are said to be certain Solubba, keepers of a
few cattle, towards Mesopotamia; living under their
own sheukh, and riders upon dromedaries. I have
seen a sheykhly northern man, honourably clad, at
HS,yil, who was a Solubby; he invited me (I think at the
great Emir's bidding) to ride with him in the next
mountains, seeking for metals. I asked, 'Upon what
beast? ' He said I should ride upon an ass, 'we have
no other.' I would gladly have ridden out of H§,yil
into the free air; but I thought a man's life was not to
trust with abjects, men not of the Beduin tradition in
faithful fellowship. Even the Solubba hold to circuits,
and lodge by their tribes and oases. There are Solubby
families which have their home station, at some settle-
ment, as Teyma; but the most remain in the desert.
The Sunna are some settled in the villages, and some
are wandering men with the tribes, leading their lives
as nomads, and possessors of cattle. The Solubba
outcast from the commonwealth of mankind, and in
disgrace of the world, their looks are of destitute humil-
ity. Their ragged hareem, in what encampment they

alight,will beg somewhat, with a lamentable voice,
from beyt to beyt, of the poor tolerant Beduw: yet
other (as those from Wejh) are too well clad, and well-
faring honest persons, that their wives should go a-mum-
ming. I have seen young men, which were S16yb, in
the Syrian wilderness, clad in coats of gazelle-skins.
The small Solubby booth is mostly very well stored,
and they have daily meat to put under their teeth,
which have not the most poor Beduins."
Leaving out all conjectures by the different authors,
and the ^leb varying accounts of their origin, we are
left only with the positive fact that the ^leb are either a
pure or mixed race of Gypsies. What a pity that no
study of their dialect has been made! It would be
interesting to know whether any Gypsy words are left
among them. Considering the fact that the Egyptian
Gypsies, especially the Nawars, have but a few words of
Indian origin left in their vocabulary, it is not at all im-
possible that no such words would be preserved in
Arabia, even as Gypsies may be found in Europe who
no longer speak Romani.
Gypsies are found as far south as Abyssinia, but un-
fortunately the information about them is scanty:
"I saw here the first gipsy encampment in Abyssinia,
curious people with a red brown complexion, long
straight black hair with regular gipsy features. The
Abyssinians disUke them and believe they are capable
of doing all sorts of mischief by magic and other means.
They had with them a lot of waterproof grass baskets
and wooden bowls and platters, which they manufac-
ture and sell at the markets. ..they live by catching
animals and they have the reputation of being great
thieves, helping themselves at night time to the growing
crops; in habits therefore they resemble the English
gipsies. There are only a few bands of them left in the
;

country, and I regret that Hailou would not allow me to
enter into conversation with them; he pulled out his
crucifix from the inside of his shirt and held it between
"^
himself and them until he got out of their sight.
It cannot always be said that these pariahs are of
Hindu origin, because intermixtures have taken place
in the past, and even where totally unrelated races have
been reduced to abject servility, it is remarkable that
under Arabic influence the profession of the blacksmith
has been connected with this low estate. Such is the
case with at least some of the low caste Somalis: " The
great Somali social order is divided in two large ethnic
groups distinctly separated in the two component races.
The pure Somali race, called Gob, consists of various
tribes, which, although tradition assigns to them differ-
ent origins, differ very little in their social organization,
and the impure race of the pariahs. Gum, consists of
the proscribed and rejected people, known under the
name of Tumal, Midgan, and Yibir. The population
of the first group. Gob, may conveniently be divided
into three classes, namely: 1) Sedentary Somalis, a
small class that lives in the coast villages, or nearby, or
connected with confraternities (the most civilized);
2) People occupied in trade, who take large caravans
from the interior to the coast, during the commercial
period of cool weather, and vice versa (merchants,
hence few in number, but real vehicles of progress)
3) Nomad Somalis, who keep cattle, goats, sheep, and
camels, and raise horses and oxen, and lead the lives of
Bedouins, roaming through the bush according to the
rainy season in search of pasturage for their beasts
(shepherds, naturally the most restive). In the second
group, the Gum, the origin of the subcastes is unknown.
They are easily distinguished from the true Somalis
by their ignoble appearance, their less proud and bold
' A. Wylde, Modern Abyssinia, London 1901, p. 338 f.
aspect, their coarse features, but they are not slaves.

They are not organized, and live by families, confusedly
scattered in the whole country amidst the known tribes.
The Tumal (also known under the name of Tum-tum)
are exclusively blacksmiths, manufacturers, and manip-
ulators of all kinds of weapons, such as arrows, lances,
knives, horse bits, hooks, daggers, and all kinds of small
utensils for the people. The Midgan, who resemble
the Somalis, but are of smaller stature, are the most
numerous among the proscribed class, and have no
especial occupation, and usually enter service, drive
cattle, and do similar things. They are armed with a
bow and poisoned arrows, which they carry in a quiver,
and a small knife, and generally Uve by the chase.
Some of them raise certain wild dogs, which they train
for the chase. Their chief game is the oryx, a large
ox-like antelope, which is provided with horns of half to
a whole meter in length. These two classes intermarry,
and they marry occasionally, but very rarely, some low
caste Somali. The Midgan are often used in the
tribal wars as messengers, explorers, ambushers and
excellent spies. The Yiber, the lowest class, are con-
sidered even more abject than the poor and despised
Tumal and Midgan. These Yiber, avoided and de-
spised by all, Uve by themselves, begging here and there
for a miserable bit, living on anything that they find,
often on the flesh of animals fallen dead on the road.
For garment they wear a skin rag over their loins, and
another very small one which serves them for a mantle
over their shoulders. They bear neither lance nor
shield and, like the Midgan, always walk armed with
a bow and poisonous arrows and a long double-edged
knife, which they carry tied to the waist by a strap.
They are leather-workers and make saddles, sheaths,
bags, straps, amulets, etc. They are nearly always
charlatans and mystifiers by profession, act as buffoons
on holidays, weddings and other solemn occasions. It
issaid that often they sell their children, and that they
are not allowed to follow the Somalis to war. It is
customary for the first Yiber who arrives in a kraal,
where a boy is born, to receive from the family of the
new-born child a small present, which may be a piece of
cotton ware or a kid, or lamb, or other small thing.
Such a customary present in these circumstances is
called samanyo, and he who receives it gives to the
donor in return, as a proof of such a receipt, an amulet,
called macran, which consists of a small piece of wood
or bark of a certain tree, enclosed in a bit of kid leather,
which is placed about the neck of the child, so that
when another Yiber passes he may see that the samanyo
has been paid and the macran received. If the new-
born child is a girl, no present is given. Similarly the
Yiber receive the same kind of present from every
youth who gets married, when they give another small
macran of the same kind, which the bride keeps in the
house for a whole year, after which it is thrown away.
The families of the Yiber, now small in number, are
'
steadily disappearing. ^ '
The structure of society in the Western Sudan is in

essence identical with that of the Somalis just studied.
Among the Malinkes, of whom the majority are Mos-
lems, the tribe is divided into two portions, those who
are free, and those who are captives, while the free
portion is again subdivided into five castes:
"(1) The Horos, who are citizens.

(2) The Sohres, weavers.
(3) The Garangis, or shoemakers.
(4) The Hrabis, or blacksmiths.
(5) The Yellimanis, or jesters.
1 R. Bricchetti, Somalia e Benadir, Milano 1899, p. 214 ff.

"The Horos are the only class from which chiefs and
headmen can be selected. They are the predominant
caste, and all the others are their menials.
" Horos can only marry in their own class. The other
people can marry amongst themselves as they please.
"The Hrabis are looked on with great contempt,
corresponding in caste to the sweeper class of India.
It is uncertain what was the origin of this, but there is
a story connected with Mohammed and a blacksmith
which probably accounts for it. It is said that the
Prophet was once pursued by some infidels, and con-
cealed himself in the trunk of a tree near the spot where
a blacksmith was at work. The latter was on the
point of betraying Mohammed's hiding-place when he
was struck blind by God. Mohammed, when he
issued from the tree, is supposed to have cursed the
blacksmith and all his kind.
"The Yellimanis are a very obnoxious class. They
spend their time in abusing those who do not give them
any money, while they sing the praises of their patrons.
Every chief has an entourage of these jesters. They
are often equipped with musical instruments, and form
a sort of band which precedes him wherever he goes. "^
For horo the Malinke dictionary^ gives foro, and
translates it by "free man." This is Arabic j^ hurr
"free," which indicates that the borrowing of the divi-
sions is from the Arabs.
Delafosse^ more justly classes Sudanese society in
three castes, of which the first, the Horos, busy them-
selves with occupations that do not demand any special
training, such as agriculture, fishing, cattle raising,
hunting, war. The second category includes the pro-
^ A. Haywood, Through Timbiidu and across the Great Sahara, London
1912, p. 57 f.
2 [Abiven], Dictionnaire frariQais-malinke et malinke-fransais, Conakry
1906.
' M. Delafosse, Haut-Sinegal-Niger, Paris 1912, series I, vol. Ill, p. 115 flf.
fessions of religion and commerce, that is, the occupa-
tions of the merchant, tailor, weaver, painter, or
Moslem preacher or teacher. The third category
comprises the true castes, and here one finds the workers
in wood, clay, leather, and the metals. Each one of
these forms a special caste, and to these must be added
the two castes of the griots, kinds of buffoons, musicians,
bards and professional dancers, who attach themselves
to kings and famous warriors and extol their exploits.
To these must be added the caste of magicians, doctors,
sorcerers, manufacturers and merchants of talismans,
fortune-tellers, etc.
The oldest reference we possess to the Mandingo

bards is contained in Ibn-Batutah 's account of his
voyage to the African kingdom of Malli: "On the
hohday, after the Sultan Duga has finished his games,
the poets come, who are called guld, in the singular
âll. They make their entrance, every one of them in
the hollow of a figure formed by feathers, which re-
sembles a SigSag, or a kind of sparrow, to which a wooden
head has been attached with a red beak, to imitate the
head of this bird. They stand before the Sovereign
in this ridiculous make-up, and recite their poetry to
him. I have been informed that it consists of a kind of
admonition and that they sây to the Sultan: 'This
penpi on which you are now seated was occupied by
such and such a king, who has done such and such
generous deeds; another has done such and such noble
acts, etc. So you, in jour turn, do likewise that you
may be remembered after death.' Then the chief of
the bards climbs the steps of the penpi and places his
head in the Sultan's lap; then he ascends the penpi it-
self and places his head on the right shoulder, then on
the left shoulder of the Sovereign, all the time speaking
in the language of the country. Finally he descends.
I have been assured that this is a very old custom, pre-
From Reeve's The Gambia.
vious to the introduction of Islam among these people,

in which they have always persisted."^
A
fuller account of them was given by Jobson, in
1623: "There is, without doubt, no people on the
earth more naturally affected to the sound of musicke
than these people; which the principall persons do hold
as an ornament of their state, so as when wee come to see
them, their musicke will seldome be wanting, wherein
they haue a perfect resemblance to the Irish Rimer sit-
ting in the same maner as they doe vpon the ground,
somewhat remote from the company; and as they vse
singing of Songs vnto their musicke, the ground and
effect whereof is the rehearsall of the auncient stocke of
the King, exalting his antientry, and recounting ouer all
the worthy and famous acts by him or them hath been
atchieued: singing Hkewise extempore vpon any occasion
is offered, whereby the principall may bee pleased; where-
in diuerse times they will not forget in our presence to
sing in the praise of vs white men, for which he will ex-
pect from vs some manner of gratification. Also, if
at any time the Kings or principall persons come vnto vs
trading in the Riuer, they will haue their musicke play-
ing before them, and will follow in order after their
manner, presenting a shew of State. They haue little
varietie of instruments, that which is most common in
vse, is made of a great gourd, and a necke thereunto
fastned, resembling, in some sort, our Bandora; but
they haue no manner of fret, and the strings they are
either such as the place yeeldes or their inuention can
attaine to make, being very vnapt to yeeld a sweete and
musicall sound, notwithstanding with pinnes they
winde and bring to agree in tunable notes, hauing not
aboue sixe strings vpon their greatest instrument: In
eonsortship with this they haue many times another
' C. DeMmery and B. Sanguinetti, Voyages d'Ibn Batoutah, Paris 1879,

vol. IV, p. 413 f.
who playes vpon a little which, he holds vnder
drumme
his left arme, and with a crooked sticke in his right
hand, and his naked fingers on the left he strikes the
drumme, & with his mouth gaping open, makes a rude
noyse, resembling much the manner and countenance
of those kinde of distressed people which amongst vs
are called Changelings; I do the rather recite this that
it may please you to marke, what opinion the people
haue of the men of this profession, and how they dispose
of them after they are dead but first I would acquaint
:
you of their most principall instrument, which is called

Ballards made to stand a foot aboue the ground, hollow
vnder, and hath vppon the top some seuenteene wood-
den keyes standing like the Organ, vpon which hee that
playes sitting vpon the ground, iust against the middle
of the instrument, strikes with a sticke in either hand,
about a foote long, at the end whereof is made fast a
round ball, couered with some soft stuffe, to auoyd the
clattering noyse the bare stickes would make: and vpon
either arme hee hath great rings of Iron: out of which
are wrought pretty hansomly smaller Irons to stand out,
who hold vpon them smaller rings and iuggling toyes,
which as hee stirreth his armes, makes a kinde of musicall
sound agreeing to their barbarous content: the sound
that proceeds from this instrument is worth the ob-
seruing, for we can heare it a good English mile, the
making of this instrument being one of the most in-
genious things amongst them: for to euery one of these
keyes there belongs a small Iron the bignesse of a quill,
and is a foote long, the breadth of the instrument,
vpon which hangs two gourdes vnder the hollow, like
bottles, who receiues the sound, and retumes it againe
with that extraordinary loudnesse; there are not many of
these, as we can perceiue, because they are not common,
but when they doe come to any place, the resort vnto
them is to be admired; for both day and night, more
especially all the night the people continue dauncing,

vntill he that playes be quite tyred out; the most de-
sirous of dancing are the women, who dance without
men, and but one alone, with crooked knees and bended
bodies they foot it nimbly, while the standers by seeme
to grace the dancer, by clapping their hands together
after the manner of keeping time; and when the men
dance they doe it with their swords naked in their hands,
with which they vse some action, and both men and
women when they haue ended their first dance, do giue
somewhat vnto the player: whereby they are held and
esteemed amongst them to be rich; and their wiues
haue more Cristall blew stones and beades about them,
then the Kings wiues: but if there be any licentious
libertie, it is vnto these women, whose outward carriage
is such wee may well conceit it: and this one especiall
note, howsoeuer the people affect musicke, yet so basely
doe they esteeme of the player, that when any of them
die, they doe not vouchsafe them buriall, as other
people haue, but set his dead corps vpright in a hollow
tree, where hee is left to consume: when they haue beene
demanded a reason for so doing, they will answer, they
are a people, who haue alwayes a familiar conuersation
with their diuell Ho-re: and therefore they doe so dispose
of them: which opinion of theirs caused vs to neglect
and especially in their hearing to play vpon any Lute or
Instrument which some of vs for our priuate exercise
did carry with vs, in regard if they had hapned to see
vs, they would in a manner of scorne say, hee that
played was a luddy: The greatest resort of people,
with the most aboundance of these luddies, is at their
times of Circumcision."^
The best modern description of the griots, summar-
izing a number of earlier authors, is that by Charon:
> R. Jobson, The Golden Trade: or, A Discouery of the Riuer Gambra, and
the Golden Trade of the Aethiopians, London 1623, p. 105 ff.
"The griots form the most degraded portion of the
people, an abject, useless caste. Theyare parasites
who produce nothing whatsoever, because they are not
allowed to do any manual labor. They gain all their
livelihood by exploiting the Blacks chief fault, vanity,
'
and impose themselves on the great and small alike.

The griot proceeds in the most ingenious manner. He
begins by showering praises upon him from whom he
wishes to obtain a present, glorifying his and his an-
cestors generosity, bravery, nobility.
' If he does not
succeed, he will make a disadvantageous comparison be-
tween his liberality and that of one of his ancestors or
of a well-known citizen. If this means is not success-
ful, he will ultimately attempt blackmail, conveying
to him in hidden words that he knows an anecdote,
the publication of which would cause him displeasure
and trouble. At last the victim gives in and the griot
goes away, singing the greatness of his benefactor.
But flattery is not the only means of a griot 's existence,
for he has other strings to his bow.
"He is indispensable at feasts and ceremonies, be-
cause, besides the buffooneries he produces and the
songs he knows, he can play music on instruments which
he manufactures, a kind of xylophone, guitars, flutes
and drums, with which he accompanies the dances
which are highly appreciated in the black continent.
He is found not only at all kinds of festivities, birth,
baptism, circumcision, marriage, but he is equally in-
dispensable at funerals where he tries to console the
family by exalting the virtues of the deceased.
"He is also the chronicler who knows thoroughly all
the deeds and happenings of the past and who sings of
the exploits of heroes. He is also a business man to
whom all kinds of business is entrusted and who exe-
cutes it with dexterity. He both brings about marri-
ages and executes the orders of lovers. Finally, he is
tlie magician, the charlatan who cures more or less

empirically and sells drugs, be they efficacious or not.
In war the griot encouraged the warriors and excited
their bravery. He never took part in battle, for he
bore no arms. If he was captured, he was not only
sure of having his life spared, but also of not being
reduced to slavery. He then attached himself to his
victors and, without any scruple, sang their bravery
and glory. He was also employed as a spy and mes-
senger.
"Naturally the griot is rarely met with Uving by
himself, for he nearly always depends on a house from
which he draws his sustenance and for whose profit he
exercises his manifold industries. The wealthier it
is the more griots it has, for it is a sign of opulence to be
able to feed useless mouths for its mere pleasure. The
oldest griot always plays a considerable part in the
household, for he is the recorder of the family, the
trusty of the master, who makes of him a factotum and
disburdens to him some of his privileges, by transferring
to him some of his authority and power. With the
chief the griot becomes an intimate counselor, a minister;
he is a more important personage than the greatest
dignitary and, although always feared, enjoys every-
body's respect and consideration.
"As to the griot's wife, whom he can choose only in
his caste, she generally has very loose manners, to
the great profit of her husband. She lives on his in-
sinuations, calumnies, and menaces, as well as on the
sale of her good offices. She is often the confidante of
a free woman, but more especially she is a hair-dresser
and dancer.
" One can see that the griots form the dregs of society.
Indeed, the Blacks consider them to be abject and de-
graded beings, all the same having fear of them and,
consequently, certain respect for them. In fact, the
right which they possess to overwhelm with injuries
those of whom they have cause to complain is the
cause of their being well treated during their life-time
and being shown certain consideration, but revenge
is taken upon them after their death by greater affronts.
The griot's body being regarded impure, he cannot be
allowed to rest in the earth, which he would defile; and
"^
so he is allowed to rot away in the hollow of a tree.
"Each group on the shores of the Senegal has its
griots, the Moors as well as the Peuls, the Mandingos
as well as the Tukolors; but, as one proceeds southward,
this institution is seen to change, the griot by degrees
becoming fetishist, he who is in contact with the spirits
and who recognizes and furnishes the sorcerers. His
action on his compatriots has not by any means lost
by this change on the contrary, the fetishist has greater
;
profit and kills off more easily those whom he dislikes

among the idolatrous tribes than the griot in the Mos-
lem conglomerates of Senegambia.
"The griot exists not only in the countries that the
Senegal washes, but, to tell the truth, also in all Central
Africa. He is found in the Gulf of Benin as at Futa-
Dyalon, in the Sudan as at Bornu, in Wadai and
Darfur, on the shores of the Atlantic as on the shores of
Lake Chad and the shores of the Red Sea, at Zanzibar,
etc., etc. What is more, some of them travel with the
caravans over truly enormous distances, retailing on
their marches legends, musical airs and songs which are
the admiration of the natives whom they meet. "^
So far as I know griot or, rather, guiriot was first
used by Saint Lo.^ Its etymology will be given in the
^ G. Ch6ron, La sociiU noire de I'Afrique Occidentale Frangaise, Paris
1908, p. 31 ff.
^ L. B&enger-Feraud, Les peuplades de la SSnSgambie, Paris 1879, p. 375 f.
' "Les Guiriots qui les font dancer sont en ignominie parmy le commun, &
quand les Francois veulent fascher quelque Negre, il I'appellent Guiriot.
Or ces Guiriots sont extr6mement importuns, car n'ayans pas accoustumS
de voir quelqu'vn, ils I'accostent, & en luy chantant toutes les louanges dont
next chapter. Here we are interested only in the

name these griots assumed among the Mandingos. We
have already seen that Ibn-Batutah called them âli,
that is, jahlee in English pronunciation. Jobson
understood this as juddy. In Malinke it is dyeli or,
more nearly, djeli, Bambara dyeli. This is Berber
dejjal "pygmy, human caricature, Antichrist," tjal
"pygmy, ill-built, used as an expression of contempt for
a low fellow. " This leads us at once to Arabic <>r-5
daâla "he lied," <J^-> "a

da^gal great deceiver,
liar,
Antichrist." The dagâl is a very common personage
in Arabic eschatology, being a kind of monster or false
Messiah, whom the true Messiah overcomes. In
some accounts he is supposed to live in an island of the
Indian Ocean, where the Arabic sailors can hear music
and dancing. It is clear that this Arabic appellation
of the griot, and the latter, as we shall later see, is also
an Arabic word, disposes of the Negro origin of the
bard and buffoon. It is, in connotation, an exact
parallel to Arabic v-?^ haluh "deceitful," ^ ^ halla-
bah "enchanter," which led to the Helebi of Egypt,
that is, we have here again a Gypsy. Indeed, among
the Peuls this griot is called haulube or kolibante, which
is derived from the same Arabic halub, etc.
Of course, at the present time and for centuries back,
the griots have ceased to show any Gypsy characteris-
tics in language or blood, but there can be no doubt that
the caste draws its origin from a Gypsy ancestry, which,
under Arabic rule, left the Egyptian home and migrated
to the Niger. We have such persistent references
ilsse peuuent aduiser, ne le quittent iamais qu'ils n'ayent receu quelque
chose, & voyant que I'on ne leur veut rien donner ils crient si long temps
aupres des personnes que a la fin on est contraint de leur donner. Quand ils
sont morts Ton les estime indignes de sepulture, car on les met debout dans
quelque arbre creux, i'en ay veu le corps d'vn en ceste sorte de sepulchre,"
A. de S. Lo, Relation dv voyage dv Cap-Verd, Paris 1637, p. 87 f.
among the Peuls and others to a migration from Egypt
or the east that there must be a measure of historical
truth connected with them. The Niger River is in
Mandingo called Dyeli-ba, literally "the river of the
griots," which indicates that a migration of Gypsies
from Egypt to the Niger basin took place some time in
the past. The caste of the griots is distinctly one of
"Touch-me-nots," and has remained immutable through
the centuries. Individuals within the caste have not
been able to free themselves from the curse attached to
them in the Cave of Treasures, though Negroes from
without must have found it profitable to join the privi-
leged, though scorned, parasites of bards. Once in-
cluded in the caste, their exit from it was impossible,
and in the course of time its ethnological constituency
was completely changed, whereas the caste as such
remained immutable.
We are now in a position to trace the Gypsy migration
from Egypt to the Niger and to show the enormous
consequences it has had upon the social and religious
conditions of the Western Sudan and upon the Negroes
throughout Africa.
According to one native legend, the Peuls of the
Western Sudan, whose red skin has long ago roused the
suspicion that they originally belonged to the white
race and, according to Delafosse,^ were Judaeo-Syrians,
were descended from Jacob, the son of Israel, the son
of Isaac, the son of Abraham, and from Suleiman, who
came from Syria and had settled in Egypt about the
time of Joseph. After the death of Joseph, the
Egyptians tried to subjugate the Sons of Israel, and
burdened them with taxes. It was then that the
Judaeo-Syrians escaped from Egypt. Some of them
regained Canaan and Syria, under the guidance of
Moses. The others crossed the Nile under the guid-
» Op. cit., vol. I, p. 189 fl.
ance of Suleiman and marched westward. Pharaoh

pursued them, but was drowned upon trying to cross
the Nile. The Judaeo-Syrians, with their herds, came
to Soritu (Cyrenaica) and since that time assumed the
name of Fudh or Fut, in memory of their flight. Then
they marched toward the Tuat, while a part travelled to
Bornu, under the guidance of two chiefs, Gadya and
Gaye. Kara or Karake, a son of Gadya, and Gama, a
son and successor of Gaye, led their people from Bornu
to Massina, where they were hospitably received by the
Soninkes.^
This apocryphal story has a foundation of truth, as
far as the migration from Egypt westward is concerned,
for the old name of the Peuls, namely Fudh or Fut, is
derived from Arabic '^y faut "escape," which is also

found in Coptic pot, fot "escape. " But this is only an
Arabic or Coptic transformation of a Syriac legend, for
the Biblical Put was by the Syrians transferred to the
Zott, because Syriac /m< means "he expressed contempt,"
hence the nation so called belonged to the "Touch-me-
nots, " that is, was identified with the Gypsies. This
Syriac meaning was changed to the Arabic or Coptic,
where the word means "escape," even as the Greek
translation of this word in Nahum gives t^c ^^r^c
"flight."
There is another persistent Peul legend of their mig-
'
ration which has an historical basis According to all
: '
the traditions collected at different periods among the

Peuls in the different regions of the Sudan, the Peul
tribes, scattered from the lower Senegal and the Futa-
Dyalon in the west to the country between the Chad
and the Nile in the east, declare unanimously to have
come from the Senegalese Futa or from Malil, that is,
from the countries situated between the Atlantic and
1 Ibid., p. 214 f.
the Upper Niger. But they all at the same time pre-
tend that their western ancestors themselves were de-
scended from others who originally arrived from the
north or east, or, in general, the north-east. The
enormous majority of these traditions assigns to these
primitive ancestors, as their original home, the country
of Sam or Ham, that is, Syria, as considered in its largest
aspect. From there, according to these indigenous
traditions, they went to the country of Tor (the Sinai
Peninsula), then from Tor to Misira (Egypt), and from
Egypt to Soritu (no doubt, the Syrtes, Cyrenaica),
whence they much later reached the country of Diaka,
Diaga, Dia (western Massina), where we shall find
them later.
"Those of the Peuls who have been strongly Islam-
ized have amalgamated with their traditions recollec-
tions from Islamic history. Thus many of them pre-
tend that their first ancestors were still at Sinai, after
the death of Mohammed, when, in 639, the Khalif
Omar-ben-el-Khattab (634-644) sent from the Hejaz,
by the Red Sea, an army commanded by Amru-ben-el-
Assi, in order to convert the Jews and infidels of Sinai
and of Egypt. Amru landed at Tor a part of his troops,
under the leadership of Okba-ben- Yasser. He con-
verted to Islamism the majority of the Jews of Sinai,
while those who refused to abjure Judaism were mas-
sacred. When Amru returned from his expedition to
Arabia, in order to get the news of Okba's successes,
the king of Tor asked the Arab general to leave in his
country someone capable of finishing the religious in-
struction of the new converts, and thus Amru left Okba
at Sinai and marched toward Medina, where Khalif
Omar was residing. Okba, left in Sinai, there married
Tadiuma, daughter of the king of Tor, of whom he had
four children."^
^ Ibid., p. 2ns.
The historical substratum is this: "In 638 the

Oswaris, a non-Persian contingent of the Persian army,
concluded a treaty with the Moslem general, which was
confirmed by Khalif Omar, promising to embrace Islam
and enter the service of the conquerors on condition of
receiving the highest pay given to soldiers, of remaining
free, of being permitted to associate themselves with
any Arab tribe they chose, and of having to serve only
against non-Arabs. Their example was followed by
the Zotts and Sayabidjas, who had established them-
selves in the ports of the Persian Gulf and in the Iraq,
and they all associated themselves with the tribe of the
Tamim."!
Whoever the Oswaris and the Sayabidjas may have
been, the Zotts were a Hindu race from among whom
came the Gypsies, but de Goeje considers even the first
as Hindus.^ The Zotts were given to the rearing of
buffaloes in swampy regions, and the Arabs had to
contend with them bitterly before they were able to dis-
lodge them from their lurking places in the Euphrates
and Tiger valleys.^ In 710 a mass of these Zotts were
transported with their buffaloes to Antioeh. No
earlier transportations are recorded, but there cannot
be the slightest doubt that in the Persian army of occu-
pation in Egypt, in the beginning of the VII. century,
there were not only Gypsies, but also Zotts, as, indeed,
we hear of their entering the armies of the Arabs under
very favorable conditions in 638.
Apparently the Zotts were not properly treated by
the Arab authorities some time after 638, and they set
out with their buffaloes and in company with the
Gypsies in search of new places to rear their buffaloes.
It is this migration that led to the establishment of the
'
M. de Goeje, Mimoire sur les migrations des Tsiganes d travers I'Asie,
Leide 1903, p. 86.
2 Ibid., p. 17 f.
>Ihid., p. 20 £F.
red-skinned Peuls in the Niger valley, whither they
brought their hump-backed buffaloes, where they are
still to be found. They adopted the language of their
Tukolor surroundings and later became settled among
the Mandingos. It is they who carried the Mohamme-
dan institutions and superstitions into the Western
Sudan, where they have flourished more particularly
among the Mandingos. They remained free men,
horos, but the castes of musicians and blacksmiths,
although free in name, were, by Islamic prejudice, re-
duced to "slaves of slaves" and "Touch-me-nots," as
which they have persisted up to the present time.
The blacksmith, although also a "Touch-me-not,"
stands higher among the Negroes than a griot. Among
the Bambaras, a blacksmith, called numu, is free from
the death penalty, just as a member of the princely
order, an exception being made only in case of adultery
with a woman from another caste. The chief of the
numus crowns the kings and renders justice in cases
between members of his caste. ^ They tell fortunes,
cure diseases, circumcise, act as go-betweens in marri-
ages and intermediaries in divorces, and embalm the
dead.^ The Mandingo name numu can be easily ex-
plained when we consider the Arabic name (^ mdlam
for "blacksmith" in Timbuktu and among the Sosos,
which means "learned, skilful."
Similarly the Mandingo numo, Malinke, Bambara
numu "blacksmith" is derived from Arabic •^" talmldz
"a student, " talammudz " to study, " through the Zenaga
edejmun, atejmud "pupil, student," anmu4, enmud
"blacksmith, potter," almud "armorer."^ We have
' A. Rafifenel, Nouveau voyage dans le pays des Nigres, Paris 1856, vol. I,
p. 384.
" G. Deherme, L'Afrique Occidentale Frangaise, Paris 1908, p. 298 L
' R. Basset, Mission au Senegal, vol. I, part I, in Publications deV&cole
des Lettres d' Alger, vol. XXXIX, pp. 240, 275.
also the Hausa "disciple, beggar."

almajiri The
Mande numu isobviously an abbreviated form for
numudz or numud. That such is the case follows from
the fact that in Bambara numu also means "forge,"
while numuliba is recorded as "blacksmith." As the
latter is a + fto derivative, expressing the agent, it follows
that a form numuli must have preceded it, that is, one
derived from the Zenaga word. The Zenaga edejmun
"student" has produced a series of "griot" words in
several African languages. In Songay we have djam,
zam, tarn "blacksmith, worker, artisan," tamu "slave."
These two terms are in reality in opposition to each
other because the artisans were not slaves but "slaves
of slaves, " yet free. The same confusion will be ob-
served over a large territory. We get, from the Songay,
Malinke, Bambara dyo, Mandingo jongro "slave," yet
Bambara dyamuru "corv6e, forced labor," and yet
dyamuru bugu "a free village." In Wolof we have
dyamburu "free" and dyame "slave." In Asante we
have dzwuma "business, occupation, employment, duty,
ofB.ce, trade, profession," odzwumfo "artist, artisan,
workman, smith, saddler," otomfo "smith, black-
etc.,
smith;" in Akra, tsumo, Uu "labor, work, occupation,"
tiulo "servant, slave. " Thus we come to the extraor-
dinary contradiction that "the wisest of men" is at the
same time "the most contemptible" of all, which, as we
have seen, is a direct result of the curse which rests upon
the Egyptian diviner who manufactured the golden calf.
Although among the Berbers the blacksmiths are also
looked down upon with contempt, and, as Beni Niydt,
form a kind of special group outside of society,^ their
status is not so well defined as in Egypt, Somaliland,
Arabia, or the Western Sudan. The same is true of
the Beni 'Ades, who practise the art of tattooing,
horsetrading, circumcision, while their wives tell for-
1 E. Doutt^, Magie et religion dans I'Afrique du Nord, Alger 1909, p. 42 ff.
114 AFRICA AND THE DISCOVERY OP AMERICA
tunes, whereas among the Beni 'Amer the women
tattoo, while the men are horse doctors.^ Doutte
himself proposed the question whether these tribes, al-
though not considered accursed, were not identical with
the Gypsies, and to these he added the mysterious
Zkdra, an anti-Islamic tribe, who consider themselves to
be fathers of 'Amer ben Sliman, a disciple of Sidi Ahmed
ben Yusef, whom the Beni 'Ades consider to be their
saint. Doutte also propounds the question whether
Zkara is not a corruption of Zingari or some such
form.^
Unfortunately the material is too scanty to admit of
any verification, and this uncertainty is further ac-
centuated by the kar, kir forms in the African languages,
which refer to the blacksmith and his art, but are of
various origins. Thus Hausa makeri, maikira "black-
smith" is a ma- derivative of Hausa kir a "to forge,
metal," Berber kir "bellows," from Arabic kir "bel-
lows. " There are "among the Gwari, near Abuja and
in other parts, communities of Koro who live apart and
are at once feared and despised by the agricultural
Gwari. These Koro are light-colored and are black-
smiths."* The Peul haulube "enchanter," that is,' the
Arabic ^
hallabah "enchanter," seems to have sur-
"H'
vived in Soso khabi, Malinke hrahi "blacksmith."

It thus appears that among the Berbers the Gypsies
never acquired the significance in the social order that
they obtained in the Sudan, no doubt because their
migration from Egypt, as indicated in the Peul legends,
did not take place from the north, but from the north-
east over Barka, approximately over the path taken
nowadays by the Hausas who traffic between Tripoli
1 Ibid., p. 43 f.
» Ihid.,
p. 48.
' H. R. Palmer, Notes on the Kororofawa and Jukon, in Journal of the
African Society, vol. XI, p. 403.
and the Niger valley. But in the Western Sudan the

whole historical development of Negro civilization,
especially among the Mandingos, is intimately related
to the legendary and religious separation of the divining
and metal-working Gypsies as a race of "Touch-me-
nots."
CHAPTER IV.
African Fetishism and Totemism.
Probably the oldest description of African fetishism

is found in de Marees' A Description of Guinea,^ which
appeared in 1602. It runs in Purchas' translation^
as follows : Although they are altogether wild, rough,
'
'
and uncivill, having neither Scripture" uor Brookes,, nor

any notable Lawes that mi^Tbe set downe, or declared
to shew the manner pf_their,£olicie^ and~lmng,~^
when they have past the six dales of the weeke in labour
and paines taking, to get their livings, the seventh
day they leave working, and reckon that to bee
their day of ease, and abstinence from worke, or their
Sunday, which they call Dio Fetissos, which in our
speech should signifle Sunday, but they observe it not
upon our Sunday, nor-upon the Jewes Sabbath Day,
but hold it upon Tuesday,,)the second working day in
the weeke; what law-or' opinion they have to moove
them thereunto, I know not, but they hold Tuesday
for their Sunday, and that day the Fishermen goe not
to the Sea for fish: The women and Countrie people
that day bring no Wine to the Market, but all the Wine
which that day they draw out of the trees, they deliver
it unto the King, which in the evening hee giveth unto
his Gentlemen, and they drinke it among them. That
day they doe no kind of worke, nor trafflque with other
but such as dwell on the Sea-side, refraine not for all
that to goe aboord the shippes, and to buy wares of the
' Beschryvinghe ende historische verhael van het Gout Koninckrijck van
Gunea, 's-Gravenhage 1912.
* Hakluytus Posthumus or Purchas His Pilgrimes, Glasgow 1905, vol. VI.
AFRICAN FETISHISM AND TOTEMISM 117
Netherlanders. In their Markets they have a square

place foure foot every way, supported with foure
Pillars, and about two cubits high from the ground,
flat on the top and covered close with Reedes, and
hanged round about Wispes or Fetissos of Straw,
whereon they lay Millia with Palme-oile or water, and
give their god that to eate and drinke to sustaine him
withall, that he should not die for hunger or thirst,
thinking that he eateth and drinketh it and lives by it,
but the Birds of the Aire eate the graine, and drinkes the
water, and when it isêatenjth^y^ômt jthe Altar with
Oile,. and seTmore mea te and drinke uponJiTthmE^
thereby to doe their god gTeM--Samflc6^nd sOTyicg..__
"
Th«y have also a Priest, who in their speech they call
a,\ hee upon their Sabbath day sits upon a
Fetissero,\
stdoie, in the middle of the Market before the Altar or
place whereupon they sacrifice unto their Fetisso, then
all the men, women and children come and sit round
about him, and there he speaketh unto them, & they sit
stil to heare him: but what it is, or what it meaneth
that cannot I learne, nor perceive, neither can you get
it from them, for I have oftentimes asked them about
it, but they will not tell, but are ashamed to declare it.
But I have seene this Fetissero, have a pot with a

certaine drink, (wherein there was a Snake) standing
by him, and a Wispe, and some women with their little
children went to him, which children hee streaked with
colour, or with some of that drinke, and so they went
away, which I ghesse to be a kind of Salve against their
Fetisso, for they esteeme their Fetissos to be both good
and evill. And when their Fetissero hath made an
Oration unto them, then he stands up and smeareth
the Altar with his Wispe, and drinke out of his pot,
and then the people using certaine words and making a
great noise among them, clapping their hands together.
cry I.ou, I. ou, and therewith their preaching is done,
and so everie one goes home to his house.
"They hang many straw Wispes upon their heads, and
thinke thereby to bee free and safe as long as they
weare them, and that their Fetissos can doe them no
harme. In the morning betimes when they have
washt their bodies cleane, they stroake their faces with
white stripes, made of earth like ehalke, which they do
in honour of their Fetisso, and use it in stead of praiers
in a morning; when they eat any thing they present
their Fetisso (the straw Wispes which they weare about
their legges) the first bit, and also the first draught that
they drinke, giving him to drinke, which if they doe not,
they thinke they shall have no good lucke that day,
for they perswade themselves that their Fetisso would
not otherwise suffer them to be quiet. When the
Fishermen take but small store of fish, then they thinke
that their Fetisso is angrie, and therefore will give them
no fish, then they make a great crie among them, and
goe to their Fetissero, and give him Gold to conjure
their Fetisso, to send them store of fish. This Con-
jurer presently goes, and makes all his Wives (two,
three or foure, or as many as he hath) put on their best
apparell and ornaments, and with them goes howling
and crying round about the Towne, striking them-
selves upon their brests, and clapping their hands flat
together, and so making a great stirre and noise, goe to
the Sea-side, and taking boughes from the trees, hang
them about their neckes those trees they esteeme to be
:
their Fetissos Dusianam, who they thinke send them fish.

Then, the Conjiirer or he that should bewitch the
Fetissos, comes with a Drumme, and plaies or sounds
before the trees, which they esteeme to be good for
that purpose, which done hee goes to his Wives upon
the strand, and when they have spoken one unto the
other a good while, he casteth Millia into the Sea for
his meate, with other colours, thinking that thereby

their god is appeased, and will let them take fish enough.
"When the King receiveth not custome enough, to
maintaine himselfe withall, then he goeth to a tree
which he esteemeth to be his Fetisso, and sacrificeth
unto it, carrying it meat and drink; then the Conjurers
come and conjures the tree, to tell them whether there
will any Merchants come or not, which to doe they
make a heape of ashes, in forme like a Sugar-loafe,
and cutting a bough from the tree stieke that in it, then
they take a Bason of water and drinke out of it, and
therewith sprinkle the bough of the tree, which done
they speake each to other, and then againe they sprinkle
more upon it, after that they take some of the ashes,
and be-dawbe their faces therewith, and in that manner
use many foolish and vaine Ceremonies, and not long
after they shall heare a voice which is the Devill,
that saieth something unto them, and therewith they goe
home againe, and bring word what their Fetisso hath
said. They hang many of those things about their
children for diseases, as is said before, as also of their
drinke of jealousie. \
"When any man dieth, they also make a Fetisso, and
desire it to bring the bodie into the other world, and not
to trouble it in the way as it goeth, then the next our
neerest kinsman killeth a Hen, and dresseth it ready to
be sodden, which done, they goe and sit in a corner of
their house: and with him take all his Fetissos, and
place them in order, as their greatest god in the middle,
and the rest of meaner sort by it, then he takes certaine
beades, some made of shels, some of Beanes and great
Pease, and others of feathers, mixed with Buttons made
of barkes of trees, and hangs them upon the Fetissos.
After that they take the bloud of the dead Hen, and
therewith sprinkle their Fetisso (for a dead man must
offer bloud unto his god) then hee f etcheth certaine
Herbs out of the fields, and hangs them about his neoke
like a chaine. In the meane while the raan is in
tinae,
this sort made readie, the Hen is sodden, and being
sodden then he brings it, and putting it in a Platter,
sets it in the middle of the Fetissos, which done, hee be-
ginneth to conjure, using many words, and casteth
water or wine of Palme upon his Fetisso, then he takes
two or three of the greene leaves, which he hath about
his necke, and roUeth them betweene his hands, making
a little bowle or bale thereof, which he takes in the two
forefingers of both his hands, and thrusts it betweene his
legges, twice or thrice one after the other, saying, to his
, FeiissoP Aucie, which is as much as if he should say,
AHr'Baile. After this he wringeth the sap out of that
ball, and lets it drop upon his Fetisso; which done he
-kiies the ball upon the ground, and takes two or three
leaves more of the Herbs he hath about his necke, and
roUes them in his hands, and having made them in a
Ball, thrusts them betweene his legges, speaking
certaine words as aforesaid, and then lets the sap drop
upon his Fetisso, and this he doth untill such time as he
hath rolled and wrung all the greene Herbs in that sort,
which he had about his neck; then he takes all the balls
or leaves together in his hand, and thereof maketh a
ball as bigge as a mans fist, wherewith he wipeth his
face, and that also is a Fetisso, which being_done tha_
dead bodie shall rest in peace, and therewith he packeth
up all his trinkets, and laieth them aside untill another
time, that some other bodie setteth him a worke. This
kind of Superstition they esteeme for a great holinesse
for their bodies, for when they goe to warre they hang
such beades about their neckes, armes and legges, think-
ing that their Fetisso will defend them thereby, and
preserve them from killing, and thinke that they need
not feare any thing. They esteeme the Pittoir also for
a god, for when they goe from one Towne to another,
and heare it call,they are exceeding joyfuU and glad,

for they say that it is a Fetisso, which speaketh unto
them, saying, that all those that then travell in the way,
shall have no hurt, nor need f eare any danger, for he will
defend them from all men that seeke to molest them,
and wheresoever they heare it crie, there they seLMiHia^
for him to eate, and pots fuU of water to drinke, and
dfire not passe~that place without giving it somitHiDi;,
whereby in some corners of streets, and in the Woods
you shall see a number of pots, and other meates as
Millia, Mais, &c. which they set there to honour the
Fetisso the Petoir, whereby it appeareth that they
make great account of Birds, and also of some fishes, as
of the*ônny, which they by no meanes will take, but
esteeme it to be their Fetisso or Sea-god. They take
many Sword-fishes, and cutting of the Swords they
drie them, which they also esteeme for a great Fetisso.
Others put their trust in some trees, and when they de-
sire to know any thing they goe to those trees, where the
Devill oftentimes appeares in forme of a blacke Dogge,
or of such like things, and many times invisibly, and
maketh answere unto such things as they aske him. (So^
that if_youjiske, them any thingtou^mgjtheir beliele7
and they give you nolihswere^^^lgii maketh^ any shew
of truth^th'en they say that their Fetisso said so, and
willed tnem to doe it, for they esteeme Jî^_f or their
godiând use maayjoqlishtoyes and vaine shewes when
they pray to him, and serve him, thinking that it doth
them good, and that they merit much thereby, yet it
helpeth them not. But they rather find themselves
deceived, and as they deale with the Devill, and put
their trust in him, so he rewardeth them, and yet they
desire not to heare of him, but feare him much.
"There are some hils in those Countries, whereon
oftentimes it thundereth and lighteneth, and thereby
manie times some Fishermen, or other Moores, are cast
away or receive some great hurt, which causeth them to
thinke that their god is angry, and would have some
meate and drinke, or wanteth some other thing, and by
that meanes they hold manie hils to be their gods, and
set meate and drinke upon them to pacifle them withall,
and they dare not passe along by them, without going
up and giving them something, fearing that if they did
it not, they would doe them some hurt, and make each
other beleeve such things, and whatsoever they beleeve,
and once conceive in their heads, it will never be ex-
tirped, but have as firme an opinion of their Fetissos as
possible may be. But when the Netherlanders saw
them use such vaine toyes, which were so foolish, and
laught and jested at them, they were ashamed, and
durst make no more Fetissos in our presence, but were
ashamed of their owne apishnesse.
"We asked them of their Belief e, and what opinion
they had of divers things as first, when they died what
;
became of their bodies and soules. They made us

answere, that the bodie is dead, but they knew not what
any resurrection at the latter day meant, as wee doe:
but when they die they know that they goe into another
World, but they know not whither, and that therein
they differ from brute beasts, but they cannot tell you
to what place they goe^ whither under the Earth or up
into Heaven, but when they die, they use to give the
dead bbdie something to carrie with_him, whereby it Ts
to be marked that tliey beleeve that there is another
life after this, and that there they have need of such
things as they have here on Earth, for when they lose
any thing, or when any of their friends die, then they
thinke that those that are dead came and fetcht it
away, and that they had need of it, but they know not
what the Soule nor the Resurrection is.
"Secondly, asking -tiiem of their god, they made
answere, that he is blacke like themselves, and that he
was not good, but did them much hurt. Whereunto

we said, thai; our goiia white as we are, that he is gao4i
that he doth us much good, that he descended downe
upon Earthy to save us, and how he was put to death
by the Jewes f or our sakes, that when wee die wee goe
to'dwell with him in Heaven, and that there we neither
need meat nor drink, whereat they wondred, and will-
ingly heard us speake of those things, and said that we
were Gods children, and that he told us all things, but
yet they murmured, saying, why doth not your God
tell and give us all things (as well as he doth to you) and
why doth he not also give us Linnen, Cloth, Iron,
Basons, and other kinds of wares; whereunto we made
answere, that our God sent us all those things, and yet
that he forgot not them, (although they knew him not)
and sent them Gold, Palme-wine, Millia, Mais,
Hennes, Oxen, Goats, Bannanas, Juiamas, and other
fruits, to sustaine them withall, but that they denied,
or else they could not conceive that such things came
from God; but to the contrarie said that God gave them
no Gold, but that th^!,Ilarth,gave it them, wherein they
digge to find it: that hee gave them no Millia nor Corne,
but that-they sowed it, and reapt it themselves, and that
the ârtl^gave it them; that the trees which they had
planfed-'fave them their fruits, and were first brought
thither by the P^)rtugals; that yong beasts came of the
old, that the fSea^ gave. Jhem_flsh, which they tooke
themselves, with 'jfiany other such like things, which
they would not acknowledge came from God, but from
the Earth and the Sea, each accorxiing to their natures,
but they acknowledge that Elaine came from ovir
Saviour Christ, and that by meanes_ of our God they^
had much Gold, for that by meanes of the Raine they
found their Gold, and their Fruits and Plants grew,
and waxed ripe by meanes of the moysture, and for that
we brought them everie thing readie made to their'
hands, therefore they thinke that wee find all such
things, and need but goe into the fields to fetch them,
as they doe their Fruits.
"And when it happened that some of our Hollanders
being in their Houses, when it beganne to Raine, Blow,
Thunder and Lighten (whereof they are in great feare)
went forth through the streets, not once shunning the
Thunder and Lightning, neither did it once hurt them,
they wondred thereat, for they were afraid that if they
should come out of their Houses at that time, that it
would not be good for them, for that many times,
(when it Thundreth and Lightneth there) it hapneth
that some of them that are travelling abroad, are
carried away by the Devill, and throwne dead upon
the ground, whereby they are as much afraid thereof,
as any man possible can be. And for that they know
that our God dwels above in Heaven, when it Thunders
and Lightens they point upwards, and call him Juan_
Goemain. And once wee had a Negro aboord ouFsEip,
whom we kept prisoner because he brought false Gold,
and gave it out for good, which Negro everie morning
tooke a Tub with water in it, and washt his face therein,
which done, he tooke his hands full of water, and cast it
over his head, speaking divers words unto himselfe, and
after that spit in the water, and used many other Apish
toyes, which wee seeing, asked him why hee did it, and
hee made answere, that hee prayed his Fetisso that it
riiight raine, that so his friends might find much Gold
to release him, that hee might goe home againe.
"They circumcise their young children, therein
following the Mahometicall Law, with divers other
opinions which they hold thereof, as thinking it evill to
spit upon the Earth, besides many other Superstitions
which they use, but af&rme, that they altogether use
those toyes, and only trust in their Fetissos, were an
untruth, for many of them that can speake Portugall
(as having dealt with them and also daily traf&que with
us) beginne to leave those foolish toyes, and to have
some understanding of Gods Word, which they doe by
reason that wee moeke and jest at their foolish Cere-
monies, and for that they say that wee are Gods Chil-
dren, therefore they beleeve much of that which we say
unto them, and begin to know God, but it is without
any ground, for they grounded in their owne Supersti-
tions, because they_ajce not otherwise instructed.)
"But the Negros which dwell among the Portugals,

know much of God, and can speake of his Commande-
ments, as I have found some among them, that could
tell ofthe birth of Christ, of the Lords Supper, of his
bitter Passion, and death of his Resurrection, and
divers other such like points, concerning our Christian
faith; specially, one whom I knew well, and that was
my good friend: for he could write and read Portugall,
and was indifferent well learned in the Scriptures.
And which is more, when he spake unto him, and argued
upon some points against the Romish faith, or against
the Religion which the Portugals had taught him (for he
had dwelt with a Monk in the Castle of Mina) he would
dispute the contrary with us, and shew that it was other-
wise set downe in such a Gospel, and in such an Epistle
of the Apostles, & that it must so be understood:
whereby we may perceive, that those among them that
have any understanding of the Christian faith, are
sharpe witted, and will soone comprehend any thing:
but it seemeth, that it hath not pleased God to call
them to the understanding of the Christian faith, and
therefore we are much bound to prayse and thanke
God, that it hath pleased him to vouchsafe us the
knowledge of his holy Word, and to understand and
know what belongeth unto our salvation."^
Ibid., p. 289 ff.

"The children being a moneth or two old, then they
hang a Net about the bodie thereof, like a little shirt,
which is made of the barke of a tree, which they hang
full of their Fetissos, as golden Crosses, strings with
Corall about their hands, feet, and neckes, and their
haire is filled full of shels, whereof they make great ac-
count, for they say, that as long as the young childe
has that Net about him, the Devill cannot take nor
beare the child away, and leaving it off, the Devill
would Carrie it away, for they say, the childe being so
little, it would not bee strong enough to resist the
Devill, but having that Net upon the bodie, it is armed,
and then the Devill hath no power over it; the Corals
which they hang about the child, which they call a
Fetisso, they esteeme much, for that hanging such a
Fetisso about the childes necke, they say, it is good
against vomiting; the second Fetisso, which they^hanrg
about his necke, they say, it is good against falling; the
third, they say, is good againsf bleeding; the fourth is
very good to procure sleepe, which they hang about the
necke thereof, in the night-time, that it may sleepe well;
the fift, is good against wild beasts, and the unwhol-
somenesse of the Aire, with divers other such like
Fetissos, each having a name a-part, to shew what
vertue it hath, and what they are good for, and they
credibly beleeve them to be good against vomiting,
falling, bleeding, (which they presently helpe) and for
.
sleeping."^
"About their neckes they weare a string of Beades, of
divers colours, which our Netherlanders bring them; but
the Gentlemen weare Rings of gold about their necks,
on their feet, they weare many strange wreathes, which
they call Fetissos, (which name they derive from their
Idolatry) for when they eate or drinke, then they power
' Ibid., p. 260.

meat and drinke upon them: and first give them to eate
and drinke."^
"After long disputation by them made, th.e{Fetissero\
(which_ is the Priest J±Lat-.coniureth their Fetissos or
gods) came thither with a certaine drinke in a pot, and
setit downe before the Captaine, the woman tooke the
pot and drunke thereof, to justifie that he had not
contented her for the losse of her honour; and if hee
would have drunke thereof before the woman drunke, to
justifie that he had paid her, and owed her nothing,
then he had beene quit from paying any thing; but
knowing himself e to be guiltie, he durst not drinke, but
was found guiltie, and was judged to pay a Fine of
three Bendaes, which is sixe ounces of gold.
"This Drinke among them is as much as an Oath,
and is called Enchionkenou; which they make of the
same greene herbs whereof they make their Fetissos;
and as they say, it hath such a force, that if a man drink-
eth it falsely, their Fetisso causeth him presently to die;
but if they drinke it innocently, then their Fetisso
"^
suffereth them
to live.
De Marees' Dio Fetissos for the African weekly
holiday is obviously intended for the Portuguese dia
feitigo "fetish day. " The fact that the Guinea Negroes
are mentipaed as observing a seven-day week at once
points to Arabip influence, whence alone this division
of time c8t[W; have reached them. Indeed, the name
f6fTruegday-4s-itt Maiidrn:g07''Wolof, Soninke, Hausa
talata, Dahome tlata-ghe, from Arabic •^'^* saldsa'u
'
' Tuesday. De Marees did not quite get the idea about
'
'
the Negro holiday, when he

identified it with the
Christian Tuesday. not likely that the Negroes,
It is
who had borrowed the Arabic week, should have
changed the day of rest from Friday to Tuesday. If
1 Ibid., p. 266 f.
' Ibid., p. 315 i.
.

the flshermen did not go out to fish, the wine mercliants
did not bring wine to market on that day, it was so be-
cause Tuesday is an unlucky day with the Arabs, hence
truly a "fetish" day with the Negroes.
De Marees loose use of the word fetisso, which he
'
inherited from the Portuguese, is responsible for the

nondescript conception of fetishism as a peculiar aspect
of African religion. De Marees understands_ Jb;^
fetisso any charm which is intende;d.ia ward^ff^djseases
of protect against wild beasts, hence anyjbrinkets worn
on the legs. But he also transfers the meaning to the
Negro gods, who are propitiated by meat and drink and
wi^ps of straw. The fetisso is both a protecting spirit
or object and a malign spirit or object, and also serves
man as a-,totem. The priest or sorcerer, who mediates
between theêople and the fetishes, is a fetissero, but
de Marees could not ascertain what views he held in
regard to the after life and God, although he makes it
clear that in his time Mohammedan and Christian ideas
had already permeated the Negroes primitive religion,
'
if there ever had been such.

The origin of the Portuguese word feitigo "fetish"
should be from the Arabic, since an overwhelming mass
of Arabic "charm" words have found their way into
the European and African languages. I have already
shown that Fee Morgain arose from the Arabic, where it
had the meaning of "amulet."^ In the African lan-
guages I have already treated the "amulet" words
that have arisen from the meaning "writing."^ To
these must be added Hausa laya, from Arabic S"^'
al-'ayah "token, sign," Hausa hatumi, katumi, from
Arabic f'^ hatim "seal," Dahome vodU "good or
bad spirit, fetish, idol," from Arabic "'^j^ 'udzah "pro-
' See my Contributions, vol. IV, p. 141 f
^ See my Africa and the Discovery of America, vol. I, p. 108 f.
"
tection, phylactery." This produced the American

voodoo.
The Latin grammarians used the Latin tGrmfacticius
for anything made up, such as an otiqmatopoetic vôrd,
and similarly the early Christian writers and 'f'liny
employed the word in the sense of "anything created"
as opposed to "eternal," or of "artificial" as opposed
to "natural." It never occurs with any other mean-
ing in Latin literature. This Latin word produced
Spanish hechizo, Portuguese feitigo, OFrench faitis
"artificial," but in the Latin- Arabic vocabulary of the
XI. century^ we find not only "factio opus fallax,
which is sporadically found in the older Latin, but also
faccio, a back-formation ivovafacticius, with the meaning
"adolando inpedio decipio, " which is identical with
"fascino adolando inpedio laudando decipio." The
latter occurs several times in Latin vocabularies, the
first is found nowhere else, and facticius "amulet,
charm" is nowhere record:ed\ NQw^we have Arabic
*-^ used JLcar^fascinating and
/a<saA.-"-ar certain-bead-^
restraining men;one of tlifi„beads.of the Arabs "^"the
de^HjjfntFwEi^h, W0Jiifiii_aTe j^^ by the Arabs to
fascinate and restrain men." This Arabic word is
already recorded in the VIII. century, and is centuries
older, since it is already found in early Egyptian texts
as petes "globule, pill." Obviously this Arabic word
entered into Portuguese at a time when Latin facticius
had already changed to feitigo, and thus the latter, and
with it liSitin facticius, acquired the meaning of "amu-
let, charm."
Outside of feitigo there is another word which gained
currency in Europe for the African "fetish," namely
grigri. We find it for the first time in 1575 in Thevet,^
• Seybold, op. cit.
^ "Ce peuple n'a point est6 si estrangement tourmentS de ces fantosmes:
lesquels vsent de pareil traitement sur les pauures Mores Idolastres de la
where has the meaning of "a tormenting spirit."
it
This no doubt, from Mandingo grigri, girigiri "to
is,
shake, tremble, thunder," which ^very^-aQon was applied

to anything relating to primitive.religio^ Braun, in
1624, still appliedT the term to aspirit which caused
terror,^ but Jobson speaks of gregory as a mere Arabic
talisman.^ This produced French griot "sorcerer."
In all those cases where we have an Arabic word for
"fetish, medicine, amulet, " it would be absurd to speak
of an aboriginal fetishism, dating back to prehistoric
times. All folk religions resemble one another, and
the gnosticism of the Arabs is not different from that of
the Greeks or Egyptians or Babylonians. In dealing
with a contemporaneous state of superstitions, it is
necessary first to reject the late borrowings, as evidenced
in specific practices, but more especially in the linguistic
Guinee, & sur tout par les boys, les effroyant auec des visions espouuentables,
& les battant souuent, qu'ils noment en leur langue Grigry, le craignant, &
abhorrant sur toute chose," A. Thevet, La cosmographie universelle, Paris
1575, vol. II, fol. 921 b.
^ "Die Konig daselbsten opfferen jhrem Abgott Crycry mehr alsz 2000.
Menschen," Samuel Brun, Schiffarten, Basel 1624, reprinted in Werken

uitgegeven door de Linschotertr-Vereeniging, 's-Gravenhage 1913, vol. VI, p. 30.
Similariy p. 38 f.
2 "The Gregories bee things of great esteeme amongst them, for the most
part they are made of leather of seuerall fashions, wounderous neatly, they
are hollow, and within them is placed, and sowed vp close, certaine writings,
or spels which they receiue from their Mary-buckes, whereof they conceiue
such a religious respect, that they do confidently beleeue no hurt can betide
them, whilst these Gregories are about them, and it seemes to encrease their
superstition; the Mary-buckes do deuide these blessings for euery seuerall
and particular part, for vppon their heads they weare them, in manner of a
crosse, aswell from the fore-head to the necke, as from one eare to another,
likewise about their neckes, and crosse both shoulders about their bodies,
round their middles, great store, as also vppon their armes, both aboue and
below the elbow, so that in a manner, they seeme as it were laden, and carri-
yng an outward burthen of religious blessings, whereof there is none so thr-
oughly laden as the Kings, although of all sorts they are furnished with some,
both men and weomen, and this more I haue taken notice of, that if any of
them be possest of any malady, or haue any swelling or sore vpon them, the
remedy they haue, is onely by placing one of these blessed Gregories, where the
grief e lies, which they conceite will helpe them: and for ought I can perceiue,
this is all the Physicke they haue amongst them, and they doe not onely
obserue this for themselues, but their horses doe vsually weare of these about
their neckes, and most of their bowes are hanged and furnished with them,"
op. cit., p. SO f.
derivations, and then similar earlier borrowings must be

carefully eliminated, before the original state of a belief
may be even hypothetioally put. In eliminating such
accretions, it must always be borne in mind that super-
stitions, amulets, fetish rites are very frequently the
residuum of religious conceptions, medical practices and
social customs of a superior civilization insidiously
working its way into a lower civilization through the
witch-doctor and quack. Even thus have Arabic
medicine and religion worked their way into the
Western Sudan. No doubt, before that, Christianity
may have filtered into the Guinea Coast from the At-
lantic border, but that influence must have been weak,
since no powerful Christian state had developed in
northern Africa. But long before that, for whole
millenniums, Egypt must have carried its religion and its
medical science into the heart of Africa. In the Sahara
and the two Sudans it will be rather difficult to trace
this earlier infiltration because of the later, more
powerful Islamic interpenetration, but among the
Bantus, away from the coast, we may, at least in the
vocabulary, still observe the effect of Egyptian culture.
Sir Harry Johnston^ has collected a large number of
words among 276 Bantu and 24 Semi-Bantu languages,
among which we shall study a few for the connotations
"medicine, magic." Only characteristic words of a
group, without the classifier, will be given here, as the
reference to a specific language or dialect would only
burden the text unnecessarily. A characteristic form
rozi for "magic" is found in the following groups:
Nyanza loyi, rogo, logo, rdji, rozi, loko; Wunyamwezi
lozi; Ruflji-Ruvuma loha; Tanganyika-Bangweulu lozi,
losi, doci; North-west Nyasa dozi, rozi, dosi; South
Nyasaland lozi; Southern Rodesia rowa, royi, rowi, loi;
' A Comparative Study of the Bantu and Semi-Bantu Languages, Oxford
1919.
^

Sengwe-Ronga Becuana loi; Zulu-Kafir loza,
loyi, roi;
lozi, loya, lowi; Zambezia losi, lozi, loti; Kongo doki;
Luba-Lunda dohi, luwi; South Kongo loji; Central
Kongo nok\ noke; Wele-Aruwimi ogo; Kwa-Kasai lo,
logo, lono; Ogowe-Gaboon longo, doka, lok'; Duala
doki; Rumpi dowe. As one reaches the Western Sudan
this group entirely disappears. One at once recognizes
in these words Coptic loz, lozi " to cure, " from Egyptian
dri utcha "to heal, " literally "to make healthy, strong,
intact." As this is a compound word, it is evident that
the Bantu words are all borrowings from the Egyptian,
hence we have here an ancient infiltration of Egyptian
medicine. Unfortunately the time of the borrowing by
the Bantus cannot be ascertained, as it may have come
through the Arabic 'Hag "doctor," which itself is de-
rived from the Demotic lek, leg, and which produced
the Germanic "medicine" words.
Forms of ganga have in the Bantu languages the
meaning "medicine," but also "magic," hence "gun-
powder." Such are Sotho ngaka, Tlapi naka, Pedi
nak'a, Zulu nyanga, Ronga nanga, Mochi hanga,
Subiya anga, Nyika ganga, Kongo nganga, etc. All
these are from Egyptian heka "magic, the power of
working magic, sorcery, spell, incantation, charm, word
of power," from which we get Coptic Mk "magician,
demon." It would, therefore, appear that in the
Bantu superstitions we have reminiscences of mil-
lenniums past, but the appearances are deceptive, since
the words testify only to the uninterrupted existence of
superstitions which may have been modified by later
events. This is made a certainty by the presence of
Arabic dawd' "medicine" among the "medicine,
'jj
charm" words in Bantu. We have Pokomo, Zangian

dawa, that is, in the neighborhood of the Arabic colonies
1 See my Contributions, vol. IV, p. 326 ff.

^
in East Africa the Arabic word lias been preserved in

its purity.
The most interesting "medicine" word in the Bantu

languages is the one which in many cases coincides with
"tree," because here we can study the overwhelming
Arabic medical influence throughout Africa. The
Arabic jr^ saâr "tree" is found in the Sahara oases,
where we get Soa sedar, Wadai êdar, Adirar sadsdrun,

plural ladsaru, Beran sadsar, plural sdar. We also
find in Soa siddri "doctor. " The word would be very
puzzling, if we did not have constantly, in Ibn-al-Baitar,
-'^=^ Saggar "botanist." It is not surprising to find at
Beran heâb "medicine," but this is not from Arabic
v^ huSh "tree," but from v^' 'dSah "plants," since

"botany" is in Arabic called "the science of plants."
Now, we have the Arabic term \J^ §a^^drlah "the
plants or substances which enter into the composition
of a medicament, " hence we get the confusion of "tree"
and "medicine" throughout Africa in those cases where
these words are derived from the Arabic j^ Sagar.
In the African languages, especially in Mandingo,
Arabic § generally turns into t, and occasionally into s,
hence Soa sedar should appear as tara or some such form
in the native languages. Sir Harry Johnston^ records
for "tree:" Nyanza ti, sali, sala (also "medicine");
Wunyamwezi ti; British East Africa ti, di, hi (taiga,
di "medicine"); Kilimanjaro ti, di, ri, hi; Zangian ti,
ci, iri, rro; IJsagara-Ugogo ti (also "medicine");
Rufiji-Ruvuma ti, tera, tela, nandi, landi {tera,
tela, tende "medicine"); North Ruvuma tera, thende
(also "medicine"); Ukinga thende (also "medicine");
1 S. W. Koelle, Polyglotta africana, London 1854.

a Op. cit.
Tanganyika-Bangweulu ti, sola (ti "medicine") North- ;
west Nyasa ti; Yao-Ngindo and Mogambique tera,

tela, tengo, tongo, tali, ri, iri {tera, tela "medicine");
Soutli Nyasaland ti, tengo, ri, muti, huti; Southern

Rodesia ti, ri; Sengwe-Ronga dongo, ri; Becuana-
TransvaaHJ' are {tl'are, re, li "medicine"); Zulu-Kafir
ti, tsi, {ti, tsi "medicine");
hlahla, hlohla West Central
samo "medicine"); Western Zambezia te, sako,
(also
tondo {sambo, tondo "medicine") North-west Zambezia
;
ti, tondo, sako {tumbu, emba "medicine"); South-west

Africa ti (also "medicine"); Angola ji, H {lango, longo,
hemba "medicine"); Kongo ti, ci, te {longo, nti "medi-
cine"); Luba-Lunda ci (also "medicine"); South
Congoland tondo, tonda,ji, ci {Ion, ono "medicine") Up- ;
per Kwango and Kwango-Kasai ti (also "medicine");

Central Congoland tamba, sungu; Manyema te; Elila-
Lowa ti; Ruwenzori ri, ele, ti; Upper Ituri i, e; Wele-
Aruwimi le; Aruwimi-Lomami sandu, te {iso, ti, te, sisa
"medicine"); North Central Congoland ite, ete, tele,
tamba {le, te, ele, no, Id, olo "medicine"); Kwa-Kasai
ti, te (also "medicine"); Central Ogowe ti, ri, eli, ere,
rere, longu {longu, eli "medicine"); Spanish Guinea
etse, eci, ele, li, le, e, ere, tu (le "medicine"); Manenguba
eV, ed', al {an, e, el "medicine"); Middle Sanaga ete;

Pangwe le, li {an "medicine"); Kadei-Sanga-Lobai le,
ti; Fernandian te, ti {wele, bele, beli "medicine");
Cameroons-Cross River ti, te, tsi, tya, txo, tij, tse, ji, Set,
ale, ri, ni, nei, td, tu {je "medicine"); Northern Cross
River Basin Si, ji, ti, tete {ji, jik "medicine"); Central
Nigerian ji, se; Nalu ti, ri; Upper Gambia ri, s. To
this must be added the words for "wood," which in
Ruflji-Ruvuma is sagala, where the relation to the Ar-
abic word for "tree" is obvious. The successive de-
teriorationsfrom the Arabic word may be approxi-

mately arranged in tbe following scheme:
To the north of the Bantus the Arabic word is equally

universal. In the Mande languages we can study the
word in Delafosse.^ Here we have for "tree" Son-
inke yite, Muin, Mau iri, Gbin, Dyula yiri, Bambara
tige, Foro tiri, Takponin yire, Gimini kagdra, to which
must be added Malinke djiri, iri, and, in Koelle, Tor-
onka yiyi, Kabungo Iro, Tene tewuri, Gbandi nguru,
Gbese uru, Toma guru, Gio giri, Soso wuri, wudi. To
understand the Mande forms, we must turn to the
Berber dialects whence, no doubt, they were derived.
In some of the dialects, as at Redames, we have esejret,
which is pretty close to the Arabic word, but the Hass-
ania-Arabic sejar becomes in Zenaga §ar, in Chauya
tjert, plural tjur, tjari, while in Kabyl we have tejra,
plural tjur. These at once explain Malinke djiri,
Tene tewuri, of which the others are progressive de-
teriorations, but in Gimini kdgara we still recognize the
original Arabic Sugar. In Peul leg gal, lekki "tree,
amulet, medicine, " we have, no doubt, a reduction of a
form tige, as in Bambara, just as in the Bantu languages
1 M. Delafosse, Voeabulaires comparatifs, Paris 1904.
"

le, li are found for te, ti. In the Mossi languages we
still observe the same forms as in Mande. We have
Mossi tera, Dselana tiv, Legba taro, tan, Kaure tero, ten,
taro,Kiamba teo, Bagbalan tla, Kasm tenu, Yula teo.
In Nupe thghe "medicine," etc. we have an original
sebe "writing" word, but it led to tsigbon, Hghon "tree,
hence Kupa tsigmo, Esitako tsugba, Goali dsimo, suva,
"
Basa tsigwo, Ebe tugma, as well as Gbari shingwa " tree,
are similarly derived.
When the series is viewed in the inverse direction, from
the Berber countries to the Bantus, it will be at once ob-
served that the fuller forms are found in and near the
Berber countries, hence the direction of the development
must have proceeded from the Mediterranean southward,
and then we observe that the medicine words are all
*
'
'
'
contaminations with Arabic 'j^ tirah "a thing from

which one augurs good or evil, " hence Berber tira " dest-
iny, evil omen, charm, amulet, writing, " from which de-
velops a verb aru " to fix one's destiny, to write, "hence
Yoruba tira " a Mohammedan charm, anything written,
a book."
A still greater number of "tree" words could be
shown to be similarly related to the Arabic, but enough
has been given to indicate that African fetishism, in so
far as it deals with the curative powers of material
objects, is above all else the residue of Arabic medical
practice, just as the grigri and sebe are the residue of
llslamism, as passed through the hands of the religious
Wpostor. Hence the consideration of African fetish-
ism independently of Arabic medicine and magic can
only lead to disastrous results. This is well illustrated
by the havoc produced in the conceptions of African
totemism as though it were of native origin. ^ We shall
'
For the status of the scientific aspect of totemism, see A. van
Gennep, L'etat actuel du probUme iotemique, Paris 1920, where (p. 341) not
less than thirty-seven learned theories of totemism are classified.
"
approach, the subject chronologically, from its Arabic

source until the present time, Quatrem^re^ has col-
lected a number of passages Jrom Arabic authors which
show that the Persian rang "icolor, paint" produced
Arabic '^ J, br arms, banner, mark of dis-
rank"'' coat
tinction." "TFe fonA; of the Abassides was a black
garment. Melik-Daher 's rank was the figure of a lion,
wherefore he had a stone lion placed on a bridge called
"The Lion Bridge."^ The rank of another prince
"was composed of a white circle, cut by a green band
on which there was a red dagger. This rank was very
fine, and the women, even courtesans, were pleased to
engrave it on their daggers." Selar's rank was black
and white. In their processions the participants were
distinguished by their ranks, and the shop signs bore
their distinctive ranks and niSdns.
NiSan, in Persian, hence in Arabic, means "sign,
stamp, ensign, flag, standard, colors, armorial bearings,
and niSdnddr is "a standard-bearer, ensign." Both
rank and nisdnddr early entered into the European
languages. In the XII. century OFrench renc means
"the line of procession in a tournament," hence "rank
(and file), distinction."^ NiSdnddr occurs as OFrench
estandart, at first almost exclusively as the Saracen
standard, with a golden apple on the pole or, more
frequently, as the carriage which bore this ensign.
In the Moslem countries the niSdns representing
animals, fiowers, or any object were of common use in
the armies,* and among the cattle-raising Bedouins the
custom of marking the animals with a waSm was uni-
versal. "These wasms of the Arab tribes may often
be seen on men and women, either tattooed or in deep
' Op. vol. II, part I, p. 14 f.
cit.,
part II, p. 153.
2 Ibid., vol. I,
3 The correct etymology of OFrench renc was already observed by Yacoub
Artin Pacha, Contribution d I'etude du blason en Orient, Londres 1902, p. 12.
*Ibid.,v- 83 ff.
cuts which leave on the sMn marks of cicatrization;
This habit is particularly in use in the Egyptian Sudan,
because the tattooing does not show on the skin of the
Sudanese Negroes, while in the north of Egypt, either
in the valley or the desert, tattooing with indigo is of
frequent occurrence. These tattooings are. -adorn-
ments imposed by custom, intfl which is introduced the
waSm oi the tribe to which the individual belongs."^
^he- author proceeds to quote from C. W. Wilson, On
the Tribes of the Nile Valley, North of Khartum,^ to the
effect that the Semitic tribes of the Shagiah and Ja'ali
"have adopted the African custom of gashing the cheeks
of their children; the Shagiah gashes are vertical, the
Ja'ah horizontal, and the latter say they adopted the
custom from the former."
Tattooing and cicatrization are universal practices
and are recorded from antiquity, but we know them
chiefly as a manner of adornment or as a rehgious prac-
tice. In Africa the custom is distinctly one of clan or
tribe distinction and, as such, cannot be separated from
the Arabic loaSm. Delafosse^ says: "Tattooing of
-the face by. scarification is widely distributed over the
High-Senegal-iSfiger, but is not met with everywhere.
Certain peoples completely reject these mutilations:
the Moors, Tuaregs, Peuls, Tukolor, Bozo, Mahnke,
Fulanke, Birifo, Lobi, Puguli. Among others the
scarifications exist only in a part of the population or in
certain families. Thus, though the Songay are in
general not tattooed, some have very long vertical
cicatrices on each temple, or upon the brow a long
vertical incision, surrounded by a dotted line, from
right to left; the Soninke are not tattooed, except the
Dyawara, who have three short incisions between the
1 Ibid., p. 191 f.
2 In The Journal of the Anthropological Institute of Great Britain and Ire-
land, vol. XVII, p. 18.
»
Haut-SSnegal-Niger, vol. I, p. 332 f.
.
two eyebrows; the Dyula of pure Mande origin are not

tattooed, but those who are allied to the Senufu have
generally on each cheek three large fan-like cicatrices
issuing from the corners of the mouth; with the Kagoro
the majority of individuals have no scarification what-
soever, but some of them have adopted the Bambara
tattooing or have on each cheek a double fan of varied
design; the Dyan and the Gan have adopted both the
Bobo and Senufu tattoos. The natives belonging to
the other peoples of the High-Senegal- Niger have nearly
all ethnic scarifications, of which the details follow, but
I must observe that it is not at all rare, even among the
latter, to meet with individuals who are not tattooed,
and, moreover, the mark of one people or tribe has fre-
quently been adopted by the members of another people
or tribe or has been imposed upon slaves of foreign
origin, so that the tattoo marks of an individual are not
an absolutely certain indication of the ethnic group to
which he belongs."
As the waSm was a mere sign of recognition and did
not bear withit any honor or distinction, the practice of
cicatrization has in Africa been neither universal nor
imperative within the tribe, hence it is quite difficult to
trace any progressive use of this type of recognition in
Africa, whereas the Arabic nisdn has left important
results wherever it was introduced. In the Berber
country nican, from this Arabic ;]ôrd, means "target,
distinction, decoration. " In t^ Mande languages the
word hasthe specific meaning of " aMfflarorptherôbj^t
serving as distinctivejmarkjoi -the. family.',' Thus we
Have Bambara n^tene, tene, for Arabic nUan, since here
§ changes to t, Malinke tana, tene, tne.
Brun^ summarizes the concept of the totem in the
Western Sudan as follows:
^ J. Brun, Le totemisme chez qtielques peuples du Soudan Occidental, in
Anthropos, vol. V, p. 863 f

• ^'1. Totemism does not appear as a precise stage of
religious evolution exclusive of all other beliefs. It is
^simply an element of beliefs.
-^' 2. It appears, indeed, as a universal fact in French
West Africa.
"3. Except for some details, the beliefs and practices
ha,v6~ among these people a remarkable character of
jihity.
"4, Although certain social institutions are placed
"under the protection of totemic beliefs, the whole of
such institutions is not based on totemism.
"5. It is a definite fact that at the present time
there is no common totem for a whole people.
"6. It is possible, but not demonstrated, that the
names of several peoples are reminiscences of totemism.
" 7. Totemism is essentially familiar, in the sense of
the gens of the Latins.
"8. The totem is transmitted from the father to the
children of both sexes.
- "9. Totemism does not present any exogamic
character.
"10.The native legends do not attach the origin of
totemism to an animal descent, but to a benefit rend-
ered by the animal to the family ancestor.
"11. Finally, it seems that the sense of 'parent-
age, when referring to the totem, should be explained
'
^ the belief in metempsychosis."

Brun says that he has never found any representation
of the animal totem, either in painting or sculpture, or
; as a symbol.^ But Ibn-Batutah tells of a golden bird
'
which surmounted the umbrella of the King of Malli,^
and which apparently was of the nature of a totem.
We shall later see that the Mexicans have preserved
for us representations of Mandingo blazons on their
1 Ibid., p. 860.
^ Defremery and Sanguinetti, op. eit., p. 406.
sMelds, wMcli are of the nature of totems, in the sense

of representing personal or clan dignity. According to
Artin,^ in the XI. century " the Mamelucks of S3rria and
of Egypt, following in this the example which had been
set them by the Oriental princes, also adopted these
coats of arms, which they painted on their shields,
their streamers, their standards, even on the garments of
their slaves,and sometimes on their coins, and on monu-
ments which they built. " In the XIV. centtiry, when
Ibn-Batutah visited the Mandingos, these were far
more fully under Arabic influence than they are at
present, and they imitated the custom of emblazoning
their weapons and garments, which only in the XII ^
century became a universal custom in Egypt. This is

evident, as we shall see, from the Mexican records, which,
therefore, cannot be of a date earlier than the XII.
century of our era. At a later time the Mandingos,
with the loss of their empire, discontinued the method
of representing their coats of arms, preserving only a
vague reminiscence of animal antecedents, which, how-
ever, is neither universal nor definite, and which has
wrongfully led to a conception of prehistoric totemism
in the Western Sudan.
Ôp.cit., p. llff.
CHAPTER V.
The Boei.
Henry's work on the Bambaras^ gives a complete ac-

count of the Mandingo bori, hence it will be summarized
here to the extent to which it touches upon the religion
of the Sudanese Negroes, and, ultimately, of the
American Indians.
The benevolent spirits of the Bambaras are called
nyena, better known under the Arabic name of jine
(drr). These from the Islamic spirits in that
dififer
they are conceived as intermediaries between angels and

men, or between a vaguely conceived divinity and men.
The jine sometimes assume a human form, when they
are represented as white beings. The nya or boli, like
the jine, may be male or female, but they are in constant
opposition to God and the jine, and torment men with
their mischievous acts. When they appear as human
beings, they are represented as black and exceedingly
ugly. While animals of a white color are sacrificed to
the jine, the holi receive only black or red victims.
There is, however, no strict line of delimitation between
the two kinds of spirits, and women and children apply
the term nyena to both. Men, too, are not always
making a distinction between them. Some sorcerers
invoke only the jine, others call upon both, but more
frequently upon the boli, while others again travel from
village to village dressed in the skin of a goat or wild
cat, from whose paws hang little bells which announce
their coming, and talk in the name of the boli, predicting
' J. Henry, Les Bambara, Miinster 1910.
THE BORI 143
the future or reading the past. Such, a sorcerer is

known as bolitigi, literally "master of the holi."
The Bambaras have retrograded in religion, for the
jine refer to a period when the Mandingos drew their
religious concepts frona the Koran. Since then they
have almost entirely lost the idea of Godor the positive
injunctions of the Islamic faith. They have preserved
the idea of sacrifice, the saraka, the Arabic ^j^ §araqah
"alms, " but it has deteriorated into a mere contribution
to the sorcerer, to propitiate his good will. In every
sacrifice there is a priest who is charged with the duty,
but in the saraka there is, properly speaking, no priest.
In the indirect saraka, by the request of a fetish, the
person offering it addresses himself, out of respect, to
the chief Of the village or the family. In the direct
saraka the sorcerer indicates the particular persons,
provided they be circumcised or excised, who are to offer
it. In a public saraka a woman is never chosen, but
only a revered old man. After a public sacrifice the
sacrificer generally receives a chosen morsel and the skin
of the animal, and the assistants roast the meat and
devour it on the spot, whereas in the saraka, except by
special order of the sorcerer, the meat, the flour, the
kola nuts, the cowries are abandoned at the place of the
saraka, and only the poor may take it all and eat it.
Generally the children of both sexes are called to par-
take of the food. What is left is taken up and is de-
posited at the cross-roads. In a saraka offered by the
whole village it is nearly always a goat that is killed, and
the hide folded into four or six parts is fastened with
strings of the Bauhinia reticulata and is suspended on the
end of a pole at the entrance to the village.
The saraka is never given to the fetishes boli, but only
to the fetishes jine and to God, and it is either bloody or
bloodless and is given directly or indirectly. The di-
rect saraka is given by either men or women, provided.
according to sex, they are circumcised or excised. In
an indirect saraka, whether at the instigation of a jine
or a holi (one given to God is never given through a
boli), one must have recourse to the owner or sacrificer
of such a jine or boli. The animal of the saraka is killed
to the fetish taken as a mediator, and this fetish takes
the responsibility of averting evil from him who has
presented to him this tribute or impost. In the saraka
two things are given, the material and the immaterial,
and when it is indirect, there is necessarily bloodshed-
ding in favor of the mediator, in order that he may
seize the life of the beast offered as tribute, and offer it
in turn to the proper person. The saraka is the unique
religious act of the Bambara towards his creator and
sovereign master. Although it is done at the fork or
crossing of paths, it may be given anywhere, and the
god of the Bambaras remains without priest, without
altar, without sacrifice, properly speaking.
Certain sorcerers of renown prescribe a saraka to a

whole village, and at times to a whole region. The
entire wisdom of such a sorcerer, who can tell every-
thing by rummaging in the entrails of chickens and
crushing them between his fingers, may be summarized
in these words: "The jine are dissatisfied, offer a
saraka to them, offer it quick, for, if you do not do so,
all the children under three years will die. . .goats,
.
cows, horses, sheep will perish. . .the springs will

.
dry up. . ..the karite butter will fail," etc. Some-

times, after much persuasion, everybody offers a
saraka, even the Moslems taking part in it. Some-
times a saraka is given in order to hurt a neighbor, in
which case a magical object is placed nearby. The
Bambara is persuaded that if his enemy touches such a
magical object, his vengeance will be so much more
effective. Certain merely superstitious practices,
which do not have a proper religious character, are also
THE BORI 145
called saraka. Rags and clusters of fruit suspended

from trees are not so mucli an homage rendered by the
Negro to his jine, as a tribute offered to his own personal,
inarticulate fear.
The Bambaras use a number of Arabic terms for
religious concepts, but this they do only for diplomatic
reasons, for the adaptation of these forms has nothing
in common with acceptation. The Bambaras have
always struggled to maintain their religious independ-
ence and to save their faith and their fetishes. These
religious struggles in themselves allow us to consider as
an indisputable fact the statement that before the
arrival of the Moslems they were animists and sacrificed
to the genii.
In this r6sum6 of a part of Henry's book the data are
highly interesting, while the author's explanations are
generally inadequate or wrong. The absence of an
altar in the saraka does not point to a native origin of
the saraka, since the altar is almost unknown to the
Arabs.^ The saraka, from Arabic ^-^ sadaqah (not
'^^ which in Henry is a misprint), is unquestion-
,
ably of the nature of an expiatory sacrifice, for in

the Islamic orthodoxy the victim in such a case is
generally distributed to the poor,^ and in North
Africa the ââqah is the tithe on the capital for the
current year, which is directly distributed to the poor in
the form of alms.^ Hence the Bambaras, by turning
the sacrifice into a saraka, only imitate an Islamic cus-
tom. The killing of a goat at the sacrifice is based on
the Koranic prescription that only a camel, an ox, or a
goat may be sacrificed. As the pre-Islamic ginns
have permeated the Islamic religion, it is only natural
for the Bambaras, with their low intellectual develop-
1 Doutt6, op. cit, p. 463.
' Ibid., p. 474.
' Ihid., p. 493.
ment, to give these sarakas to the jine rather than to
God. Thus we have so far not a particle left of a non-
Islamic origin of the Bambara sacrifice.
According to Henry, the Bambaras believe in a great
variety of nyena or jine. There is the wokolo, the
mischievous jine of the brush the unnamed jine of the
;
village who must be propitiated, in order to keep the

children from diseases; the were- wolf, suruku; the
powerful jine who can do good; the jine of the springs,
the crops; jine protecting the children, the circumcised
and excised children, the women, the twins, etc. All
such jine may be found among the Arabs, even as they
form the folk-lore of Europe, hence there is nothing
specifically Negro in them.
Far more interesting is the protecting jine of the vil-
lage, the dasiri, who lives on a rock or, more generally,
in a baobab or cailcedra tree. He is the property of a
family with which he shares the name, and his sacrificer
has the name of dugu-tigi "village chief," whereas the
civic head of the village is called so-tigi "house chief."
The great sacrifice to the dasiri, whose mount is gener-
ally a serpent, rat, lizard, ass, but more generally a
horrible billy-goat that has the freedom of the village,
takes place every year. For at least three days the
village revels: men, women and children dance, gorge
themselves with food, drink, and get drunk. Every-
body takes part in the sacrifice, and the women, who
may not enter the sacred grove, at least go to its edge.
Everything is abundance: honeyed flour, milk, millet
in
beer, and meats. In the name of the village they
sacrifice at least a goat, and sometimes a bull, and the
chiefs of the families, according to their means, offer
goats and chickens.
Who does not see that we have here a close reminis-
cence of the Moslem 'id-al-âgir, the "minor festival"
after the fast of the Ramadan? In the Mohammedan
THE BORI 147
law only two holidays are specifically provided for, ^ the

'id-al-âgir "the minor festival," on the first of
Shawwal after the Ramadan, hence also called 'id-al-
fitr "feast of the breaking of the fast," and the 'Id-al-
kdblr "the great festival." For the latter it is pre-
scribed for every free Moslem who may be considered as
a father of a family to sacrifice an animal, provided his
means permit. A goat is sufficient for one person, while
seven persons may combine to sacrifice a camel or ox.
It is recommended to distribute the meat of the sacri-
ficialanimal as a sadaqah among the poor and needy,
the sacrificer retaining a part for the sake of the blessing
it contains. As a matter of fact, the minor festival,
which comes at the end of a long fast, has gained greater
importance in all Moslem countries.
"On the first three days of Showwal (the tenth month,
the next after Rumadan) is celebrated the minor of the
two grand festivals which are ordained, by the religion
of the Mooslims, to be observed with general rejoicing.
It is commonly called el-'Eed es-Soogheiyir; but more
properly, el-Eed es-Sagheer. The expiration of the
fast of Rumadan is the occasion of this festival. Soon
after sunrise on the first day, the people having all
dressed in new, or in their best, clothes, the men as-
semble in the mosques, and perform the prayers of two
rek'ahs, a soonneh ordinance of the 'eed; after which,
the Khateeb delivers an exhortation. Friends, meet-
ing in the mosque, or in the street, or in each other's
houses, congratulate and embrace and kiss each other.
They generally visit each other for this purpose. Some,
even of the lower classes, dress themselves entirely in a
new suit of clothes; and almost everyone wears some-
thing new, if it be only a pair of shoes. The servant
is presented with one or more new articles of clothing
1 Th. W. Juynboll, Handbuch des islamischen Gesetzes, Leiden, Leipzig 1910,

p. 126 fl.
;

by the master, and receives a few piasters from each of
his master's friends, if they visit the house; or even
goes to those friends, to congratulate them, and receives
his present if he have served a former master, he also
:
visits him, and is in like manner rewarded for his trouble

and sometimes he brings a present of a dish of kahhk
(or sweet cakes), and obtains, in return, money of twice
the value, or more. On the days of this 'eed, most of
the people of Cairo eat feseekh (or salted fish), and
kahhks, fateerehs (or thin, folded pancakes), and shoor-
eyks (a kind of bun). Some families also prepare a dish
called moomezzezeh, consisting of stewed meat, with
onions, and a quantity of treacle, vinegar, and coarse
flour; and the master usually procures dried fruits
(noockl), such as nuts, raisins, &c., for his family.
Most of the shops in the metropolis are closed, excepting
those at which eatables and sherbet are sold; but the
streets present a gay appearance, from the crowds of
passengers in their holiday clothes.
"On one or more days of this festival, some or all of
the members of most families, but chiefly the women,
visit the tombs of their relatives. This they also do on
the occasion of the other grand festival, of which an
account will be given hereafter. The visitors, or their
servants, carry palm-branches, and sometimes sweet
basil (reehhan) to lay upon the tomb which they go to
visit. The palm-branch is broken into several pieces,
or its leaves are stripped off, and then placed on the
tomb. Numerous groups of women are seen on these
occasions, bearing palm-branches, on their way to the
cemeteries in the neighbourhood of the metropolis. They
are also provided, according to their circumstances,
with kahhks, shooreyks, fateerehs, bread, dates, or some
other kind of food, to distribute to the poor who resort
to the burial-grounds on these days. Sometimes,
tents are pitched for them: the tent surrounds the tomb
THE BORI 149
which is the object of the visit. The visitors recite the

Fat'hhah; or, if they can afford it, employ a person to
recite first the Soorat Ya-Seen, or a larger portion of the
Ckoor-an. Often, a khutmeh (or recital of the whole
of the Ckoor-an) is performed at the tomb, or in the
house, by several fickees. The men generally return
immediately after these rites have been performed and
the fragments or leaves of the palm-branch laid on the
tomb: the women usually go to the tomb early in the
morning, and do not return until the afternoon: some
of them (but these are not generally esteemed women
of correct conduct) if they have a tent, pass the night
,
in it, and remain until the end of the festival, or until

the afternoon of the following Friday: so also do the
women of a family possessed of a private, enclosed
burial-ground, with a house within it (for there are
many such enclosures, and not a few with houses for the
accommodation of the females, in the midst of the public
cemeteries of Cairo). Intrigues are said to be not un-
common with the females who spend the night in tents
among the tombs. The great cemetery of Bab en-Nusr,
in the desert tract immediately on the north of the
metropolis, presents a remarkable scene on the two
'eeds. Ill a part next the city-gate from which the
burial-ground takes its name, many swings and whirli-
gigs are erected, and several large tents; in some of
which, dancers, reciters of Aboo Zeyd, and other per-
formers, amuse a dense crowd of spectators; and
throughout the bvirial-ground are seen numerous tents
"^
for the reception of the visitors of the tombs.
Among the Bambaras we have the same indulgence in
sweetmeats, the same sacrifice of a goat by a chief of a
family, the same procession of the women to the sacred
grove, instead of the cemetery, the same revelry for
1 E. W. Lane, An Account of the Manners and Customs of the Modern

Egyptians, London 1837, vol. II, p. 240 flf.
three days. Mere accidents of resemblance are ex-
cluded. The only difference is, as usual, that the
Bambaras, whose Islamism is skin-deep, have recourse
to a specific înn, even as the uneducated Christian or
Mohammedan masses favor a particular saint for some
holiday. But the very name of dasiri given to the
protecting village înn is significant, for dasiri, in
Bambara, also means "to gag, to stuff," from Arabic
j^^ dasara "to push, propel, stab." In Berber the
related Arabic >•= dagara produces deger "to push,
shake, drive away," which leads in Hausa to dakara "a
guard, soldier." Hence we have in Bambara dasiri
"the one who drives away disease, a protector."
Henry records another series of fetishes, which he
identifies as belonging to the cult of Satan, namely the
holi or nya, who prefer to take up their residence in an
inanimate object with which they become one and the
same, so that in the sacrifices one addresses oneself to
the object as a whole, whereas in sacrificing to a jine
one addresses the inherent spirit. The boli are in-
trinsically bad, in revolt against the Supreme Being,
the Creator and Master of all things. They reside in
the fire, in hell, and every black man will tell you that
the boli is a demon (setane).
There are the minor boli, bad fetishes that have
nothing terrifying in them, in so far as they rarely cause
anybody's death. A large number of these are above
all medicine, and everybody has some of them. Such
are tails of cows, goats, dogs, without any special name,
and called fur a (medicine). To these may be added the
kana, ox-horns and hoofs, filled with an unguent of
which it is hard to give the composition, blood, grease,
and dried and powdered fire-flies, mixed with iron and
coal dust and other ingredients. These kana are
different from the kana which every major boli posses-
THE BORI 151
sesand are no longer boli, but a part of a boli. This

word kana means "that which, protects, preserves,
keeps evil away from us." These small fetishes are
not celebrated, every owner merely besprinkling them
with blood and sacrificing to them. The most power-
ful of these boli are the boli of the house, to whom the
guatigi, the chief of the family, sacrifices a goat and at
least a chicken and a kola nut for every male child.
Henry calls the "major boli" all those nya whose
cult is secret and in whose mysteries only the initiated
participate, after having solemnly sworn not to betray
them. These receive an endless number of victims,
and some of them are constantly covered with blood.
At the sacrifices five or six sacriflcers may participate.
This is the great and true Bambara cult, which gives
rise to orgies lasting from four to eight days. These
boli are very much feared. They render justice and
preside at all disputes, they have the right of life and
death over men, and through them the Bambaras wreak
their vengeance against a real or imaginary enemy.
No victims are too noble for these boli, and there was a
time when human beings were sacrificed for them.
These blood-slaked boli are made from tree-bark, roots,
goat-horns, etc. Some look like roughly made dolls,
or a hippopotamus, a cow, an elephant, etc.
The cult of these boli is complicated. We have the
boli-tigi, the owner of the fetish, and the murukala-tigi,
the sacriflcer, and the two wara da, the intermediaries
between the fetish and its possessor. Through them
the boli makes known its wishes and gives its orders,
and its mouth-piece is in this case always seized with
fits of epilepsy, followed by catalepsy. When he has
spoken, he is brought out of this state by sprinkling
some holy water upon him, when he takes his place as
though nothing had happened; although but a minute
ago his body was covered with perspiration, his limbs
were agitated witli a nervous spasm, his haggard eyes
seemed to roll from their orbits, a bloody spittle
flowed from his mouth, and his respiration was wheezy
and broken. The boli has his beadle, the darotigi, who
rings the bell for the sacrifice and sings the praises of
the fetish, wherefore he is also called nyd dyeli "boli
griot."
Henry divides all the holi into three groups. In the
firsthe places the kingly or governmental fetishes which
formerly were propitiated with human sacrifices. In
the second are the Tyiwara and the Duga. The first is
represented by a well-made mask of a goat or fawn.
The chief boli belong to the third group, among which
are the Komo and the The sacrifice to the
Nama.
latter is described as follows: In an enclosure where
there is a bee-hive which serves as a tabernacle, only
the officiating sacrifioers may enter. The crowd of
the fraternity members stay outside and approach
only to present to the sacrificer a chicken and kola nuts
of the sacrifice. While the Nama is sprinkled with
blood, two men, naked up to their waists and facing the
crowd, stand motionless on each side of the sacrificer,
holding in their hands two pieces of wood on which three
sheep-horns and two other horns are tied. It is a
great honor to hold these pieces of wood, and these
people are called dyenfa tyeu. Two sacrifices a year
are offered to these boli, one before the rainy season, at
a time when, as the Negroes say, the male and female
idols copulate, and the other about three months later,
just before the grain is ripe.
There are also talking boli, who are reputed to put to
death those who show them contempt, but are chiefly
known for causing sickness and misfortune. They are
in the hands of self-appointed sorcerers, veritable blood-
suckers who run from village to village during the dry
season, always stretching out a hand and begging.
THE BORI 153
The generally an ox-tail on which the

talking boli is
sacrifice is The owner generally has his
made.
paraphernalia in a case of fawn leather, from which one
end protrudes a little. This is his ensign, his mark of
identity, unless it be the bell suspended from the flap of
his leather case, which he rings in order to announce
his coming. The Bambaras consult these men, in
order to find out their future or have their fortunes told.
As before, Henry's data are valuable, while his
explanations lack foundation. It is clear that the
Bambara boli, that is, a Satanic înn, is no other than
Arabic Jt»- habal "a malicious ginn, Satan," which
among the Bambaras is also evidently applied to an
amulet, where fur a "medicine" and kana "to protect"
are exact translations of the various Arabic words for
"amulet." The bloody sacrifices to the boli are
reminiscences of pre-Islandc practices not entirely done
away with by Mohammed, for these are expressly
sanctioned in the Koran. ^ The Bambaras, however,
have added a number of f eatiires in their sacrifices taken
from the dervishes, whose frantic dances and madness
are in no way different from the epileptic and cataleptic
fits of the Bambara sorcerer. "Who does not know
the dervish, not the wild sectarian of Persia .... but
the dervish by derision of northern Africa, whom the
sceptics riddle with sarcasms, and the common people,
considering him sincere, adore and venerate."^ The
"^
intelligent people call him mahbul, that is, "crazy,
but the uneducated take this term merely to mean "pos-
sessed by a habal," that is, "a religious epileptic." The
identity of the term mahbul, for which we get Hausa
maiboli, maibori "obsessed, " with the one possessed by
1 J. Wellhausen, Skizzen und Vorarbeiten, Berlin 1887, vol. Ill, p. 110 ff.
18^ 0. Depont et X. Coppolani, Les eonfreries religieuses musulmanes, Alger
97, p. 96.
' Ibid., p. 97.
"

a the Bambara practice, does not admit of
hori, as in
any other conclusion than this, that we have among the
Mandingos, chiefly the result of the activity of the
Moslem marabut, the fakir of the Sudan. That bori
originally meant "obsessed, insane" is shown by the
fact that the Tuaregs understand by buri "a vertiginous
disease of the brain among the Negroes, which drives
them mad."^ We also have the fuller form in Berber
hebel "to have a disturbed mind, be insane," hebbuel
" to turn topsy-turvy, " hebbel " to get mixed up, be off,
by the side of hebbala "ingredient which enters into the
composition of a philter," that is, "something which
makes one lose one's mind. " Thus we are brought back
to the Bambara boli "amulet, medicine." In Tama-
zirt we have amehbul "crazy." Thus the relationship
of Bambara boli to the Arabic Jt»- habala is put beyond
any possible doubt.
As we shall later, in America, have to deal with the
talking bori, we shall here place an early account of him,
as given by Jobson: "And to make vp the number at
all these meetings, there is one sure card that neuer
f ailes, which is their roaring deuill, that before I spake
of, whose attendance may seeme to keepe the youth in
awe, and he is called by the name of Ho-re, whose
strange report I proceede vnto: There is at all these
meetings, some distance of from the place, heard the
noyse of a roaring voice, resembling the greatest base
of a mans voice; when we demand of them what it is,
they will answer, with a kinde of feare, it is Ho-re, and
then describe him to be a f earefuU spirit, that none may
come neere, without danger of being destroyde, carryed
away, or torne in pieces: there is at all their meetings,
vpon the first notice of his voice, a preparation for him
of all manner of victuals, they haue amongst them, euery
' H. Duveyrier, Les Touareg du Nord, Paris 1864, p. 436.
THE BORI 155
one imparting somewhat, all which is carryed towards

the voyce, and there vnder a tree set downe, and within
small time, bee it of what quantitie soeuer; it will bee
found deuoured, and not so much as a bone to bee
scene, vneaten, or left behind, and if they be not ready
forthwith to carry him such prouisions, as shall content
him, some of their vncircumcised sons are instantly
taken away; females he meddles not with, and saide to
remaine in Ho-reyes belly, some of them nine or tenne
dales, from whence they must be redeemed with some
belly prouision: and it is strange to heare, how confi-
dently they will report vnto you, that they haue beene
carryed away, and beene abiding there wherein this is
:
obserued, that looke how many dayes he hath beene

kept away, or remaining, as they say in Ho-reyes belly,
so many dayes after they returne, it must be, before
they will, or dare open their mouths, to speake a word.
For confirmation of which, this I haue scene as I walkt
:
one day into the countrey from our dwelling to Feramb-

ras house, distant some foure mile, in the way we were
to passe through a towne of the Fulbies, among the
people that lookt vpon vs, I was shewed a youth of
some eighteene yeares of age, who they said, came but
the night before out of Ho-reyes belly: I went towards
him, and vrged him to speake vnto me, but still he went
backe from mee, and kept his finger before his mouth,
and notwithstanding I made what meanes I could, by
pulling and pinching of him, and more to terrifle him,
making proffers with a false fyer to shute at him, beeing
naturally exceeding fearefuU of our gunnes, I could not
preuaile, neither make him open his mouth: notwith-
standing afterwards, the same fellow did often come,
and haue commerce amongst vs: nay our people, who
were lying, and dwelling in the countrey, had beene at
seuerall times frighted with the voyce of this Ho-rey, for
hauing staide in their fowling, or being abroade, vntill
night hath ouertaken them, in their eomming home, as
they haue saide, they haue heard the voyce of Ho-re, as
they might conceiue, some mile from them, and before
they could passe tenne steppes, hee hath seemed to be
in their very baekes, with fright whereof, maintained by
their imagination, of their report went of him, they
haue not, without a gastly dread, recoueredhome: vnto
which place of dwelling, he neuer was so bold to make
any attempt: and verily my opinion is, that it is onely
some illusion, either by the Marybuckes, or among the
elder sort, to forme and keepe in obedience those
younger sort for better approbation of what I suppose,
:
I will craue the patience, to set downe what I obserued

at the circumcision of our blacke boy: The nights were
very light, the Moone being then about the full towards
midnight, eomming from Bo lohns house to the place at
Faye, Ho-reyes voyce was wondrous busie, as it seemed
to me, not farre of. I spake vnto my consorts, we
would secretly take our armes, and steale downe, to see
what it was, one of our three was backeward and vn-
wilUng, whereby it came to passe, our Marybucke
vnderstood what we intended, who came earnestly vnto
mee, intreating, I would giue ouer that dangerous at-
tempt, saying, I could not finde him, for one cry would
be hard by me, and another instantly beyond the riuer,
which was a mile of, and there was great danger, he
would carry me into the Riuer with him: when hee
perceiued, he could not alter my resolution, he held mee
by the arme, and pointing to a blacke, not farre from
mee, held downe his head. I went to that man, being
a very lusty fellow, to speake vnto him, whose voyce
was growne so horse, by crying like Ho-re, he had no
vtterance, whereupon I returned to my Marybucke,
and saide, there is one of your Deuils; who with a smile
went his way from me. "^
•Op. cit., p. 115 ff.
THE BORI 157
Delafosse's account^ of Mandingo

fetishism on the
whole coincides with that Henry, but he gives more
of
space to the Mandingo suharha "sorcerer:" "The
subarha or fortune-tellers are initiated into the magic
of possession, who voluntarily allow themselves to be
possessed by the nyama, the life-spirit of a dead person,
in order to increase their power and to execute the
vengeance for which the nyama is thirsting. They
are very much feared, and certain religious associations,
such as the Komo and Nama, devote themselves to the
discovery and execution of such subarha. The latter,
whatever may be said of them, are not anthropophagous,
properly speaking; when they say that they have eaten
an individual, that is merely a simple way of saying that
they have thrown upon him a korte, a mysterious and
mortal poison, or one reputed to be such. Similarly
a man afflicted by a bad sore thinks that a suharha has
sucked his blood. If a child has been eaten up by a
hyena, they say that the author of this exploit is a
subarha for the time being transformed into a hyena.
These behefs are exploited by the namatigi or priests
of the Nama and by the gbassatigi or owners of amulets
against all kinds of evils, who, if they have a grudge
against some person, accuse him of being a subarha, in
order to put him to death. Diseases and death are
generally ascribed to the anger of a ginn or the spirit of
"^
a dead person, or to the sorcery of a subarha.
The Malinke dictionary gives subaha "a night-man,
were- wolf, sorcerer who runs through the night in quest
of some evil deed and is accused of eating the souls,"
while the Bambara dictionary has suba "night sorcerer,
were- wolf. " It would seem that this is a native word
derived from su "night," even as Soso kuera-mikhi
"sorcerer" seems to be compounded of kue "night" and
' Haut-SenegalrNiger, vol. Ill, p. 161 ff.
2 Ibid., p. 182 f.
"

mikhi "man." Yet we certainly have here Hausa
subahi "the dawn, " from Arabic ^^ subh "the dawn,
which was dreaded because even then the pre-Islamic
Arabs made their c*.^ sabih, their terrible predatory
raid."^ No wonderthat the sorcerer should be identi-
fied with the masked raiders of the night.^
Before passing to the investigation of the popularity
of the Mandingo boli, which we have seen to be of
Arabic origin, it must be pointed out that even tigi of
boUtigi is an Arabic word. Mandingo tigi means
"chief, master, possessor." The Portuguese called all
the African chiefs xeque, from Arabic ^ saih "elder,
chief." This Arabic word produced MaHnke, Bam-
bara tigi, Mandingo tio. In Columbus' cacique, as I
have already shown, we have a compound, in all prob-
ability, kun-tigi "village chief,"* where cique is still
closer to Arabic iaih. Thus the Mandingo bolitigi
"possessor of a bori," which in other Mande languages
assumes the meaning "sorcerer," bears at once a
double reminiscence of Arabic influence.
The religion of the bori, if this gnostic Islamism may
thus be called, is found scattered over a large territory.
It reaches down into the Bantu countries of the Congo,
where we have a detailed account of it.^ The national
fetish of the Fans is called bieri or bieti, among the Fiots
bwiti. As these terms also denominate the members
of the association who know the mystery of this fetish,
For the history of this terror in Europe read my Contributions, vol. IV,
'
p. 324 S.
* In an article, Sur les pretendus loups-garous et sorciers nocturnes au
Soudan, in L' Anthropohgie, vol. XXXI, p. 489 ff., Delafosse withdraws his
previous statement in regard to the Sudanese were-wolves and denies their
existence in native belief. But the discussion of Hausa amina at the end of
this chapter shows that Delafosse was more nearly correct before than he is
now.
» See vol. I, p. 71 f.
* H. Trilles, Le totemisme chez les Fan, Munster 1912.

.
THE BORI 159
it we have here confusions of Mandingo

follows that
boatio, Bambara bolitigi "fetishist" with Bambara boli,
Mandingo boa "fetish. " The secret society of the bieri
is also found among the tribes of the Ngunye and the
High Ogowe, totally unrelated to the Fans. The Fan
with the Ahamhu m'biri, m'biti, the Ba-
bieri is identical
Kal6 m'bwiri, m'bwiti, the Galea om'biri, the A-duma
m'bweri, the Bat6k6 bwH4, the Pahuin m'bieri.^ "The
chief fetish of the Pahuins is the bieri, whom only the
initiated may know or invoke. The bieri is merely the
skull of an ancestor, religiously preserved by the new
head of the family, after the decomposition has done
its work in the tomb. It is rehgiously kept in a box
carefully coated with palm-oil and redwood powder,
which is deposited in a rudimentary temple in the
neighborhood of the head of the family. The bieri
represents the Lares of the ancients, the protecting
genius of the house and family: it keeps away the
sorcerers, destroys black magic, makes women fertile,
and procures riches. It is invoked in all difficult
circumstances, and, as they are persuaded that the soul
of the ancestor ordinarily lives in it, they sacrifice to it
food and different objects which the Pahuins consider as
especially precious. Nothing of importance is done
without consulting them, and they can always tell, by
certain signs, whether the bieri approves or condemns
a certain undertaking."^ Except for the ancestor
worship, which seems to be a specific belief of these
Bantus, everything else is identical with the worship of
the dasiri among the Bambaras. This is further
shown by the connection with the totem of a snake or
other animal, in this case among the Fans,^ just as the
dasiri is connected with a lizard, snake, goat, etc.
^ R. Avelot, Note sur les pratiques religieuses de Ba-KaU, in Bulletins ei
memoires de la Society d'Anthropologie de Paris, series VI, vol. II, p. 213 f.
* V. Largeau, Eneyclopedie pahouine, Paris 1901, p. 336 f
' Trilles, op. eit., p. 49 flf.
Among the Malinke, the Nama takes the place of the
Bambara dasiri, although all the Mande people also
have the Nama, which, they agree, comes from Misango,
in the province of Beledugu, that is, it was originally a
local fetish.^ "The Nama
is the great chief of the
village, the beneficent genius, the anti-sorcerer, the
occult force and supreme element in the service of
goodness against evil. "^ The appearance of the
namatigi "the chief sorcerer" in Mexico as amanteca^
makes it necessary to assume nama is not the original
word in Malinke for what, to judge from Bambara
dasiri, should mean "protection, religion," or some-
thing like it. Now, we have Arabic

'aman "pro- iJ^'
tection, safeguard, freedom from fear," which is found

in Berber aman, laman "religion, faith, security, safe
conduct, aid," Hausa lamuni "security," amana
"confidence, trust, security." In Malinke initial I
and n are often interchangeable, as in la, na "in,"
laho, nabo "go out," luntd, nunta "stranger," nu, lu
"habitation," loho, nofio "need," hence Berber laman
could appear as naman, and we really have nama.
In the Hausa country the bori worship has received
its most elaborate development,* but without adding
any substantial changes. From here the bori worship
has spread to the west and north. The possessor of a
bori, or, to be more correct, one possessed by a bori, is in
Hausa called maibori, in the plural masubori, and these
Hausa words may be observed far away from their
native home. Thus the women initiated into the Mendi
poro society are known as mabori^ or marbori.^
1 Henry, op. cit, p. 150.
2F. de Kersaint-Gilly, Le Nama, in Bulletin du ComiU d' Etudes historiques
et scientifiques, Paris IOIq, No. 4, p. 429.
' See p.
" A. J. N. Tremearne, The Ban of the Bori, London 1914.
6 T. J. AUdridge, A Transformed Colony, London 1910, p. 209.
• C. B. Wallis, The Advance of our West African Empire, London 1903, p.
246.
;
THE BORI 161
This chapter was all finished when I discovered ad-

ditional proof of the fact that the Mande Nama wor-
ship was, on the one hand, a degraded Islamic rite, and,
on the other, was related to the "hyena" by a philologi-
cal vicious circle. "The hyaena is the buffoon of the
animal world, and is deceived by the goat, the jerboa,
the ostrich, the jackal, the scorpion, the lizard, the dog,
even the donkey, and, of course, man; but he some-
times manages to avenge himself on the two latter.
The hyaena is a noted thief, and has a bad name, and
she very vain, being quite overcome by flattery.
is She
is fond of dancing and of music, and she once returns a
child to its mother because the latter has taught her a
song. She has some magic power of appearing and
disappearing (though this is not shown in the tales),
and is sometimes called amina, the friend, though for
what reason I could not discover. One man informed
me that the name is given because she tries to come
into a man's house at night, but it may be that the
Hausa magician resembles his colleague in North- West
Uganda in being able to make the hyaena take the place
of a dog, and in that case amina would be better trans-
lated by 'familiar,' perhaps. Another man said that
Amina was simply one of the names of the beast, she
having taken several so that she may have an advantage
in the division of food, as is shown in the following
story. Some of the animals had found a carcase, and
the hyaena, being the biggest present, said I will divide '
it up.' She took one quarter, and said 'This is for

Amina'; she took another fourth part, and said 'This
is for Burungu' (despoiler); she took a third quarter,
and said This is for Maihi derri (Traveller by night)
' '
and then she took the remainder and said 'Now the rest
'"^
is yours.
lA. J. N. Tremearne, Hausa Superstitions and Customs, London 1913,

p. 35 f.
Tremearne's surprise at the Hausa name amina can
fortunately be explained. One of the Arabic sobriquets
of the hyena is y^ r' 'ummu ^dmir, but we also have
y^ 'amir "the hyena." But the latter also means
"worshipper." In Hausa the more common Arabic
•^1 'aminah "the faithful one, believer," which sounds
very much like ^^ 'amir, has assumed the additional
meaning of "hyena" common Hausa kura.
for the
The Mandes have taken over the Hausa word for the
Islamic faith and, side by side with it, the Hausa kura
"hyena, " which, however, in Mande, is near in pronun-
ciation to koro "old," and thus the new fetish became
amankoro, later changed to namakoro "the old hyena."
The Hausa and Mande connections of the cult with
the hyena is not entirely gratuitous. The --^ 'umrah
was an ancient Arabic cult, consisting in the visiting of
Mecca,^ and, at the same time, y^ 'amir is "a înn,
inhabiting a house," which easily led to the idea of
"prowling," hence "a worshipper" and "a hyena"
suggested themselves alike from the same philological
root.
1 Wellhausen, op. cit., p. 74 f.
"
CHAPTER VI.
Fetishism and Sxjfism.
Arabic 'J qara' "to read, receive instruction, recite

from the Koran, teach., greet, present in the name of
another," but more especially "to recite magic words,
to recite prayers overJi^Lêad, " ''J qurra' "holy man,
devotee" have in(Asantej^rodug ed an enfuaaous-num-
ber of words relati^g-te-fSushismT The original mean-
ing is'THfrsBTved ~mlcrEricana~" to pray, recite, repeat
prayers, ask or inquire of God, prophesy, soothsay
(said of Mohammedans)," hence Kramo, Kramoni
"Mohammedan." The form hand leads to kan "to
read, count, number, reckon," kankye "to pray, re-
hearse or speak a prayer, invoke or call upon a fetish,
kenkan "to read, count, tell, wail for the dead, " ka "to
speak, say, emit a sound, and (in compounds) by spoken
words to cause or incur debts, mischief, recommend,
reprove, reprimand, censure, utter an oath, swear,
foretell, predict," nka "to perceive, learn, hear, scent,
smell." In the Akra language we have similarly kla
"to divine, especially used of the fortune- telling and
divining of the Mohammedans," klamo "diviner, sooth-
sayer, especiallyMohammedan fortune-tellers." The
same root is found in the Mande languages, where we
have Mandingo karang, Soso kharan, Bambara kalan,
kran, Malinke kara "to read, study, teach," Yei kar a,
karan " to learn. " The word has penetrated wherever
Arabic influence has been exerted, hence we have Hausa
karatu "reading, story, language, words, to learn,"
Peul karamoko "reader of the Koran," Gbari karatu
"reading, education."
In the ^'sante language^ this root has received its
.
fullestdeveTo^ment: ^rom "to read, to pray" we

pass to "to give a message," hence kara, kra "to tell a
message, send word, advertise, inform, give notice,
appoint, ordain beforehand, predestinate, take leave, bid
farewell;" nkra "errand, mandate, order, commission,
word, message, information, notice;" kra, kana, kena
"mark, visible sign made upon a thing for some pur-
pose, significant token, character made instead of
signature by one who cannot write; " akrasem "secrecy,
secret. " Similarly we get Akra ka "to say, reprove,
Bambara ka "word, voice, tone, accent, neck, responsi-
bility," Malinke kd "word, voice, noise, throat, neck,"
Soso khui "word, voice, noise, language."
We have already passed into the meaning "predict,
foretell,"hence it will be seen at once how we get
Asante okra, okara "destiny, fate, lot, luck;" nkra-hea
"fate, destiny, appointed lot, allotted life, final lot,
manner death;" akrade "luck, good luck, fortune,
of
godsend, a final present given by a trader or retail-dealer
to the peddler employed by him, a beloved, favorite
thing, a thii^-%elaaging to the soul." From this we
at once -pass vfKQkra, gk€trxr'HM.squlojt,m.a,n. Accord-
ing to the notions of "the natives tlie"4ia?*^ bf a person
existsJiefnre.Jhis_birth-and.jpiay be the soul or spirit of a
relation or othra:. person already dead that is in heavenT
or'with "GBdIand obtains leaye to eome again into this
^ world; when he is thus dismissed in heaven, he takes
with him" his errand, i. e. his destination or future fate
is fixed beforehand from this the name okara seems to
;
be drawn, and the realization of his errand or destiny

on earth is then called obra or abra-bo. The kara, put
by God or by the help of a fetish into a child, can be
askecJ.'While it is yet in the mothei-.'s-womb^ In life
the (karajs considOTed_p8aLtly_a^_0e sgjil or^spiriit of a
persotirpaJtly^ as a separate being, distincTlroin the
FETISHISM AND SUFISM 165
person, wlio protects him, gives him good or bad advice,

causes his undertakings to prosper or shghts and neg-
lects him, and, therefore, in the case of prosperity, re-
ceives thanks and thank-offerings like a fetish. When
the person is about to die, the kara leaves him gradually,
before he breathes his last, but may be called or drawn
back. When he has entirely left (whereby the person
^LLes)rie-^s~Jio more called kara, but sesa or osaman."
^ PI. akrg^,_a_male sl ave chosen b y his master to be his
ctyggant compariion""anT~destined to be sacrificed on
his~3eath in order to accompany and serv£LMD3,„iB-the
other_sailiL 'V" OA;ra,ôfearati;a, a J[e^m.aie~siflKe,_d.esiiiied
to bVsacnflced^n the death of her master."^ "AA;ra-
kwd,li slave, considered as the king^s okara; a soul-slave,
body-sklM^ page, valet de chambre/'^ Similarly we
have i^^c^'kla, okla "ghost, spirit, soul, genius, demon,
the slave~chosen by his master to be his continual com-
—
jjahion Snd according to the notion of some trjbfis^f
western Afric a—
to be s acrificed ovjai-hia-grave^haiAe
mayi>ccompanyJhircLia.tEe^.9rld to come. The word
IS one of the greatest difficulty to be_defined. Accord-
ing to the notion of the natives the%^ of a person exists
before his birth ajid m
ay be the soul or spirit of a^rela-^"
tionjor ^jiherjpiônalreadX- dead as "sooBras-arwr^manTs
;,
with child, she goes to a fetishpriest and asks the kla

of her child which is called by the priest, sujid-r-y^gues-
tions, which are answered by the priest whoCpretends)t6^
hear the kla etc. In life the kla is considereaptcrtly as
the soul or spirit of a person, partly as a being apart of
and without him, who protects him, gives him good or
bad advices, etc. receives thanks and thankofferings as
;
a fatish. Ever^jerson is_jnoreove r_gjipp^sed_to_haye

'two AZa, ja male and a feniale, the former being of a bad,-
~1ire4affer~ol a gooSTdisposition. After dmiiiJtiiS^Mgy.
1 J. G. Christaller, A Dictionary of the Asante arwTFante Language, Basel
1881, p. 254 f.
2 Ibid., p. 256.
heQomes_}§i&aS^ In the language of Christianity the
wor3~Hah^^brnlerly been used= doificov, afterwards it
was left unemployed."^

Asante kra-befwye "wonder, wonderful sight, worthy
to be advertised to persons dwelling elsewhere to come
and see" at once explains Bambara kaba "wonder,
miracle," Malinke kava "to admire," Soso kabe "to
admire," Bambara kaba "to admire, be astonished,
cry from fear," hence kabako "wonder, miracle."
The form of the word in the Mande languages is ob-
viously borrowed from the Asante kra-befwye or nkra-bea
"destiny)" and this is of importance as indicating the
difeFfion from which many Arabic words may have
come.
The conception that the spirit existed before birth
and that merely accompanying life, to be freed
it is
after death, is taken from the Sufls, and we shall soon
come across some Sufi terms in connection with African
fetishism. According to the Asante idea th^^^^Lla
the jjCconapanying-protecting spirit in life, hence we get
the corresponding Mande terms to mean "protector."
We have kalfa "protector, guardian, patron, to entrust
into one's care" and kana "to protect," Malinke karfa
"to confide," kanta "to protect, guard," Soso kanta
"to watch, guard, " hence we get Bambara kana "cure-
all, fetish medicine,"^ that is, "protecting amulet."
The close relationship of Asante fetishism with
Sufism may be gleaned from the Asante root kom. We
have kom " to dance wildly in a state of frenzy or ecstasy,
ascribed by the Negroes to the agency of a fetish;"
akom "in the state of being possessed by a fetish, " that
is, "a temporary madness or ecstasy, expressing itself
in dancing and wild gestures;" rikom "oracle, com-

munication, revelation or message delivered by God or
^ J. Zimmermann, A Grammatical Sketch of the Akra- or GSrLanguage,
Stuttgart 1858, vol. II, p. 151.
^ Henry, op. cit., p. 45 £f.
a fetish to a prophet or fetish man, prophecy, predic-

tion;" okomfo "a fetish man possessed with or proph-
—
esying by a fetish, soothsayer, diviner, charmer,
sorcerer; thev^w/o)pretends to be the interpreter jand
mou th-piece ^^hef onKe~guardian"'spirrt of a nation,
town or family, or of a soothsaying spirit resorted to in
sickness or other calamities;" nkommo "talk, chat,
concern, care, sorrow, soheitude, complain, lament,
moan;" nkongya, rikonya "miracle, wonder;" kum
"to tire, weary, wear out, cause to cease, defeat, over-
come, destroy, kill, disfigure, defile, pollute, to hinder
from using, stop, prevent, render ineffective, finish,
accomplish, put out the fire, be effaced." Similarly
we have Akra 'komo, nkomo "sadness, grief, to relate a
sad story, complain, commune with each other in
confidence," 'komoyeli "sadness, grief, sad story, story,
discourse."
At first thought, it would seem impossible to derive all
these connotations from a common root, but the whole
matter becomes clear as soon as we have before us
the Arabic group from which they are derived. We
have Arabic V gamm, "^ gammah, .gummah "grief,
mourning, lamentation, unhappiness, " "^ gumma'

"calamity, misfortune, a hard affair in relation to which
one knows not the right course to pursue, " J^ gumma
"he swooned, became senseless, and then recovered his
senses; an abstraction or absence of mind that over-
takes a man, with languor of the limbs by reason of a
malady." This is sometimes confused with "^
gamrah "difficulty, trouble, distress, the rigors of death,"
^ gumr "inexperienced in affairs, one in whom is no
good nor profit with respect to inteUigence or judgment
or work," from ^ gamara "to become much in
. "

quantity." In the circular dance of the Sufis, known
as ^ sama' , the dervishes worked themselves up into a
state of senselessness caused by ecstasy, '-r^ gamrah,^
when they had communion with God and uttered mys-
tical sayings, J^ gimar.
Thus we see how the Asante connotations have
changed from weakness " to " prophecy, fetish.
'
' The '
'
same root is found over a large territory. We have

Berber garni, gammi, gum, gumi " to be at the end of one 's
rope, to be harassed or fatigued, to refuse to do, resist,
Bambara komo "disagreeable, difficult, disgust, con-
tempt, " kuma "word, discourse, " Malinke kuma "word,
discourse." There can, therefore, be little doubt that
Malinke koma "chief of sorcerers," Bambara koma,
komo "fetish" are identical with the Asante kom. In-
deed the worship of the Komo is accompanied by las-
civious dances, howling, and ear-rending sounds. "At
dawn the warada thanks the audience in the name of the
god Komo, while talking through a horn tube, another
repeats the phrases through a reed pipe, and a griot of
the Komo renders the words of God more clearly to the
audience. And the crowd expresses thanks, utters
cries, puts forward its demands, and everybody retires
crushed, contrite, with his throat aflame, the spittle on
his lips, and with tottering limbs, —
during these festivals
one must become intoxicated, in order to please the
god."^
We have here the same ecstasy which we observed in
the case of the dasiri festival among the Bambaras, and
it is clear that we have here again a reminiscence of the
greater and lesser festivals of the Moslems. This is
made a certainty by the t5ai_grfiat^ festivals of the As-
' R. A. Nicholson, The kitdb al-luma' fi 'l-ta?awiimf, in "E. J. W. Gibb
Memorial" Series, Leyden, London 1914, vol. XXII, p. 145.
2 Henry, op. cit., p. 190 f
FETISHISM AND^ SUFISM 169
antes^Jkno wn as the ^dae festivals). "OnjtlieJareat an^

Little Adae~testivals, liis customary to offer sacjificea,
TiojEEe tfTBal and aational deities, the chiefs and men of
rank offering human victims, and_the; pporer^lasses
sheep and poultry. The festivals^ last^threedaj^ In
Coomassie, the commencement of an Adae feast is an-
nounced by the beating of the large state drum at sun-
set; and, upon this signal being heard, shouts, songs,
and discharges of musketry break out from all quarters
of the town. The Ashanti Government has utilised
thes©-festiv§ls_fgr the purpose of keeping up a species of
^TY§n^lce oVer all strangers in the capital. "About
'SIX days before theÂ^daejthe^M palm-wine/
-^s^'sMiJLlat whicH «very straBgeK-in, Coomassie is-b»und,/
to pay his respects in person to thekingj receiving ih
return a jar of paTm-wine, or some other small present.
On the Adae itself the king visjtsthe buildings within the
palace enclosure, where the ôols^f the f ormeE^Mngs
are preserved, and sprinMe^lreiat^th palm-wine, or
wmrtire-blbod of sheep and fowls, wEich" arekilled an^_
êookedrôr the asrahmanrfo-ai~lAie deceased monarchy
iBlood is also poured upon the stool of the dynastyJ
This duty accomplished, the king then proceeds wrEn
his chiefs, preceded by bands of music, to a part of the
city called Mogya-woh, literally, 'The blood dries.'
Here he seats himself, surrounded by his chiefs, and
strangers again present themselves, and are treated
with palm-wine. "^
Adae is "a festival day, returning every forty- third
day; one feast, called adae kese, akwasidae (adwedae)
is celebrated on Sunday; another, 24 days later, called
awukudae, falls on Wednesday. The king receives all
his elders and honoured guests in his residence and
gives them drink and presents. "^ This is from Arabic
' A. B. Ellis, The Tshi-Speaking Peoples of the Gold Coast of West Africa,
London 1887, p. 228 f.
2 Christaller, op. cit., p. 60.
"

•>^ 'id "fair held every ninth day, festival in general."
The greater and by this name have
lesser festivals
already been referred to.^ Thus we see once more
that the fetishism of JbheAsantes is mere "denatured"
Islamism. /
n,
"
We have Asantei^ni/aTOe,*''heavenj sky, the Supreme
Bdfl.g^Qxidî^^n2/aTOo;^:46^ small, faint,
drooping, languid," onyankopon "the visible expanse
of the sky, God, rain." In some of the connotations
this is confused with Asante Uwam "to become dry,
lean, languish, pine away." Tswam is from Arabic
1^ zaman "the withering, drying up of the lip from
thirst," ur*^ garni' "thirsty, lean, fleshless, " '^
zam'a "a hot, scorching wind," ^ zama' "long for
ardently, crave for;" 'c^ zim', pi. '^l azma', is "the
time, interval, or period between two drinkings, keep-
ing the camels from the water until the extreme hmit of
the coming thereto," hence the idea of "long journey"
naturally evolves from it, and we also get "the period
from birth to death." On the other hand, the root
nyam is evolved from Arabic !^ §am' "collection, as-
semblage, multitude, army," hence ^^ ^dmi' "the
collector of the created beings for the day of reckoning,
God, the mosque, a great town," "^^ gamd' ah "the
orthodox faith, the Moslem community, school, world.
The two roots have become equally confused in the
Mande languages. We have Bambara dyama "as-
sembly, reunion, village, " dyamani "country, province,"
but Songay dyam "artisan, smith," etc., which are ob-
viously from Arabic ^
gam' and dyama "to go on a
,
long journey," dyan "long, high," dyamandyan "very

' See p. 147 f.
long," which are derived from Arabic -^ zim' "long

jovirney. " "^^ gamd' ah also
Just as the Arabic word
means "the Moslem community, school," so dyama
has in Bambara and Malinke come to mean "church,"
hence we have Bambara dyamaso "church" and Wolof
dydmu "to pray to God, to render to God the worship
due to him," hence dyamome "the word used to greet
the king with," hence there can be little doubt that in
Malinke, Bambara dyamu "family name" we have a
derivative from the same Arabic word.
The tendency of African languages to abbreviate
words leads to a large number of contaminations, where
it is not at all easy to unravel them. Thus it is hard to
tell whether Hausa damina "rainy season" is the origi-
nal Arabic '^

sama', or whether there is here a con-
tamination with the Arabic words just discussed.
Similarly Soso nyamena "rainy season," by the side of
the sama words in the other Mande languages, makes it
hard to determine whether nyamena is derived from
there or from the same word as the Asante word for
"heaven," which is certainly at least a confusion with
the "dry" word in Arabic. But Soninke kamme
"rain" is, no doubt, a mere transformation of sama,
since here s and k constantly interchange, especially
since we have also Soninke kamu "heaven." Accord-
ing to Henry, the dynamic force
^ is in Bambara called
nyama or dya. The latter word will be discussed later.
Nydma may merely be a corruption of dya under the
influence of the "heaven" words, which is the more
likely, since we have in Asante a conception of such a
dynamic force, which is derived from a word "iio create
(said of God), " namely p&ra-i,the coming into the world,
the-state of existence, or life in this world:; maiiher of
life, conveT'saiiion, behavior, conduct;" bra, bar a "to
1 Op. cit., p. 26 fi.
'

make, enact a law or laws, to order with authority, to
lay an injunction, command, forbid, prohibit, settle,
come, become habitual, deceive, withhold, keep back;"
ohra "moral law;" abrabo "life in this world;" Akra
bla "to come again into the world, be born once more, to
be or behave in the world, to attach oneself to, behavior,
character of a person."
words are from Arabic ''-^. bara' "a writing
All these
conferring immunity or exemption, a manifestation of
excuse and a warning from God and his apostle;"
'^y. "(God) created mankind or the beings or
bara'
things that are created after no similitude or model,
but out of pre-existing matter; " 'y. birr, barr " goodness
in the service of God, in paying regard to relations,
acting well to them, and in dealing with strangers."
The group is represented in Malinke bila "to author-
ize, send a commission, leave alone, bring up an
animal;" Bambara bla "to authorize, put on one's ac-
count, accuse, give guarantee, condemn, simulate,
leave alone, permit." In neither language is the word
applied to the rebirth into a new body. In Asante
bra, Mandingo nydma we have a philosophic speculation
on the soul, which most likely belongs to the Sufis, for
we have already seen that life was represented as a span
of time between two drinkings, and Mandingo nydma
and dya go back to the same conception.
The was by many African
corLGep-tioajof 'theaveiij'
tribes directly derived from the Arabic. Arabic ^^^
qltgannah^''^]psbr8b6ise,'' from drr §an "to hide, conceal,"
isresponsible for an enormous number of words through-
out the northern part of Africa. Among the Berbers
we have igenni, ijenni heaven, firmament, atmosphere.
'
'
'
Koelle reports for "heaven" Biafada harddsenna,

Timne arlanna, Mandingo arldsenne, arddsenna, aid-
senye, aldsekne, Bambara ardsene, Vei aldênna, Soso

anyanna, andsanna, Hwida dsinukuzu, Mahi Sdsinik-
usu, Nupe aldsenna, Puka aldsenna, 'BornvC/tsannd,
dânna, Pika aledsanna, Bode aldsenna, ^lina, Kanika
dsinnawe^, Wolof hddsanna, Soninke aldsenna, Boko
alezonda, Kandin aldsenna, Idna, Timbuktu al§^nne,
Mandara aldsena, Bagrimi, Hausa, Peul aldseniia.
Many of these words can be rectified and expanded in
the light of later authorities. Timne ariann "heaven,
paradise" has also d-aren, r-aren "rainy season,"^
which, as we shall later learn, is the usual second conno-
tation Arabic '^ sama'.
of Bambara ardyana,
ardyine leads to a confusion with dyine "invisible spirit,
ginn." The Ewe languages (Hwida dsinukuzu, Mahi
edsinikusu) are most instructive. We have Dahome
jinukusu "heaven, firmament," which Delafosse^ takes
to be trovi ji-nuku-nusu "heaven-eye-cover," but this
is mere popular etymology, the root being jinu, which,
as usual in the African languages, is abbreviated and is
found as ji "heaven, rain, top, above," hence jijoho
"rain and wind, storm," jijoo, jijowd "storm," jikpa
"heaven," jikpame "garden." We get a far better
account of the root in the Ewe language,^ of which
Dahome is a dialect. Here we get dzinkusi, dzinko,
dzingo, dziiho'^ "heaven, firmament" and the apocopated
dzi "heaven, clouds, siirface, above, at." Just as
from "spirit" we got the concept "protector," so
here "heaven" leads to "help," hence dzi "to be at
one's side, to help." There is an enormous mass of
compounds from this root, such as dzidudu "to have
the upper hand, to reign," dzihose "faith. Christian
religion," dzime "above," dzimeanyisi, literally "wife
1 N. W. Thomas, Timne-English Dictionary, in Anthropological Report on
Sierra Leone,London 1916, part II, p. 3.
2 Manuel dahomien, Paris, 1894.
3 D. Westermann, Worierinich der Ewe-Sprache, Berlin 1905.
^ This ii is a bilabial /.
of heaven and earth, " but meaning "a priest of heaven
and earth, worshipped as a divinity," dziibadola
"angel. " In Akra we have similarly dzen, dze "world,
everything visible, atmosphere, weather, outward ap-
pearance, behavior, manner, circumstances, life, com-
monwealth, " hence dzenbd "behavior, character, com-
ing into the world, " bo dzen "to create the world, begin,
exist, live, behave, " hence there cannot be the slightest
doubt that dze "to come out, appear, become, arise"
is from the same root, and this brings us back to Ewe
dzi "to bear, be born."
Wehave already seen that Arabic "^^ al-gannah
produced Soso ariyanna, Timne arian "heaven," aren
"rain." This led to Xoruba prMnY"the sky, heaven,
bow, the invisible world, hades. " In.Asani0_no word
begms with a£^ Here Yoruba gruii has become oiKmT"
aJm^mu^âMmum "the air, atmosphere, the apparently
empty space above the earth," ahum "a strong wind,
gale, storm," honhom "a spirit," ahonhom "a mild,
gentle wind," but more commonly, as if from Timne
arian or Soso ariyanna, we have ewyim "the air, at-
mosphere, firmament, heaven, weather, the course of
things, " ewyi " the apparently vacant space encompass-
ing the earth, atmosphere, firmament, the revolving,
lucid air, the apparent arch or vault of heaven, the
course of things, the world," ewyiase "what is under
the sky or heaven, the world," wyiasefo "inhabitants
of this nether world, mankind."
The Arabic '^ sama' "heaven, firmament, roof,"

plural samiy, suman, also means "a cloud, rain,
heavy downpour. " In the latter sense it has survived
in a large number of languages. We have Bambara
samian, saminya, samia, samyen, somyen, Malinke
sama, samanya "rainy season." Koelle reports Kono
sama, Vei samara, Soso nyamena, Gbandi samai, Mende
"
hdma, Gbese sama, Bagrimi, Hausa damina "rainy

season. " We have already seen that in Timne d-aren,
Dahomeii "rain" we, by analogy, have similar deriva-
tions from Arabic -^'-^1
al-gannah. Koelle simi-
larly gives for "rainy season" Dahome dsi, odsi,
ozi, Mahi edyi, esi, Anfue edsinuali, Adampe esinole.
The latter two are dialectic forms of the fuller Dahome
juvenu "the long rainy season from March until June.
The-sh^rter forms at once explain Yrvriiba. n.d?.q^ edzi
" rain "although it also has the other Arabic word,
H,agps!y sarima "cloud."
In Bambara we have not only the above-mentioned
"rain" words, but also suma "fresh, cold, shade, odor,
rest" and "to measure, weigh, appraise, adjust, com-
pare, assimilate." We have also, from samian "rainy
season," sandyi "rain," sanfe "in the air, above,"
sangirigiri "thunder. " We find here a confusion with
another Arabic root, namely iSj^ §ara "to buy, sell,"
which produces Malinke, Bambara sara "to pay,"
Soso sara "to buy, sell, prepare." In Malinke and
Bambara this has become contracted to san "to buy, ex-
change, " sd "marketplace," sani "purchase, sale."
Similarly we have Malinke sama, samanya "rainy
seas9(£7^^^ air, atmosphere, heaven, year, to buy, sell."
Iri Asante)the connotation "heaven ''-•leads to a whole
series\§L-sjOTitual words. We have 4urn(i " tqjiide, in
sgcrelj^' sunsumma" shadow, '^ nsttnsManl" the wateiLOf
a heavy shower of xain. overflowing thfi_grou,nd,- but
quickly~flSwmg away," s w»SMma_ "ahadei_stLadQw,''
esiim "darkn^s. " From these we pass th-suman
"cEaJffiu amul©tr-taliaman, worn_as_a remedy or pres-
any protecting
ervative_agaiDs±.-jevils jor- mischief,
po^vBT'lnduding the abosom," osumanni "the owner of
a cEarfflT one who understands how to make amulets
and^sells them, sorcerer, magician." But we have
';
similarly the Arabic ^^ sama' "heaven:" asaman

"the world of spirits, the nether world, the lower regions,
the place of the dead; by some it is conceived to be in
the upper regions, the milky way being the road leading
to it. It is said: In the realm of the dead there are
kings as well as subjects (slaves). If you were sick in
this world for a long time, you will be restored to health
there after three years; but one who died in battle or by
accident will be well again in a short time, perhaps in
a month or so. It is said: the realm of the dead is
below (in the earth) some say: it is above (in heaven)
;
about this there is no surety. Where one is taken to,

when he dies, there his spirit is; when you die and they
take you to the spirits grove, then your spirit is in the
'
grove. The town (or country) of the departed spirits

is not in the grove, but in the earth it is a large town
;
(city), a long way off, and in going there a mountain

has to be ascended. The way of one who died a com-
mon death, is dark in heaven; but if one who died in
battle or by accident takes that way, some of the white
clay, with which he is rubbed, drops down, therefore
his way (the milky way) appears white. —
In the spirits
grove the departed spirits do not stay always; only on
certain single days they come and assemble there for
drinking or eating or playing;"^ osamaii "departed
spirit, ghost, goblin, spectre, apparition, skeleton of a
man. There are, according to the opinions of the
heathen Negroes^, three—diSecent kinds of de parted
spirits a) _^those who ffilMn battle (or by an accident,
:t.
as by a falling tree) ("^bL-Oommon spiritp; c) lingering

spirits. The last named are not admitted to the world of
spirits, where the others are, but hover about behind the
dwellings; the spirits of those who were killed do not
associate with the common spirits; they walk about,
rubbed with white clay and in white garments; they
1 Christaller, op. cit., p. 407.
are not afraid, whilst the common spirits flee when they
see a man, and do not wish even to be seen. "^ Here we
are once more brought back to a Moslem speculation,
for according to some views there are three categories
of spirits of the faithful: the prophets enter heaven at
once, the martyrs after a while, while all others linger
by the graves or with Adam in heaven.
The same confusion of the two Arabic roots is found
in Akra, where we get susuma "shade, character, re-
flection, soul," and susumo "measuring, thinking,
thought." In Yoruba, where we have alreadyJound
the "heaven" word, we have the ArabJi!Q;^_^£asurfi_J>
word best preserved, for we have aiuwo'^^: oSuwqti, and
the apocopated won "to measujcerêigh. " The ah-
breviated Akra-*ws:jf "to shadow, measure, think,"
\4gaJite-^s MSM, susuw }Fa,nte sitsu "to measure, think,
,
imagine,~suppt>"ser'nieditate, guess, utter a suspicion"

are also found in Ewe susu "to measure, length, mass, to
consider, regard, think, believe."
In Akra we find won, plural wodSi, "fetish, idol,
demon, something holy or belonging to the fetish^
The African theology is shortly the following: Goa
(Nyonmo, Nanyonmo, Mawu, Nyonmo Mawu) is the
highest Being, the only one, the creator of heaven and
earth; the fetishes, heaven, earth, sea, rivers, trees etc.
but considered as spiritual or personal Beings, are his
sub-deities, whom he has given the government and
care of the world, demons, good and bad, male and
female; there are such common to all (f.i. earth, sea)^/
or to a part of men (rivers etc.) to a tribe, a town, a
;
family, a single person a person

; may possess a fetish or
demon or be possessed by one. Besides there are
innumerable things holy to, or belonging to, or made
effectual by, a fetish, as cords, to be tied about the
' Ibid.
body or the house; teeth, chains, rings, etc. worn and
the like; which gave rise to the absurd that the
belief,
African makes any thing, even a bottle, a cork etc. his
God: and hasty travellers and other people not having
time to ask and to learn have sustained this saying,
whilst a comparison with religious things and super-
stitions in the very heart of Christendom would have
fully explained the matter without casting the African
together no more with men, but with brutes. "^ From
this we get a vast number of derivatives, such as
wondzamo, wonsuomo, wonUumo "fetish service, idol-
atry, heathenish reUgion, " wontSe, wontsemei "possessor
of a fetish," woyeli, wonyeli "fetish-eating, "wolomo
"highest fetish priest." This is all derived from
Arabic °'^y 'udzah "a kind of amulet, phylactery, or
charm, bearing an inscription, which is hung upon a
man, or woman, or child, or horse, to charm the wearer
against the evil eye and against fright and diabolical
possession." That this is the real origin of the Akra
word follows from Akra wulo, wolo, plural wodzi, "skin,
hide, leather, parchment, paper, book, note." The
plural in both words has preserved the Arabic form,
while in the singular the word has been abbreviated to
wo and lengthened by new suffixes. In the Adanme
dialectwe find womi "skin, parchment, paper, book,
letter"and wo "fetish, idol, demon."
In Ororuba\thfi_j\r-0rd-appears as J)7ide" fetish tied to
the bo dyj" in Dahome as vodu "good or bad spirit,
fetish," hence voduhwe "temple," voduno "priest,"
while Ewe has the abbreviated dzo "fetish, magic,"
with a very large number of derived words. Asanlfi.
has received- its words from-Akua^f or we have woma,
rihoma, Fante ahoma, nwowa "skin, leather, paper,
^ Zimmermann, op. eit., p. 337.

book, "but Akra wonsuomo "îdolatry"

letter, epistle,
produced^^ Asante abonsam'\ wizardV'sofcere r, dem on,
'
devil, "ahosonsom "fetish service,id_oiElry,IieatJiei]iim7^'

a5osoTO"'*ti^SlSr:j9r _giiardiaii.spiri_t.pf a town or family,
imaginary spirits subordinate to. God worshipped or
consulted by the Negroes. " In the latter case we have
a phonetic confusion with Asante words derived from
Arabic sama, even as this process may be observed in
the case of Yoruba won,^ where a similar confusion of
Arabic samd and 'udzah has taken place. It is this
common phenomenon of phonetic confusions that makes
African philology so extremely difficult, as already ob-
served by Steinthal.
1 See p. 177.
CHAPTER VII.
The Caeaibs.
L^ry^ gives the following account of the Caraibs in

Brazil: "It must be known that they [the Brazilians]
have among themselves certain false prophets whom they
call Caraibs, who, going from Adllage to village, like the
carriers of indulgences in papacy, make them believe
that by communicating with the spirits they may in
this way not only give power to whom they please, in
order to vanquish their enemies, when they go to war,
but also that it is they who make grow the large roots
and fruits, which I have said elsewhere this country of
Brazil produces. Besides, as I have heard from the
Norman truchemens, who have lived for a long time in
this country, our Totioupinambaoults have the custom
of assembling every three or four years in a great festival,
and I, finding myself there without thinking of it (as
you shall hear), can tell you the following for a fact.
So when another Frenchman, called Jaques Rousseau,
and I with a truchement were traveUng through the
country, having one night slept in a village called
Cotina, early the next morning, as we were thinking of
passing on, we saw, to begin with, savages from neigh-
boring places arriving from all sides; with whom those
of the village, leaving their houses, joined and were soon
gathered in a large space, numbering five or six hundred.
So we stayed on, to find out for what purpose this as-
sembly was collected, and when we returned we sud-
denly saw them separate into three groups, namely the
' J. de L6ry, Hisioire d'un voyage faict en la terre du Bresil, Paris 1880, vol.
II, p. 67ff.
THE CARAIBS 181
men in a house apart, the women in another, and the

children similarly. And seeing ten or twelve of these
Caraihs among the men, and suspecting that they would
do something extraordinary, I insisted on my compan-
ions staying there, in order to see the mystery, which
request I was granted. Thus, after the Caraihs had
told the women and children, before they separated,
by no means to leave the houses in which they were
stationed, but to listen to them attentively from there
when they began to sing, and after we were similarly
ordered to keep inside the house in which the women
were, and soon after we had breakfasted, without
knowing what they were going to do, we began to hear
in the house where the men were (which was not more
than thirty steps from where we were) a very low noise,
as you would say the mumbling of those who recite their
prayers.
"When the women, who were to the number of about
two hundred, heard this, they arose and pressing close
their ears gathered in a crowd. But when the men had
little by little raised their voices, and we distinctly heard
them sing together and frequently repeat the particle of
encouragement he, he, he, he, we were all so perplexed
when the women on their part answering with a tremb-
ling voice and repeating the sameinterjection he, he, he,
he, began to cry in such a fashion for a period of more
than a quarter of an hour, that we, looking at them, did
not know what to think of it. And, indeed, not only
because they wailed so, but because at the same time,
jumping up with great violence, they caused their
breasts to shake, and foamed at the mouth, and some
(like those who with us suffer from the falling sickness)
fell all in a swoon, I thought that nothing but the devil
had entered their bodies and they suddenly became mad.
When I similarly heard the children quake and torture
themselves in the same way in the dwelling where they
were separated, which, was close to us, I, in spite of the
fact that for more than half a year I had visited the
savages and had otherwise become accustomed to them,
cannot disguise the fact that I was somewhat frightened
and, not knowing the issue of the game, wished myself
back in our fort. However, when these confused noises
and bowlings were ended, the men making a short pause
(and the women and children keeping quiet), we heard
them once more singing and raising their voices in a
remarkable accord, and I, having reassured myself to
some extent, when I heard these sweet and more
pleasing sounds, it need not be asked if I wanted to see
them close by. But when I wanted to go out in order
to approach them, not only the women held me back,
but our truchement said that in the six or seven years
that he had been in the country he had never dared to
be among the savages during such a feast, and he
thought that I should not be acting wisely if I went
there, and might put myself in danger. I was some-
what in suspense, but considering the matter more
carefully it appeared to me that he had not given me
any good reason with his statement. Add to this the
fact that I was sure of the friendship of some good old
men who lived in this village in which I had been four
or five times, partly by force and partly ofmy own will;
and so I hazarded to go out.
"When I approached the place where I heard the
singing,and since the houses of the savages are very
long and round-shaped (something like the trellises of
the gardens over there) and covered with grass to the
ground, I, in order to see better and at will, made with
my hands a little hole in the covering. Making with
my finger a sign to the two Frenchmen, who were look-
ing at me, they followed my example; having taken
courage and come near without let or hindrance, we
three entered the house. Seeing that the savages (as
THE CARAIBS 183
tlietruchement judged) were not disturbed by us, but,

on the contrary, keeping their rank and file in an ad-
mirable manner, continued their songs, we nicely re-
tired into a corner, and observed them to our hearts'
content. But having promised above, when I spoke
of their dances in their drinking bouts and caouinages,
that I would also tell of their other manner of dancing,
so, better to represent them, here are their looks,
movements, and countenances as they showed them.
Close to each other, without holding each other 's hands
nor budging from the spot, placed in a circle, bending
forward, swaying the body a little, shaking only their
legs and the right foot, each one at the same time hold-
ing the right hand on the hip, the left arm and hand
merely hanging loosely, they sang and danced in this
fashion. Besides, since on account of the large crowd
they formed three circles, having within each three or
four Caraihs richly attired with robes, hats and bracelets
made of new and manicolored natural feathers, each one
holding in his hand a maraka, that is, a rattle made of a
fruit larger than an ostrich egg, of which I have spoken
elsewhere. In order that the spirit should later talk
through them and initiate them to this usage, they
rattled them at every rest, and you could not do better,
in the state in which they then were, than compare them
with those bell-ringers or those cheats who over there in
deceiving the poor people carry from place to place the
relics of Saint Anthony, Saint Bernard, and other such
instruments of idolatry. Outside of the above descrip-
tion I have tried to represent to you the following
picture of a dancer and player of the maraca.
" Then these Caraihs, advancing and leaping forward,
and falling back, did not always stay in one place, as
did the others. I also observed that, taking frequently
a wooden pipe, four or five feet long, at the end of which
there was the herb petun dried and burning, turning
round and blowing in all directions the smoke upon the
savages, they said to them: *In order that you may
overcome your enemies, receive the spirit of power;'
and thus did these master Caraibs several times.
After these ceremonies had lasted about two hours,
these five or six hundred savages never stopping their
dancing and singing, there was such a melody that,
considering the fact that they do not know music, those
who have not heard them would never believe that
they could sing so well together in time. In fact,
instead of being somewhat afraid, as in the beginning of
this celebration (when, as I said, I was still in the
women's house), I now had in turn such a joy that, in
hearing the well-measured accords of such a multitude,
and especially the cadences and refrains of the ballad,
in which they drew out their voices and said, 'heu,
heuaiire, heiira, heiiraiire, heiira, heiira, oueh,'
I was
quite enchanted. Every time I think of it my heart
quivers, and it seems to me I still hear them. When
they wanted to stop, they struck the ground with their
right foot more strongly than before, and then, after
each one had spit out before him, all in a raucous voice
said together two or three times, 'he, hua, hua, hua,' and
then stopped.
"And as I did not yet understand their language
perfectly and they had said several things which I had
not been able to understand, I asked the truchement to
tell me, and he said that in the first place they had
lamented their deceased forebears who had been so
brave. Finally they consoled themselves because
after their death they would surely go to find them be-
yond the high mountains, where they would dance and
rejoice with them. Similarly they had sworn dire
vengeance on the Ouetacas (a hostile, savage tribe, who,
as I have said elsewhere, are so brave that they have
never been able to conquer them) and that they would
,
THE CARAIBS 185
soon be taken and eaten by them, as their Caraibs had

promised them. Moreover, they had introduced into
their songs the story that the waters had once swelled
so much that they covered the whole earth and that all
people except their ancestors, who had saved themselves
upon high trees, had been drowned. This last point,
which among them approaches most the account in
Holy Writ, I have heard them reiterate several times.
And, indeed, it is likely that they had heard something
of the flood from father to son, who, from the time of
Noah, in keeping with human custom, had always
corrupted and turned the truth into a lie, and being,
besides, as we have seen before, deprived of any manner
of writing, they find it hard to keep things in their
purity, and have invented this fable, just as poets do,
that their ancestors saved themselves in the trees."
I have already pointed out the fact that Caraib as
the denomination of a tribe is a ghost word,^ and is
based on Columbus' attempt to identify the Indians
with a people of the Great Khan, hence Cambalu, for
which he was looking, appealed to him more especially.
But there must have been some specific reason why he
chose Cariba, Caniba, Canima as corruptions of such a
local name. In Brazil the words carai, caraiba, cary'ba
were not used in regard to an Indian tribe, but to
strangers, more especially such as practised magic,
hence Lory's story of the Caraibs is of great interest,
in so far as it confirms our deduction. Columbus
knew that some class of Indians, who were addicted to
cannibalism, were thus called, and his unfortunate
generalization of the name in time led to the application
of this word to the tribes hostile to the Spaniards. In
1511 a number of island Indians revolted against the
Spaniards and killed a number of white men. It does
not appear that they ate any of their bodies, but the
1 Vol. I, p. 42 f.
.

fact that they were hostile to the Spaniards at once
caused them to be classed among the Caraibs, and
they could be raided and sold into slavery, while those
who had submitted to the government could not be
seized.^ In 1520, Judge Figueroa of Hispaniola gave a
summarized statement as to what tribes he considered to
be guatiaos, that is, pacific, and what tribes were caribes,
and it is evident from his conclusion that as soon as a
caribe tribe became pacific it was no longer caribe, but
guatiao: "A las cuales dichas provincias e tierras de
suso declaradas por de caribes, debo declarar 6 declaro
que los cristianos que fueren en aquellas partes con las
licencias 6 condiciones e instrucciones que les seran
dadas, puedan ir e entrar, 6 los tomar 6 prender 6
cautivar 6 hacer guerra 6 tener, e traer, e poseer, e
vender por esclavos los indios que de las dichas tierras
6 provincias e islas asi por caribes declarados pudieren
haber en cualquiera manera, con tanto que los cristianos
que fueren a lo susodicho no vayan a lo hacer sin el
veedor 6 veedores que les fueren dados por las justicias
u oficiales de Su Majestad que para las dichas armadas
diesen la licencia; 6 que Ueven consigo de los guatiaos
de las islas 6 partes comarcanas S. los dichos caribes, para
que vean ^ se satisfagan de ver como los cristianos no
hacen mal a los guatiaos, sino ^ los caribes, pues los
dichos guatiaos se van 6 quieren ir con ellos de buena
gana. "^
Guatiao is for the first time given by Herrera under
the date of 1502. "The Indians of the province of
Higuey, seeing themselves in extreme misery and in the
hills, sent to ask for peace, and the Governor granted it
to them, offering not to do them any ill if they agreed
to give a part of their bread work to the King. Many
caciques came to see Juan de Esquivel, as General of
' Coleccidn de documentos inediios de Ultramar, series II, vol. V, p. 258 flf.
'^
Coleccidn de documentos iniditos relativos al descubrimiento, conquista y
colonization de las posesiones espanolas en America y Occeania, vol. I, p. 383 f
. "
THE CARAIBS 187
this emprise, and among them was Cotubanama,

powerful and valiant, and of honored presence, who,
from that time on was called Juan de Esquivel, because
it was a league among the Indians to exchange names,
and, when they did so, they became guatiaos, which
meant as much as 'confederates,' and 'brothers in
arms.' "^ Las Casas similarly says: "This change of
names in the common language of this island was called
to be I and such a one, who change names, guatiaos, and
thus one called the other; they considered themselves
closely allied and as establishing a league of perpetual
friendship and confederation, and thus the Captain
General and that lord became guatiaos."^
It does not take much to observe that we have here
Arabic ^.^J wadtah "subject, convention, protection,"

^.^J wadt "quiet, mild, gentle, peaceable," from t^J
wa^da "to deposit, put in safety, make a treaty, enjoy
in peace." The Arabic word is unquestionably from
the LLatin vadium, Latin vadimonium " surety, pledge,
but it has been confused with the Arabic -^J wad "prom-
ise," which has affected a large number of European
words. Thus we have Gothic wadi "pledge" by the
side of gawadjon "to promise in marriage."^ In
Africa the two produce an interesting series of words.
We have Berber uad "to promise, make a vow, con-
voke," udda "promise, vow, alms," Songay wadu
"promise." In Hausa we get wada "treasure,"
wadata, wodata "riches." Asante records aguade
"goods, merchandise," aguadi "trade," aguadini
"trader," hence egua "public place, market, trade,
council," hence oguasoni "member of a council,"
1 A. de Herrera, Historia de las Indicts, Madrid 1730, dec. I, lib. V, cap. 4.
See also dec. I, lib. VII, cap. 4.

2 B. de Las Casas, Historia de las Indias, Madrid 1875, vol. Ill, p. 47 f
' Of the European development of the two Arabic words I shall treat in
another work.
oguasonipa "gentleman." Asante egua explains Yor-
uba owo "trade." The series of words tliat interests
us here is in the Mande languages. Here we have
Malinke wondi, woli, Bambara wali, woli "one's
neighbor." In Wolof we have ande "friend." In
all these there a contamination with Arabic (Jj wall
is
"relative, neighbor, friend." Hence it is clear that in

the Mande languages the Arabic wadtah had
"^.^J
assumed the meaning "a pledge, an exchange between

two persons which constitutes friendship," which
evolved from the idea of barter that led to the Asante
"trade, market," etc., and the exchange of names
among the American Indians is just such a barter.
The Caraib soothsayer is more commonly called paje or
page, and Stade tells of them as follows: "They believe
in a thing which grows, like a pumpkin, about the size of
a half-quart pot. It is hollow inside; they pass through
it a stick, cut a hole in it like a mouth, and put therein
small stones, so that it may rattle. Herewith they
rattle when they sing and dance, and call it Tammaraka.
"
Of these, each of the men has one of his own. Now,
there are among them some who are caUed Paygi;
these are esteemed among them as fortune-tellers are
here. The same travel through the country once a
year to all the huts, and assert that a spirit had been
with them, who came from foreign places far off, and
had given them the power to cause aU the Tammaraka
(rattles), which they selected, to speak and to become
so powerful as to grant whatever was supplicated from
them. Everyone then desires that this power might
come to his own rattle. Upon this they make a great
feast, with drinking, singing, and soothsasdng, and they
perform many curious ceremonies. The soothsayers
thereupon appoint a day in a hut, which they cause to
be vacated, no women or children being allowed to re-
THE CARAIBS 189
main therein. Then the soothsayers command that

each shall paint his Tammaraka red, ornament it with
feathers, and proceed thither, and that the power of
speech shall be conferred upon them. Hereupon they
go to the hut, and the soothsayers place themselves at
the head, and have their Tammaraka sticking close to
them in the ground. The others then stick theirs also
hard by: each one gives these jugglers presents, which
are arrows, feathers, and ornaments, to hang to the
ears; so that his Tammaraka may on no account be
forgotten. Then, when they are all together, the
soothsayer takes each man's Tammaraka singly, and
fumigates it with a herb, which they call Bitten. Then
he places the rattle close to his mouth, and rattles there-
with, saying to it: 'Nee Kora, now speak, and make
thyself heard, art thou therein?' Presently he speaks
in a soft voice, and just a word or two, so that one
cannot well perceive whether it is the rattle or he who
speaks. And the other people believe that the rattle
speaks; but the soothsayer does it himself. In such
manner he proceeds with all the rattles, one after the
other: each one then behoves that his rattle contains
great power. Thereupon, the soothsayers command
them to go to war, and to capture enemies, for that the
spirits in the Tammaraka desire to eat the flesh of slaves.
Then they go forth to make war.
"Now, when the soothsaying Paygi has made gods
out of all the rattles, each one takes his rattle, calling it
his dear son, builds for it a separate hut, wherein he
places it, puts food before it, and demands from it
everything that he wants, just as we pray to the true
God. These are now their gods. Of that very God
who created heaven and earth, they know nothing,
they consider the heavens and the earth to have existed
from eternity; and they know nothing particular about
the creation of the world.
"For they say, once there had been a great water,
which had drowned their forefathers, some of whom had
escaped in a canoe, and others on high trees. Which,
I opine, must have been the Deluge.
" Now, when I first came among them, and they told
me thereof, I thought there was such a thing as a devil-
spectre (evil spirit) for they had often told me how the
;
rattles spoke. But when I went into the huts wherein

were the soothsayers, who were to make them speak,
they were all obliged to sit down; and, seeing the impos-
ture, I went from out of the huts, thinking what an un-
fortunate beguiled people it was. "^
We have here the specific statement that the canni-
balism was of a religious character, "for that the spirits
in the Tammaraka desire to eat the fiesh of slaves,"
and we are also told that the Brazilians did not eat
their enemies from hunger, but from great enmity.^
Although pure cases of cannibalism for mere enjoy-
ment may have occurred, it is significant that fallen
enemies were not devoured, but those who were speci-
fically kept for the occasion, after they had been given
a wife to solace with. The cannibalism is clearly
connected with the activity of the Caraib, the priest,
and this brings us back to the African deteriorations of
the Arabic qara', the reader of the Koran, the learned
priest who sacrificed to God. Even before Islam the
Hebrew '^'î? qara "the reader of the Bible" is found
in Syriac qarditd not only as "invocation to God,"
but also as "invocation to spirits," and this led to
LLatin caragius "soothsayer. " This word is found in
Caesar of Arelatum, in a church council of the end of the
V. century, and in Pseudo-Augustine. The latter is
certainly a late work, and the first, too, may have inter-
' A. Tootal and R. Burton, The Captivity of Hans Stade of Hesse (The
Hakluyt Society), London 1874, p. 145 B.

2 Ibid., p. 151.
THE CARAIBS 191
polations. In that ease the word got into the church

through the Arabic. This caragius is found in Norman
French as cards, queras, in OFrench as caraut. In
America the word seems to be confined to the eastern
part of South America, to the Tupi-Guarani and
Caraib languages. The variant forms cariba, caliba,
canima, carai are already found in the Mandingo lan-
guages, where we have the forms kara, kana side by side,
and ba or ma are the usual sufl&xes denoting agent. Thus
we have Mandingo karamo, Soso kharamokho "teacher,"
Bambara kanfa, kalanfa "professor," Mandingo kan-
tiba "supervisor," and Bambara forms like kariba,
kaniba are a matter of course,^ though not recorded.
Thus we are brought back to the Mandingo origin of the
name of the American fetish man.
The calabash rattle is, according to our sources, the
chief implement of the American fetish man, and in it
resides the speaking divinity. Early in the XVI.
century Cabega de Vaca found the gourd rattle in
universal use in North America: "At sunset we reached
a hundred Indian habitations. Before we arrived, all
the people who were in them came out to receive us,
with such yells as were terrific, striking the palms of
their hands violently against their thighs. They
brought us gourds bored with holes and having pebbles
in them, an instrument for the most important occasions,
produced only at the dance or to effect cures, and which
none dare touch but those who own them. They say
there is virtue in them, and because they do not grow
in that country, they come from heaven: nor do they
know where they are to be found, only that the rivers
bring them in their floods."^ "When we came near
the houses all the inhabitants ran out with delight and
great festivity to receive us. Among other things, two
1 H. Steinthal, Die Mande-Neger-Sprachen, Berlin 1867, p. 91.
2 T. B. Smith, Relation of Alvar Nunez Cabesa de Vaca, New York 1871,
p. 142.
of their physicians gave us two gourds, and thenceforth
we and added to our authority a
carried these with us,
token highly reverenced by Indians. "^ " For the pro-
tection of the messengers, and as a token to the others of
our will, we gave them a govird of those we were accus-
tomed to bear in our hands, which had been our princi-
pal insignia and evidence of rank, and with this they
went away. "^ Except for Alaska, the gourd rattle is
universal in the two Americas, and among some tribes
the sign for "rattle" also means "sacred."
The rattle, especially the gourd rattle, is the constant
accompaniment of the African fetish man, as we have
already seen. The shaman of all regions, in Asia as
well, uses the rattle in order to scare away the evil
spirits, but in Africa the use of the gourd as a
rattle
sacred instrument, in which the spirit resides, is due to a
linguistic misunderstanding. The Arabs did not use
bells to convoke the faithful, but merely called them to
prayer. But in the Mohammedan countries the
Christians for a long time preserved the use of the
sounding board, the '^J^ mitraqah, which was also
employed by magicians. In the Sudan this word is
found, both with the original meaning of "hammer,"
and, much corrupted, of "bell, rattle." We have
Berber lemterqet, Songay ndarka, Hausa matalaka,
muntalaka, Malinke manterge, Bambara mantaraka
"hammer." Songay ndarka at once explains Bam-
bara n'tana, n'dana, dana, tana, Malinke tala, Man-
dingo talango, Soso tolonyi "bell, rattle," which shows
that at an early time the Arabic mitraqah had in the
Western Sudan the meaning "rattle."
Arabic '^J qara' "to read," etc. was by the Arabs
confused with 'J qura "to tell fortunes," and the
1 Ibid., p. 149.
' Ibid., p. 191.
THE CARAIBS 193
divination called qura was permitted by Moslem law.^

This kind of divination, however, linguistically coin-
cided with the word which means "gourd, " for we have
qar, qarah, quraiah "gourd," and the latter also
means "rattle." We have, in Africa, Berber qraa
"divination," Hausa gora, kworia "calabash," kuria
"dice, lots," Bambara koro "meaning, comprehension,"
korofo "to translate, interpret," koroni "small cala-
bash," Malinke koro "meaning, sense, calabash."
In Asante we have kora "gourd, a vessel made from one
half of a dry gourd scooped out and used for various
purposes," and also apakyi, from some other root, "a
bread calabash," and mpakyiwafo "a man or woman
possessing a soothsaying fetish in a calabash, which,
when asked, he or she takes upon the head and, without
holding, lets it slip forward or backward, to the right or
left. " The Caraib rattle is obviously a direct develop-
ment of the Mandingo gourd rattle. The very name
maraca is the Arabic "^J" mitraqah. The Tupi diction-
ary records maraca "gourd rattle," maracainbara
"wizard, witch." In Guarani we have similarly
mbaraca "gourd rattle." In Arawak marraka vaesbiis
"gourd."
The imprecation of the Caraibs consists in a series of
songs or chants, of which the refrain is haiire. Simi-
larly the Mandingo Negroes call their talking devil
Hore.^ This is found in MaUnke as her a "peace,
tranquillity," in Bambara as her a, herd "peace, luck,
benediction. " When one asks, " Kori hera b6 How are
you?" the answer is "Hera Peace!" The usual greet-
ing is "Alia ma hera k^nye" or similar words, which
mean, "May God give you peace!" The same is
found in Berber as Mr, her "good," hence "Yaz-ik si-
Ihir May God reward you!" In Asante we get hyira
1 Doutte, op. cit, p. 375.
2 See p. 154 ff.
—
"to bless, wish happiness, invoke, bestow a blessing,
esteem, curse, blaspheme, decide, give validity, re-
nounce, resign." "Ohyiran'ano, 'he blesses his mouth,'
is used for some religious or ceremonial observances of
the heathenish negroes, viz. a) he washes at the water-
ing-place; —
b) he takes some water into his mouth and
squirts it into the calabash again, uttering certain peti-
tions to his soul (for money, length of life, honour, re-
covery of lost property &c.) or, he spurts the water to
;
the ground and invokes a blessing or a curse on others ;
c) he takes some consecrated fluid (water mixed with

some 'medicine') into his mouth, spurts it and mentions
something by which he brought a curse upon himself,
asking for the removal of the same, and for new bless-
ing. "^ We have also Yoruba ir^ "goodness, well-
wishing" and ire "a curse," r^ "to be good," rire
"goodness," Ewe yra "to pray for a person, ask a
blessing," yre "bad, malicious."
It will be observed that the Caraibs at the end of the
ceremony spit out, obviously to invoke a blessing or a
curse, just as in the Asante ritual. The whole thing is
of Islamic origin. "There is in the collections of
tradition a chapter, which is found in all the books of
the 'adab, generally in regard to religious injunctions
which have reference to commerce, and which is called
'istiharah. The prophet there recommends a special
prayer each time one finds oneself in indecision and
when it is necessary to make up one's mind. This
prayer is short God is asked to indicate the part one is
:
to take in such and such a circumstance, which is specifi-

cally mentioned: later one may make divinations by
writing on pieces of paper the different solutions pos-
sible in the matter, unless one feels a decisive inspiration
from above. Such is the orthodox Hstiharah: it is, to
sum up the case, a drawing of lots by invoking God, and
* Christaller, op. ciL, p. 208.
THE CARAIBS 195
resembles the qor'ah."^ "Jabir says: 'The Prophet

taught the HstiMrah, as he also did a chapter of the
Koran; and he said, "When anyone of you intends
doing a thing, he must perform two rak' ah prayers ex-
pressly for Hstihdrah, and afterwards recite the follow-
ing supplication: O God, I supplicate Thy help, in Thy
great wisdom; and I pray for ability through Thy
power. I ask a thing of Thy bounty. Thou knowest
all, but I do not. Thou art powerful, and I am not.
Thou knowest the secrets of men. O God! if the matter
I am about to undertake is good for my faith, my life,
and my futurity, then make it easy for me, and give me
success in it. But if it is bad for my faith, my life, and
my futurity, then put it away from me, and show me
what is good, and satisfy me. And the person praying
shall mention in his prayer the business which he has in
"='
hand.," '
Dozy^ says: "What

is called al-isti^drah, and at
'
Medina al-Jilrah, is a collection of religious practices by

which one consults God on things one is about to under-
take or on the issue of such an undertaking. One
purifies oneself, says the prayers of obligation (saZa/i),
or a prayer called §alah ul-'istiharah, recites a
supererogatory prayer (zikr), after which one lies
down to sleep and sees in a dream what one is to decide
upon. Or one recites three times the first and the one
hundred and twelfth chapter of the Koran and the
fifty-ninth verse of the sixth chapter, after which one
opens the Koran at random and draws an answer from
the seventh line of the right-hand page. " Lane, from
whom this information is taken, mentions the fifty-
eighth verse of the sixth chapter. As both are appropri-
ate and bear on our subject, the whole is given here:
"With him are the keys of the secret things; none
• Doutte, op. cit, p. 412 f.
" T. P. Hughes, A Dictionary of Islam, London 1885, p. 221.
5 SuppUment aux dietionnairea arabes, Leyde 1881, vol. I, p. 415.
knoweth them besides himself: he knoweth that which
ison the dry land and in the sea: there falleth no leaf
but he knoweth it; neither is there a single grain in the
dark parts of the earth, neither a green thing, nor a dry-
thing, but it is written in the perspicuous book. It is
he who causeth you to sleep by night, and knoweth
what ye merit by day; he also awaketh you therein,
that the prefixed term of your Lives may be fulfilled;
then unto him shall ye return, and he shall declare
unto you that which ye have wrought."
The divination of the 'istihdrah is in the Magreb
combined with the interpretation of dreams, wherefore
Doutt6 says: "The Magrebian 'istihdrah appears to be
nothing but an ancient incubation, not recognized by
Islam, and Islamized under the cover of the orthodox
'istihdrah, which had nothing to do with it. .If one
. .
could have good dreams and dream them aloud, it would

be a kind of oracle. In the books on magic there are
prescriptions how to make the sleepers talk, but they
are not particularly adapted for divination. Al-Bakri
says that in the Rif there were individuals called er
reqqdda, that is, sleepers, who fell into a lethargy and re-
mained in that stage for several days, making at their
awakening most astonishing prophecies. The fumi-
gations with incense, according to Ibn-Haldtin, put
certain individuals into a state of ecstasy, when they
foresaw the future."^ "The Egyptians place great
faith in dreams, which often direct them in some of the
most important actions of life. They have two large
and celebrated works on the interpretation of dreams,
by Ibn Shaheen and Ibn Seereen; the latter of whom
was the pupil of the former. These books are con-
sulted, even by the learned, with implicit confidence.
When one person says to another, I have seen a dream,'
'
the latter usually answers, 'Good' (hair), or, 'Good,
' Op. eit., p. 414 f.

—
THE CARAIBS 197
please God. ' When a person has had an evil dream, it

is customary for him to say, 'O God favour our lord
Mohhammad!' and to spit over his left shoulder three
times, to prevent an evil result."^
Just as in the Arabic practice, so in the African and
in the American there is a use of incense, in America
—
the incense par excellence tobacco, to produce or seem
to produce a lethargic state, and then to prophesy. As in
the Arabic, so in the Asante ritual, a curse or averting of
the curse is accompanied by spitting, and the same is
done by the Caraibs of the Brazilian ritual. And, to cap
the climax, the hair of the Arabic prayer, which gives the
very name to the whole ritual, becomes the name of the
"speaking devil" of Jobson. In reality, however, it
means exactly the same in Bambara as in Arabic, and in
Tupi is the refrain of the ritual songs. As in the African
festival of the tutelar deity, so in the American practice
we have the separation of the women and children, the
weird dancing, the shouting, the neiirotic state of the
comrtiunicants, the ensuing festivity, the offerings to the
fetish priest. That dreams were the chief means of
the Caraibs for predicting the future is distinctly men-
tioned by Stade: "When they desire to carry war into
their enemies' country, their chiefs assemble, and de-
liberate how they will do it. This they then make
known through all the huts, that they may arm them-
selves. And they name the fruit of some kind of tree,
when it becomes ripe, (as the time) when they will set
forth; for they have no denominations for year and day.
They often determine a time for setting forth, when a
kind of fish spawn, which are called Pratti in their lan-
guage, and the spawning time they call Pirakaen. At
such time they equip themselves with canoes and ar-
rows, and with provisions of dry root-meal, which they
call Vythan. Thereupon, they consult with the Pagy,
* Lane, op. cit., vol. I, p. 358.
the soothsayers, whether the victory will be on their
side. These, then, probably say 'Yes'; but they also
command them to pay attention to the dreams which
they dream of their enemies. If the greater number
dream that they see the flesh of their enemies roasting,
this means But when they see their own flesh
victory.
roasting, it they must remain at home.
bodes no good ;
Now, when the dreams please them well, they make

ready, brew much drink in all huts, and drink and
dance with the Tammaraka idols, each one begging of
his,that he may help him to capture an enemy. Then
they sail away. When they come close upon their
enemies' country, their chiefs command them, on the
eve of the day upon which they intend invading the
enemies' country, to recollect the dreams which they
may dream during the night. "^ Thus we are brought
to the positive proof that the Tupi ceremonial is identical
in substance with the Mandingo ritual, which itself is
based on the Moslem 'isti^drah.
> Op. cit., p. 151 flf.
;
CHAPTER VIII.
The Aeetto.
Oviedo says: **
The people [of the islands] had a good
and gentile way of remembering past and ancient events
and this was in their songs and dances, which they call
areyto, which is the same as we should say 'to dance
singing. ' This areyto they did in the following manner.
When they wished to have some pleasure, celebrating
among themselves a certain notable feast, or, without
this, just for their pastime, many Indians (and some-
times the men and women separately) came together,
and in the general feasts, such as in case of victory over
an enemy, or when the cacique or king of the province
married, or in any other event where the pleasure was
in common, and men and women congregated. And,
the further to extend their joy and pleasure, they some-
times took each other's hands, and at other times linked
arms, walking in a close file (or even in a circle), and
one of them took the office of leader (and it could be
either a man or woman), and he made certain steps
forward and back, in the manner of well-arranged
countersteps, and immediately the rest did the same,
and they walked around, singing in that high or low
voice intoned by the leader, and did as he did and said,
the steps being taken in perfect order and union, and in
keeping with the verses and words which they sang.
And as he says, the multitude responds with the same
steps and words and order; and as they respond, the
leader becomes silent, although he continues to take
the counterstep. When they finish the response, that
is, the repetition of what the leader has said, he immedi-
ately, without interval, passes to the next verse or
words, which the circle again repeats; and thus, with-
out stopping, he keeps them going for three or four
hours or more, until the master or leader of the dance
has finished his story, and sometimes he keeps them at
it from one day to another.
" Sometimes they mingle with the song a drum which
is made of a piece of round, hollow, concave wood, as
large as a man, and more or less, as they wish to make it,
and it makes a noise like the hollow drums of the
Negroes, but they put no leather upon it, and there are
only holes which pass to the hollow inside, whence it
rattles badly. With this poor instrument or without
it they in their singing (as was said) relate their past
events and histories, and in these songs they relate the
manner in which their caciques had died, and how
many there had been of them, and other things which
they do not wish to be forgotten. Sometimes these
leaders or dance-masters make a change, and, in chang-
ing the tune and the counterstep, they proceed in the
same story, or tell another (if the first is ended) in the
same tune or in another. . . .
"While these songs and countersteps or dances last,

other Indians come and give the dancers something to
drink, without their stopping while drinking, but al-
ways shuffling their feet and swallowing what is given to
them. What they drink is certain drinks used among
them, and when the feast is finished most of them are
intoxicated and without any consciousness, staying on
the ground for several hours. And when one falls
down drunk, he is removed from the dance, and the rest
continue, until the very drunkenness makes an end of
the areyto. This happens when the areyto is solemn
and made at weddings and funerals or on account of a
victory, or a victory and feast to be obtained; because
they have other areytos, without becoming intoxicated.
THE AREYTO 201
Thus some from this vice, others from studying this

kind of music, they all know this type of story-telling,
and sometimes other similar songs and dances are in-
vented by such among the Indians as are held to be
discreet and possessed of a better genius and faculties."^
Although the word areyto is not used there, we are
similarly informed by Oviedo that in Venezuela, at the
death of a cacique or chief Indian, all the people of the
village where he lived and his friends from nearby
places gather and weep through the night in a loud
voice and sing, relating in their song what the deceased
man had accomplished in his life-time.^ On other
occasions Oviedo uses areyto for the native singing and
dancing,^ but this is a mere generalization from the case
in the island.
The areyto is chiefly a eulogy on the dead and is in
substance identical with the eulogy at an Arabic funeral,
as described by Roger:* "If a man has died, all the
women, slaves, relatives, and neighbors, who are present,
begin to utter terrible cries, inviting all the other wives
to come and lament husband: who at once come
their
running from all sides, and on the way pick up handfuls
of dust, which they throw over their heads behind
them, covering those who are running behind, and
shouting loudly unknown, diabolical words, turning
their hands as though they were winding a skein of
thread, with bowlings and strange cries, so that it would
seem to be the confusion of Hell. When they are
gathered, in order to execute the ceremonies and lam-
entations which they call raqaz, they go to the hall or
yard, sometimes to a high and spacious place beyond the
house, and take a position in a circle, as though they
1 F. de Oviedo, Historia general y natural de las Indias, Madrid 1851, vol. I,
p. 127 fl.
2 IMd., vol. II, p. 297.
»
Ibid., vol. Ill, p. 142.
* F. E. Roger, La Terre Saincte, Paris 1646, p. 265 fi.
wished to dance without holding their hands. Then an
old woman, who hired for this sport, paints her face,
is
chest, hands, and arms black with soot, and, imitating
her, the same is done by the wives, sisters and daughters
of the deceased, who similarly blacken their faces, all of
them disheveled, in nothing but their shirts that are
open to the navel. The blackened old woman stands
in the middle of the dance, begins to tell all the prow-
esses and noteworthy deeds of the deceased man in the
form of a litany, and at each she makes a pause, while
the others repeat with a solemn and mournful voice,
dancing all the time with equal step. The relatives,
who are besmeared black, beat their breasts and cheeks
with the palms of their hands so that the cheeks swell
up, and continue this ceremony until the body is taken
to the grave.
"As soon as the man has breathed his last, they wash
thebody and wrap it in a pall, which they do not sew up
and do not tie at the head or feet. Then they place the
body on a stretcher, and upon it his turban and arms
which he has used, as also his scimitar, club, quiver,
and bow. If it is a woman, they put there her silver
miter, bracelets, and necklaces. After which, several
monks and mosque servants come to see him, to take
him to the tomb, without taking him first to the mosque.
His relatives and friends accompany him with gravity,
their arms hanging loosely like those of the monks, and
all of them sing the psalms of David, which they have
falsified by errors, and at intervals they stop and say
with a sad voice these words: 'Merciful God, be merci-
ful to him. There is no other God besides God!'
" The women follow the body from a distance as far
as the cemetery, where they walk over to some con-
venient place, in order to begin once more their dance,
not by the tune of a fiddle, but by the clicking sounds
which they make by striking their hands against their
THE AREYTO 203
cheeks and breasts with, such fury that they seem all
to be afire, their eyes glistening like candles and seeming
to drop out of their sockets. When the old woman tells
of something secret that has taken place during the
marriage, all the others stop, at the same time doubling
their shouts and bowlings and pronouncing diabolical
words.
"While this frightful lamentation takes place, the
turban or miter and other belongings are removed from
the body; then they place a pillow under the head,
without covering the body with dirt, because they build
a small stone chapel over it. The monks and relatives
leave the body at the cemetery and go home, the women
continuing for some time either to dance or to pray.
Sometimes the widow takes the deceased man's scimitar
and sways it with both hands like one mad, without
hurting anyone. When they are tired of their exercises
they go together home to the house of the deceased man,
where they sit down to a feast which the servants have
prepared during the lamentation and sport. Thus the
lamentation passes, to begin again next day at dawn
and to last again two or three hours and to be continued
for six or seven days in succession. Sometimes they
repeat their dances two or three times a day, especially
when some relatives come from without to console them.
It is to be noticed that the women of the Schismatic
Christians observe the same ceremonies and take part
in the dances of the Mohammedan lamentations, the
Mohammedan women similarly dancing at the Christian
funerals."
Precisely the sameritual is observed in the Western
Sudan. Bosman describes a Gold Coast funeral as
follows: "As soon as the sick Person is expired, they
set up such a dismal Crying, Lamentation, and
Squeaking, that the whole Town is filled with it; by
which 'tis soon published that some Body is lately dead:
;

besides which, the Youth of the Deceased's Acquaint-
ance generally pay their last Duty of Respect to him,
by firing several Musquet-Shot.
"If the deceased be a Man, his Wives immediately
shave their Heads very close, and smear their Bodies
with white Earth, and put on an old worn-out Garment
thus adjusted they run about the Street like mad
Women, or rather She- Furies, with their Hair hanging
upon their Cloaths; withal making a very dismal and
lamentable Noise, continually repeating the Name of
the Dead, and reciting the great Actions of his past Life:
And this confused tumultuary Noise of the Women
lasts several Days successively, even till the Corps is
buried.
"If a principal Man is killed in Battle, and his Com-
panions have no opportunity, by reason of the continu-
ance of the War, to secure, hide or bury his Body (for
the Funeral Rites must be performed in their own
Country) his Wives are then obliged in all that Interval,
to be in Mourning, and a shorn Head, though they permit
the Hair to grow again where Modesty does not allow
me to speak more plainly,
"A long time after, perhaps ten or twelve Years, as
Opportunity offers, the Funeral Ceremonies are re-
newed, with the same Pomp and Splendour as if they
had died a few Days past: On which Occasion all his
Wives again put on their Mourning, cleanse and adjust
themselves as before.
"Whilst the Women are lamenting abroad, the near-
est Relations sit by the Corps making a dismal Noise,
washing and cleansing themselves, and farther perform-
ing the usual Ceremonies The distant relations also as-
:
semble from all Places, to be present at these Mourning

Rites; he that is negligent herein being sure to bleed
very freely if he cannot urge lawful Reasons for his
Absence.
;
THE AREYTO 205
"The Towns People and Acquaintance of the De-

ceased, come also to join their Lamentations, each bring-
ing his Present of Gold, Brandy, fine Cloath, Sheets, or
something else; which 'tis pretended is given to be
carried to the Grave with the Corps; and the larger
Present of this Nature any Person makes, the more it
redounds to his Honour and Reputation.
" During this Ingress and Egress of all sorts of People;
Brandy in the Morning and Palm- Wine in the After-
noon are very briskly filled about; so that a rich
Negroes Funeral becomes very chargeable For after all :
this,they are richly cloathed when put into the Coffin

besides which several fine Cloaths, Gold Fetiches, high-
prized Corals, (of which I have several times spoken)
Conte di Terra, and several other valuable Things are
put into the Coffin to him, for his Use in the other Life,
they not doubting but he may have Occasion for them.
"
The Value and Quantity of his Coffin Furniture, is
adjusted in proportion to what the Deceased left his
Heir, or perhaps to the Heirs Conveniency. All this
being over, and the Relations and Friends met to-
gether; after two or three Days the Corps are buried;
before which a Parcel of young Soldiers go, or rather
run, continually loading and discharging their Mus-
quets, till the Deceased is laid in the Ground: A great
Multitude of Men and Women follow without the least
Order, some being silent, others Crying and Shrieking as
loud as possible, whilst others are laughing as loud; so
that all their Grief is only in Appearance.
"As soon as the Corps is in the Ground every one
goes where they please, but most to the House of
Mourning, to drink and be merry, which lasts for
several Days successively; so that this part of the
Mourning looks more like a Wedding than a Funeral."^
' W. Bosman, A New and Accurate Description of the Coast of Guinea,
London 1721, p. 220 ff.
There are two points in this description which must
be noted. In the first place, the Negresses naturally
paint themselves white, where the Arab women put on
soot. The Asantes similarly put on red paint. ^ In
the second place, the wielding of the scimitar among
the Arabs in Africa gives way to the use of the latest
weapon, the firing of guns, which is universal through-
out the Western Sudan.^ Another good description of
the funeral ceremony among the Mandingos is given by
Jobson: "One ceremony more of their Religion, I will
you please to remember, where and how I left
relate, if
the chiefe Mary-hucke sicke and full of danger, it did
manifest no lesse, for in the euening, the day after I
came from him, he died, the report whereof, was im-
mediatly spread ouer the whole countrey, who from all
parts came in, after that abundant manner, to solemnize
his funerall, so many thousands of men and women
gathered together, as in such a desart and scattered
countrey might breed admiration, which I thinke was
rather increased, in regard at that time he died, the
moone was high, and gaue her light, and they in whole
troupes trauelled, eyther the whole night, or most part
of the same together; the place or port whereat my
boat did ride, was a Passage or Ferry to the towne,
from the whole countrey, on the further side, where-
unto belonged a great Canoe, which I had hired, hauing
likewise another of my owne, both which neuer stood
still, but were vsed, night and day in passing the people,
none of them came emptie, some brought beeues, others

goates, and cockes and hennes, with rice, and all sort of
graine the country yeelded, so as there came in a
wonderfuU deale of prouision, my Mary-bucke entreated
mee, to send something of sweet sauour, to be cast vpon
his body, which the people much esteeme of; I sent
' E. Perregaux, Chez les Achanti, in Bulletin de la Societe neuchdteloise de
geographie, vol. XVII, p. 126.

2 Ellis, op. cit., p. 238 f., was at a loss to explain this custom.
"
THE AREYTO 207
some Spica Romana, and some Orras, which by his

Sonne was thankefuUy receiued: the manner of his
buriall, was after this sort, hee was layed in a house,
where a graue was digged, and a great pot of water set
in the roome, and iust after the same manner, as the
Irish doe vse, with a wonderfull noyse of cries and
lamentations, he was layed into the ground; the people,
especially the women, running about the house, and
from place to place, with their armes spread, after a
lunaticke fashion, seemd with great sorrow to bewaile
his departure. They also assembled themselues, in
the most conuenient place, to receiue the multitude,
and nearest vnto the graue, and sitting downe in a
round ring, in the middle came foorth a Mary-bucke,
who betwixt saying and singing, did rehearse as it were
certaine verses, in the praise and remembrance of him
departed, which it should seeme was done extempore;
or prouided for that assembly, because vpon diuers
words or sentences he spake, the people would make
such sodaine exultations, by clapping of their hands,
and euery one running in, to giue and present vnto him,
some one or other manner of thing, might be thought
acceptable, that one after another, euery seuerall
Mary-bucke would haue his speech, wherein they
onely went away with the gratifications, who had the
pleasingest stile, or as we terme it, the most eloquente
phrase, in setting forth the praises of him departed, in
which the people were so much delighted."^
The Arabic name of the funeral dance Jj raqaz is
preserved in Soninke rege, but the other Mande

languages have the same word for " dance " and "sing,
which indicates the close relation between the two as
established by the ritual, namely words derived from
Arabic oli' 'adzan "the chanting for prayer." We
1 Op. ciL, p. 70 ff.
have Berber âdden "to call to prayer, sing," Bambara
don "dance, play, to know, " and here the meaning "to
know," which is also found in Arabic o^^ 'adzana,
shows unmistakably the Arabic origin of the "dance,

sing" word. In Malinke we have do " to dance, " dokili
"
song, " Asante edzwom "song, hymn, psalm, " dzwonto,
dzwento "singing," hence adzwo "lament, wail." In
Akra dzo means "to dance." In Koelle we get Adampe
duo, Dahome, Anfue nduwe, Aku dso "to dance." All
these show that the Sudanese ideas of "singing" and
"dancing" are closely related to the Arabic religious
ritual. At the same time it must be remembered that in
the XIV. century Ibn-Batutah records the dancing and
reciting of a laudatory poem before the King of Malli by
a griot in a masquerade attire consisting of a bird mask
and feather ornaments. Whether of religious origin or a
development of the mime on Arabic soil, the dyala of Ibn-
Batutah represents the court fool in Europe, who, no
doubt, has a similar origin. In Africa he, as a poet,
easily becomes merged with the fetish man.
The Arabic ijj rasa^ "to sing the praises of one
deceased," '^"j' 'irsa' "elegy, in praise of a deceased
person," '^'y marsah "funeral speech, elegy on a dead
person, solemnity in honor of or lamentation for the
dead." These words are not recorded in the scanty
African dictionaries, that is, there are very few entries
under "lamentation" or similar words, hence but a few
hnguistic derivatives from this group may be found.
But we have more than enough to prove their presence
at an early time. Just as Bambara koroduga means
"griot, buffoon" as well as "dancer," so we have
Bambara malasa, from Arabic "^'^ marsah "elegy,"
' It would be better to write ratha, which in pronunciation among

non-
Arabs would lean towards rasa and rata.
AFRICAN DANCING, from Desplagnes' Le plateau
central nig'erien.
--w.-i-^
r"^""^
THE AREYTO 209
recorded as "to mystify, play," and malasaba as

"juggler, mystifier, " while Malinke marsa is "to play a
trick. " The funeral ritual, as well as the other festivi-
ties, proves through the vocabulary which expresses the
enjoyment of the dances and songs that the origin of
these is due to the activities of the griots, that is, origin-
ally of the Gypsies, as described by Ibn-Batutah. The
Mande dyala, dyeli " griot, musician " lies at the founda-
tion of words meaning "to amuse, laugh" in a large
number of languages. We
have Peul dyelli, dyalli,
Hausa dalia, daria, Wolof ree, Yoruba riri, rerin, Aku
reri, and in the Mande languages, Soso, Malinke yele,
Mandingo jelli "to laugh, make fun of. " In Bambara
we have dyelo, yele, yelema "to laugh, make fun of,"
yele ko, yele fen "amusement, that which causes one to
laugh. " In the Akra and Asante languages we can see
at once how this "laugh" word was applied to the fun-
eral ceremony. In Akra we have yara, yera, ya, yano
"funeral-custom, consisting of many ceremonies, as
washing, dressing and providing for the corps, as well
as the actual burial; weeping, lamentation, singing,
dancing, rum- or palm- wine drinking, gun-firing etc.,
sometimes days and weeks together. In later periods
all this is repeated. Formerly, and even now, when it
can be done secretly, men, especially wives and slaves
are slaughtered on the graves of people of importance
"^
to accompany and serve them in the world to come.
In Asante we have ayi "the funeral custom."
In America the funeral custom, as far as the singing
and dancing are concerned, in no way differs from the
Caraib festival, as, indeed, it is the same in Africa.
Among the Caraibs the African "laugh, make fun of"
word is used, as among the Akras and Asantes, in
connection with solemn occasions. Here we find
Gahbi eremi, ilemi, Caraib eremeri "to sing," and we
' Zimmermann, op. cit., p. 351.

get a good description of the ceremony by Breton:'
" Eremericaba lao eroutou, sing, dance, enjoy yourself,
for we are going to eat an Arawak. In the beginning,
when I was at Dominique, my host, Captain Baron,
having killed and brought from the continent an Ara-
wak, made a great celebration for all those who wanted
to be present and gave to each woman a piece of the
Arawak, to cook in her pot and eat with her husband
and family, who were at the assembly, which they did
with great joy during the day, for, after having drunk
and entertained themselves with their prowesses in
their harangues, when night fell they with faltering
steps and rolling eyes began to sing, dance, and howl
with so much vehemence and fury that I was frightened.
Leremericayem boye loubara arali racautiu, the boye
sings to make the gods come down. When the Caraibs
go to war or when they have some sick people, they call
a hoye, prepare an offering, which is placed at the further
end of the house, which is always round, and put there a
bed, while the people present seat themselves around
the wall. The boye having arrived (sometimes with
another person) he begins to sing, while one of the two
,
throws some tobacco smoke upwards in place of incense,

and in this way makes his pretended god come down
(I have heard them say that he falls down like a meal-
sack, but I have not heard the sound which they say
he makes with his fingers). The boye gives him the
bed to sit upon and the offering to eat and drink. This
spirit of darkness does not want any light, and makes
them put out the fire and close up all the avenues of
light, unless ittakes place at night. I once wanted to
enter with a fire-brand, in order to prevent this abomina-
tion, but the women stopped me."
This Caraib eremi, ilemi is, no doubt, related to the
Bambara yelema "to make fun of," but we have also
' R. Breton, Dictionaire caraibe-fravîs, Leipzig 1892, p. 216 ff.
THE AREYTO 211
Galibi oalitago, aoualetago "to sing," Arawak aritin

"to call, give a name," which show that a form areyto,
as mentioned by Peter Martyr and the other early writ-
ers, must have existed in the islands. Indeed, we have
Caraib eletouac "a solemn festival, at which six or eight
of the feasters rub their bodies with elemy gum which is
still in a liquid form, over which down or small feathers
of the phaeton bird (festu en queue) are sprinkled.
Their heads are crowned with large Arras feathers, then
they dance by twos around the council-house (Carbet),
one extending his right arm over the shoulders of the
other, and the other his left arm over his companion's
neck. The others follow in the same posture, dancing
by twos, until they come to the place where they find
the large calabashes full of wine (oiiicou), which has
to be swallowed to the last drop, even if they should
burst. I have seen them almost choke, grow pale, and
unable to stand it any longer; to get relief a savage
would embrace another from behind and press his
stomach until he would vomit a part, to make room for
more. "^
There seems to be little doubt that we have in these
Caraib and Arawak words the Arabic J-> rasa "to sing
the praises of the deceased, " since in the Tupi languages
we have a corresponding word from Arabic »^V marsah,
namely Guarani poragei, mhorasei, mborahei, in the
other Tupi languages morase, murasi, porasei "to sing,
dance, " while in some there is a confusion with maraka
"^
"rattle," hence Kamayura maraka "to sing, dance.
In the case of the Tupi word, there seems to have been an
attempt made to transform the word so as to bring it in
keeping with some popular etymology, for we have also
1 Ibid., p. 203 f . See also C. de Rochefort, Histoire naturelle et morale des
Isles Antilles de I'Amirique, Roterdam 1658, p. 455 f.
2 K. von den Steinen, Unter den Naturvolkern Zentral-Brasiliens, Berlin
1894, p. 315.
"

mogaray "to make fun of, triumph, play," mogaray
goer a "one made fun of, a fool, " mogaraytara "dancer.
A similar corruption of the original word is found in
Bakairi makanari "mask dance, anything connected
with the dance. " However this may be, we have here
the same semantic relation as in the African words.
In the case of these "song" words an absolutely certain
etymology cannot be established, not only because of
the lacunae in the African dictionaries, but also because
the "sing" words are easily confused with the haiire
refrain of such songs, from which the Caraib areyto
may equally well be derived. On the other hand,
there may be in all of these a derivation from the
African yele "song" words, since we have, by the side
of Galibi eremi, ilemi, also Cumanagoto huarage and
Bakairi ali, ari, ori "to dance, sing."
CHAPTER IX.
Feathers and Masks.
In 1189, on the eve of setting out to Jerusalem,

Richard I promulgated a specific law for those who
started out across the sea to Jerusalem, dealing with the
punishments for various offences. A thief was to have
his head shaven, and he was to be tarred and feathered,
and landed at the first port, where people would at once
know by his plight what his offence had been.^ No
such practice was in force in Europe, and it is evident
that Richard here provided a form of punishment which
was prevalent in the region to which they were going,
that is, among the Moslems. According to Islamic
law, a punishment called ta'zlr is imposed in those cases
in which there is no specific legislation. Thus ta' zlr is
applied in cases of forgery, deception, extortion, false
witness, calumny, and petty thieving. The judge has
in such cases discretionary powers. He may send the
culprit to prison, have him whipped, give him a repri-
mand, put him in the pillory, have his face blackened,
"Richardus Dei Gratia Rex Angliae & Dux Normanniae & Aquitaniae,
1
& Comes Andegaviae, omnibus hominibus suis Jerosolymam per mare ituris,
salutem. Sciatis nos de communi proborum virorum consilio, fecisse has
justitias subscriptas. Qui hominem in navi interfecerit, cum mortuo
ligatus projiciatur in mare. Si autem eum ad terram interfecerit, cum
mortuo ligatus in terra infodiatur. Si quis autem per legitimos testes con-
victus fuerit quod cultellum ad alium percutiendum extraxerit: aut quod
alium ad sanguinem percusserit, pugnum perdat. Si autem de palma per-
cusserit sine efifusione sanguinis: Tribus vicibus mergatur in mari. Si quis
autem socio opprobrium aut convitia, aut odium Dei in jecerit: Quot vicibus
ei conviciatus fuerit, tot uncias argenti ei det. Latro autem de furto con-
victus tondeatur ad modum campionis, & pix bulliens super caput ejus
effundatur, & pluma pulvinaris super caput ejus excutiatur ad cognoscendum
eum, &in prima terra, qua naves applicuerint, projiciatur. Teste meipso
apud Chinonem," T. Rymer, Foedera, London 1727, vol. I, p. 65.
etc.^ The judge has also the right to impose similar
punishments in cases not yet proven but indicating that
the offender may be dangerous to the community.^
Richard's law shows conclusively that tarring and
feathering was in use among the Moslems for petty
crimes. We have already seen that the J^-^ dagâl was
a low, contemptible fellow, and we have one distinct
reference to him as a musician, hence there can be no
doubt that the Gypsies, who later developed into the
griots, were considered dangerous to society as cheats
and offenders. Charlemagne's law of 789 shows that
as early as the VIII. century means were sought to stop
the nuisance. That the Arabs would cause the Gypsies
to be tarred and feathered or, at least, to wear a special
attire indicating this tarring and feathering is shown by
the very word tK-= dagala. In all the Semitic languages

except Arabic this root has no other meaning than "to
lie,cheat, " whereas in Arabic it has not only the mean-
ing of "he lied, told a falsehood, " but also "he smeared
his whole body with tar." As <J^^ daggal also means
"having one eye and one eyebrow, " it is likely that the
pestering Gypsies were also punished in this more
severe way.
From Ibn-Batutah we have learned that the dyala of
Malli wore the distinctive costume of feathers, and a
mask to accentuate the fact that he was "a bird."
Thus it becomes clear that a bird mask was the original
mask of the buffoon or griot, as he appears in the West-
ern Sudan. Once the origin of the custom was for-
gotten, the bird mask gave way to any other kind of
mask, but we are still able to see that the predominant
mask is that of a bird. " Masquerades now take place,
' E. Sachau, Muhammedanisches Recht, in Lehrbiicher des Seminars filr
orientalische Sprachen zu Berlin, vol. XVII, p. 849.
2 Ibid., p. 848.
—
FEATHERS AND MASKS 215
though these are mainly performed by Arabs, the only

one in which the Hausas have any part being the Bu
Sadiya, in which a man dresses up in a mask ornamented
with birds' feathers (to represent the head of an eagle),
puts vulture wings upon his shoulders and wears a coat
of pieces of various skins e.g. jackal, fox, hyena, and,
ifpossible, leopard and lion. But this is not confined
to the Salla; it may take place at any time, the main
care at present seeming to be a collection from the on-
lookers. Some of the Arab performances resemble the
dance of Jato at the bori. "^
In America the mask and the dance are intimately
connected with "tarring and feathering." We have
already seen this to be the case among the Caraibs.
Precisely the same is told of the Tupis: " They all come
painted in all kinds of colors, arrayed in feathers also of
a certain color pressed out from the sap and juice of
the fruit of a tree called genipat in their language, which
they esteem greatly, not that the fruit of this tree is
good to eat, but because of the quaintness of the hue.
This tree is large and has leaves like those of a wal-
nut tree, and the fruit like peaches over there, which
grow at the end of the branches in a strange fashion.
Having no other means for extracting this fruit, they
chew it, then reject it from the mouth and squeeze it
between their hands, as one would squeeze out a sponge,
and the juice that comes out is as clear as the purest
water: with this juice or sap they wash themselves
when going to their feasts or massacres or to visit their
friends. When this juice is dried up on their skin, it
has a deep black color which does not become perfect
unless it has had two days to soak into the skin. And
thus painted they go to a feast without any garments,
as contented as we are with our silk clothes, and the
women paint thus more often than men, and seeing
' Tremearne, The Ban of the Bori, p. 241.
them from, a distance you would think that they are
dressed in black velvet. It must be observed that
other savages who go to these feasts, in order to partake
of caouin with their friends, before leaving the villages
peel a certain tree, the inside of which is yellow or
red, and chopping it very fine they mix it with the gum
of a certain tree called usup. Then they spread these
colors over the first made of genipat. As for the usup
gum, it is very good for consohdating wounds, as I have
found out from experience. Painted in this manner,
they cover themselves with down, that is, with very
small and fine bird feathers, which they apply to the
said gum from their heads to their feet, when it becomes
a pleasure to contemplate these gentile savage parrots,
who seem all to be dressed in elegant red scarlet, and
they have also large feathers, with which they surround
their heads. "^ The same is told of the people of
Cumana: " In war-time they put on mantles and feath-
ers: for the festivals and dances they paint or blacken
themselves or smear themselves with a certain gum and
sticky unguent like birdUme, and then they put on
feathers of various colors, and these feathered Indians
do not at all look badly. "^ The custom is well-nigh
universal. In the interior of Brazil, among the Bororo,
feathering the body is of enormous importance.^ Not
only is it there used on festive occasions, but it is one of
the most important "medicines." All feather orna-
mentations, except, perhaps, an occasional ear adorn-
ment among the Bakairi, are a part of festive occasions,
including festive receptions. Feathers are of the same
importance as the painting of the body.* The same
'Thevet, op. cit., fol. 926 b.

^F. Lopez de Gomara, La historia general delas Indias, Anvers 1554, cap.
LXXIX, fol. 102 a.
' Von den Steinen, op. cit., p. 476.
' Ibid., p. 328.
J
?<
pq
o
z
5
w
<
[^
O
g
5
cq
z
s
K
<
Q
Z
<
o
z
5
Pi
FEATHERS AND MASKS 217
is true in North America, but the modern aspect of

feathers and masks does not belong within the scope of
our work. Sufficient has been shown to establish the
fact that both have their origin in the forced adorn-
ment of the African griot, as a result of the Arabic
"tarring and feathering," by which to distingmsh the
contemptible caste of the griots.
"
CHAPTER X.
The Caeaib Social Order.
In The Journal of the Second Voyage we have a refer-

ence to the native duho "a low seat. "^ Oviedo de-
scribed it for Nicaragua as "a small stool with four feet,
made of fine, smooth wood. "^ This chair was a sign
of nobility or royalty among the Indians and was also
connected with their religion. It is not dif&cidt to see
that this is the Persian- Arabic '^^" taht, plural tuhut, "a
royal throne, chair of state; sofa, bed; any place raised
above the ground for sleeping, sitting, or rechning; a
capitol, royal residence." The history of this word is
fascinating.
Ibn-Batutah uses the term tr-^ mansa for the king of
Main, and this term is still in use, for we have Malinke
masa, Bambara masa, Soso mage, Vei mandza "king."
This from Arabic '^^^^ man§a' "place where one grows
is
up, where anything originates, birthplace, one 's country,

origin, beginning, source, " from '^" naSa'
"to grow up,
originate, " as the history of the word in Asante and
Akra shows. We have Asante omansofwe "reign,
regency," amdn sdii "all people," but more commonly
the abbreviated man in omdii "town, people, kingdom,
omanha "citizen," etc., Akra mantie "king, first person
of a town," man, plural madzi, "town, people, nation,
kingdom, country."
' Raccolta di documenti dalla R. Commissione ColowMarui, Koma 1892,
part I, vol. I, p. 192.
" Op. cit., vol. IV, p. 109.
ASHAXn STOOL.
From Freeman's Travels and Life in Jshanti and Jaman.

THE CARAIB SOCIAL ORDER 219
Similarly the word for "royal throne" is in the

African languages derived from the Arabic '^^' faft^,
tuhut. The royal chair is always carried wherever the
king of Dahome goes.^ It is generally made of one
piece of wood and is highly ornamented.^ " To succeed
to the stool" is the expression in Asante for "to mount
the throne. "^ " Of their woodwork the Ashanti stool
is a fair specimen, which is cut from a solid block and
variously ornamented."* In Asante we, therefore,
have from Arabic tuhut the full form dufud "a rough
kind of seat made of a block," hence dzwa, gua "to
carve," adzwa, agua "seat, chair, stool, throne," egua
"public place, market," and this leads to Yoruba aga
"chair, stool, table." In the Mande languages the
meaning "chair" is not always preserved. Although
we have Soninke takhade, that is, Arabic taht "chair,"
we get Wolof dak, deuk "village, " and so in the Mande
languages, Malinke dugu "land, country, village,"
Bambara dugu, Mandingo duo, du "earth, country,
village, town."
Wehave no record as to the native name of the
Caraib chief, except that of cacique, which is more
likely a Spanish corruption of an African word than the
name current among the Caraibs, although it will
appear later that a similar word was in existence among
the Caraibs. We have so far an unmistakable proof
that his dignity was connected with that of the Man-
dingo stool, the duho, as amply recorded among the
early writers on America. But we also know that there
was a class of nobility, for, on December 23, 1492, the
Journal recorded: "Up to this time the Admiral was
unable to understand whether they use it [caeique] for
> L. Brunei et L. Giethlen, Dahomey et dSpendances, Paris 1901, p. 304.

=*
E. Foa, Le Dahomey, Paris 1895, p. 173 f.
3 W. Hutton, A Voyage to Africa, London 1821, p. 316.
* G. MacDonald, The Gold Coast, London, New York, Bombay 1898, p. 56.
'

but they have also another name
'king' or 'governor,'
for a 'grandee,' whom
they call nitayno. He did not
know whether they meant it for hidalgo or governor
'
'
'
or 'judge.' "^ We have already seen that among the

Mandingos n'tene represented the mark of
n'tana,
distinction, the same which in Europe led to heraldry,
hence in the nitayno we have the men who acquired
marks of distinction by colors and especial designs.
We shall later meet with them in Mexico, but here
may be pointed out the curious custom, recorded for
Venezuela, of acquiring degrees of nobility by colored
markings on the body: "I have heard of a manner of
military honor with which the natives of this land pre-
cede and are preferred and honored above the common
people and even above those of greater quality, and it
is a kind of nobility acquired by military discipline and
of this form. For a deed of valor the right arm is
painted with a certain mark or device of black color,
by drawing blood and putting into the wound ground
coal. Indeed, the painting is Like the one put on for
show by the Moorish women of Berbery in Africa:
the which painting can never be taken ofif, except by
destroying the figure. And from that time on such an
Indian is no longer a common man, but like an hidalgo
among the Spaniards, and marked as a miUtary man,
and esteemed from that time on as a brave man. And
when he gives another proof of his valor and has ob-
tained victory, he is like an hidalgo whom the king
dubs a knight, and then his breast is painted with some
such device as the arm. When he obtains his third
victory, they paint some Unes which go from the ends
of the eyes to the ears, and those who are thus adorned
are highly esteemed, and there is no other honor to
aspire to, which is as though he were a Hector or a
Bernado del Carpio, or a Cid Ruy Diaz, or whomso-
' Raccolta, part I, vol. I, p. 76.
THE CARAIB SOCIAL ORDER 221
ever you may honor. "^ As we get in the same region

the fullest and linguistically most correct form of the
fetish man, there can be little doubt that we have here
a description of the African n'tene which confers and
records nobility as the result of deeds of valor.
We have already observed that in Africa the griots
laid the foundation for a free corvâble population that
was occasionally confused with slaves.^ We there
met the Bambara word dyama "province, village,"
hence dyama horo, recorded in the compounded dyam-
uru "forced labor," dyamuru hugu "a free village,"
Wolof dyambur "free." This conception occurs in
the encomienda of 1516 or 1517,^ when the Indians
were turned over as servants to the Spaniards. Here
we frequently hear of "nahoria de casa, " that is,
household servants. These naborias are mentioned
by the side of nitaynos,^ who, together with their
caciques, were similarly placed in the encomienda of
some Spaniard. Thus we get among the Caraibs the
same social order as in the Western Sudan, except that
the naborias "free workmen, not yet distinguished by a
n'tene/' were the mass of the Indian population, who
were not slaves, whereas in Africa they were evolved
from the originally foreign griots.
'Oviedo, op. cit., vol. II, p. 322 f. In chap. XXII (p. 330) there is a
fulleraccount of this custom.
2 See p. 113.
' Coleecion de documentos inSdiios relativos al deseuhrimiento, conquista
y
eolonizacion de las posesiones espanolas en America y Occeania, vol. I, p. 50 ff.
*Ibid., p. 82.
CHAPTER XI.
The Boratio.
\ According to Columbus^ the cemi, cimi, or cimini

were certain wooden images, which, were worshipped
by inducing intoxication. The cimi represented the
cacique 's ancestor, and was made to speak by trickery.
Other fetishes were stones which were good to produce
fertility, make women bear children, and preserve
people against the sun and weather. Ramon Pane^
adds the information that the cimini are immortal, as
if in heaven, and have the bones of their ancestors,
which are made of stone or wood. The cimini come

to aid in the form of adders, just as the Fan fetish is
connected with serpents. The resemblance is still
further increased by the fact that the Fans use the skull
of an ancestor for a fetish. But the Bambara nama
and dasiri, too, represent the same qualities as the
cimi.
Rochefort,^ one hundred and fifty years later, says
that the Caraib men call their good spirits icheiri, that
is, iSeri, while the women call them chemijn, that is,
§emln, which the missionaries translate by "God,"

while the evil spirits are by the Caraibs called mapoya
or maboya. By the Galibis "spirit" is rendered by
issimei, plural issimeiri, so that the masculine iSeri may
be a corruption of the latter. In Arawak semetti is a
"witch-doctor." Hence there can be no doubt that
we have here a real Caraib-Arawak word meaning
' See my Africa, vol. I, p. 67 flf.
2• IMd., p. 5»
ima., 88 n.
flf.
a
» rtn ril
Op. cit., r>
p. Al ft
416
"
THE BORATIO 223
"good spirit, fetish." There can equally be no doubt

that in maboya we have the African bori, just as cemi
is the Bambara dyine, nyena, from the Arabic înn, but
influenced by the Arabic sama' "heaven," even as we
have observed such confusion in Asante and elsewhere.^
This confusion may already be observed in the north,
for in the Guanche of Teneriff we have aiano "heaven,
from the Tuareg aênna, etc., by the side of aSaman
"God," from the Arabic sama',^ even as the Tuaregs
use agenna? for "rain" where the other languages have
similarly derivatives from Arabic sama'.
We
have already come across the fetish man under
thename of Caraib. We shall now turn to the fetish
man connected with the maboya, that is, the African
bori. The "masterof the bori" is mentioned by
Ramon Pane* as bohuti, buhuitihu, for which Peter
Martyr uses boviti} Oviedo tells a great deal about
the bor alios of Venezuela: "They fear very much the
devil, of whom the boratios affirm that they see him and
talk with him frequently. They paint his face in their
jewels and work it in relief upon wood and upon every-
thing they esteem most. These boratios are, as it were,
their priests, and in every chief city there is a boratio to
whom all have recourse to ask for things to happen,
and whether it will rain, or whether the year will be
dry or abundant, or whether they are to go to war
against their enemies or give it up, and whether the
Christians are good people or will kill them, and the
boratio says that he will give his answer after having
consulted with the devil, and for this conversation and
consultation they lock themselves up in a room alone,
and here they make certain smokes (ahumadas) which
1 See p. 174 ff.
2 E. Laoust, Mots et choses berUres, Paris 1920, p. 187.
" Ibid., p. 188.
* Vol. I, p. 80 ff.
s
RaccoUa, part III, vol. II, p. 52 ff.
they call tahacos, with such herbs as bereave them of
their senses, and here the boratio remains a day, or two,
or three, and sometimes longer, and, after coming out,
he says that the devil has told him so and so, answering
the questions put to him, according to the desires of
those whom he wishes to satisfy and for this they give
;
the boratio some gold trinket or other things. For less

important matters the Indians have another way.
There is in this country an herb called tabaco, which is a
kind of plant as high as a man 's breast, and more or less
branching, which puts forth leaves a palm, in length and
four fingers in width, and of the shape of a lance iron,
and they are hairy. And they sow this herb, and the seed
which it makes they keep for the next year's planting,
and they watch it carefully for the following purpose:
When they reap it, they put the leaves in bunches and
dry it in the smoke in bunches, and they keep it, and it
is a much appreciated article of commerce among the
Indians. In our Hispaniola there is much of it in the
ranches, and the Negroes whom we employ value it
highly for the effect which it produces by smoking it
until they fall down hke dead, and thus they are the
greater part of the night, and they say that they do not
feel the fatigue of the previous day.
"To return to the Indians of Venezuela, to see

whether they should travel or go fishing or sowing, or
whether they would be successful in the chase, or
whether a certain woman loves them, each one is a
boratio, for, wrapping its leaves around an ear of corn,
they light one end a little and put the burning part in
their mouth and breathe it out, and when it is half
burned, they crush what is wrapped around, and if the
part of the tobacco burnt is in the form of a sickle, it is a
sign that what they wish to know will succeed and if it
;
has burnt straight, it is a sign of the opposite, and that

which was to be good is bad. And they believe this so
THE BORATIO 225
mucli that nobody and no reason can make them believe

otherwise, nor that the tabaco is a joke or vanity: on
the contrary they feel badly for being reprimanded by
those who want to undeceive them.
"The hor alios, besides doing as described above, act
in the villages as doctors and cm-e in the following
manner. When one is sick and cannot rise from the
hammock, they call the boratio and ask him to cm-e the
sick man and say that they will pay him, and he says that
he will be glad to come. Arriving where the patient is,
he asks what hurts him, and the sick man tells him.
He also asks him whether he would like him to cure
him, and he answers, Yes.
'
' He again asks whether
he knows that he can cure him since he is a very good
boratio, and the patient says that he does. If to any of
these questions the sick man says, No, the boratio goes
'
'
away and will not cure him. But if he answers in the

affirmative, the first thing the boratio does is to order
everybody in the house to fast and to eat nothing but
their maize gruel, which they call caga, and not more
than once a day. Then he turns to the patient and
asks him what pains him most, and, when he says that
the head or some other part of the body hurts him, the
boratio, while opening and closing his hands and putting
them over him, as one who wants to bring something in,
says that he is taking in the soul, and then he closes
his fist and blows upon it with his mouth, and says,
'Be gone, sickness!' And saying this the boratio utters
such shouts and yells over the sick man that he grows
hoarse and cannot cry or talk, and this lasts two hours
or more. After having done so, he asks him whether
it pains him as much as before, and if the patient says
that it does, he sucks the member or spot with his mouth,
spitting out from time to time. If at the end of five or
six days of such operations, the patient tells the boratio
that he is better, he puts a thorn or stone or whatever
'

he wishes in Ms mouth, so as to make the sick man
beheve what he wishes, without anybody's noticing it,
and, after having sucked where the pain was, spits into
his hand the thorn or stone or stick which he has taken
from his mouth, and shows it to the sick man, saying to
him, 'See what has been killing you and caused your
trouble!' Then he takes his leave and says that he
wants to go, and they pay him. If, by chance, the
patient does not say that he feels better from what
the boratio has done, the boratio, instead of spitting out
the stone or whatever he means to give as the cause of
the sickness, as frequently happens from necessity, for
what the boratio does is mere nonsense, he answers,
'I am going away, for you will not recover so soon from
your sickness as you think, for the devil has told me so,
and he says good-bye and lea<ves. "^
The resemblance to the Mandingo boritigi, bolitigi,
etc. is so close, including the very name, that nothing
needs to be added. We only need to pursue the varia-
tions of the name throughout the two Americas.
There is recorded Tupi-Guarani paje, Caraib boye,
piudai, Chayma piache, Bakairi piaze, Tamanaco
ptcMachi, Aparais puiacie, Galibi piaye "doctor,"
Baniba pinata, Yabitero epinatzi "medicine." Her-
rera mentions in Cuba the doctor as behique. Las Casas
calls him bohique. The most interesting forms to us
are Nahuatl pati "doctor," patli "medicine," Kechua
'hampi "medicine." The distribution of the African
"medicine" word over an enormous territory bears
witness to the tremendous cultural influence of the
Mandingos in America. No wonder, then, that
mandinga itself should in the Spanish of South America
have received the meaning of "witchcraft."^
1 Oviedo, op. ciU, vol. II, p. 298 f.
^ D. Granada, Vocahulario rioplatense razonado, Montevideo 1890, p. 269.

THE BORATIO 227
Tobacco forms an integral part of the boratio's ritual,

wherever it occurs, but nowhere is it referred to as being
smoked. It is blown out of the pipe or roll, in order
that the incense should reach the fetish. Tobacco,
like cotton itself, was imported for ritualistic purposes,
and smoking of tobacco, as we have seen, was of a slow
growth, due chiefly to the habits of the Negroes after
the discovery. There is no evidence whatsoever that
it was used before the discovery for any other than
ritualistic and, possibly, medicinal purposes.
"
CHAPTER XII.
The Mandingo Elements in the Mexican Civiliza-

tion.
Oviedo says that in the province of Cueva, in Castilla

del Oro, tequina was the equivalent of "master" in
whatever art.^ This word is, of course, Mande tigi
"master." Its most important development is found
in Mexico. Here we have tecutli "chief, knight,"
hence tecutilia "to ennoble," tecuuia "to act as a chief,
drill the soldiers," tecutocaitl "name of nobility,"
tecuti"to become a nobleman," tecunenenque "chief
merchants who traveled far and whom the monarchs
of Tenochtitlan considered as lords," tecuacan "chief
city, court, residence of sovereigns," teca "of some-
body," tecpan "lord's, king's palace."
The last word has entered into the Maya languages,
where we have Kiche tecpan "communal house, palace,
hence tecpanir "to increase the tribute," hence tec "to
increase, heap one thing upon another," tequeba "to
put in layers." These at once explain another series
of Nahuatl words: tecpanir explains tequitl "tribute,
impost, labor, functions, duty," tequiotl, tequiutl "ex-
ercise, labor, fatigue, servitude, contribution," tequio
"hard, dif&cult, " tequiti "to work, pay tribute,"
tequitlato "agent, distributer of tribute or work;"
1 "Deste nombre tequina se hage mucha diferenffia; porque & cualquiera
ques mas hSbil y experto en algun arte, assi como en ser mejor montero 6 pes-
cador, 6 hager mejor una red 6 una canoa H otra cosa, le llaman tequina, que
quiere degir lo mesmo que maestro: por manera que al ques maestro de las
responsiones 6 inteligengias con el diablo, lldmanle tequina en aquel arte,
porque aqueste tal es el que administra sus ydolatrlas e gerimonias e sacri-
figios, y el que habla con el diablo, segund ellos dijen, 6 ^ €1 di sus respuestas,"
Oviedo, op. cit, vol. Ill, p. 127.
"
THE MANDINGO ELEMENTS IN MEXICO 229
similarly Kiche tec explains Nahuatl te, tequi "much,"

teca "to put layers of stone or wood, to lie down in bed,
hence moteca "they unite," nic-teca "to plant." This
is still better brought out in the compound tecpana
"to put in order, arrange in layers," tecpanti "to
distribute."
Tec, as we see, represents a word designating "the
master, master mechanic. " There is hardly a language
in the neighborhood of Mexico that does not bear wit-
ness to the enormous influence wielded by those who
introduced the word together with the arts. We have
Zoque tec "house, room," tectzecpa "to build," Pokon-
chi tzakal "builder, maker," hence tzakal aj "maker of
mats," etc., tzakol "mason," tzak "worth," and Taras-
can tecari "carpenter."
Sahagun^ devotes the whole of Book IX to the
Mexican merchants and workers in gold, precious stones,
and rich feathers. The merchants, pochteca, accord-
ing to a tradition, at first appeared at Tlatelulco in the
reign of Quaquapitzauac. There were only two of
them, and they trafficked in green, blue, and red feathers
of the quetzalli. Later turquoises and green stones and
garments of cotton were added, for, heretofore, only
nequen cloth was used. This puts the origin of the
merchants and of cotton cloth in the first quarter of the
XV. century. Still later, rings, gold nuggets, »eut
stones, skins, and other articles were added to the trade.
The word pochteca is obviously a compound, but poch
cannot be explained from any Nahuatl word. In Maya
a merchant was known as p'olom, which apparently is
related to a Maya root p'ol "to swell," not recorded in
the other Maya languages. P'olomkay is given as a
forbidden song and dance, no doubt of the kind described
by Sahagun and connected with human sacrifices. The
' B. de Sahagun, Histoire ginerale des choses de la Nouvelle-Espagne, Paris
1880, p. 547 ff.
other Maya languages do not seem to record any
"merchant" word from this root except Tzotzil polman
"to buy." The Maya p'ol and Nahuatl poch are un-
questionably related and point to a form bor or for as
their root.
The Arabic J^' ij fl-al-mdl "rich" has undergone
some violent changes in Africa. While it is found in the
Arabic Wadai as fl-al-mdl, it has deteriorated to alman
"wealth" in Songay, while in the Berber languages it
has become hvêl-mal "possessor of wealth, rich,"
which has reentered the Arabic Soa as abuhumdm
"rich." This Berber hu-el-mal became, in its turn,
Wolof horom alale "rich," and has broken into borome
"chief, lord," mbor "rich," and alale "wealth."
In Soninke we have the forms fogome and naburugume
"rich," which lead to Malinke nafulu, Bambara
nafolo "wealth," Soninke nabure "merchandise,"
Soso nafuli "money." We have also the simpler
Bambara fuale, fale, Malinke falln, Mandingo faling,
Wolof wakhale, Songay bora "to exchange, traffic,
barter," Malinke fi,ri "to sell," firila "merchant,"
Hausa falala "a rich man." The Soa Arabic
abuhumdm has not only produced Soninke fogome, but
is found all along the western coast in much reduced
forms. We have Malinke, Bambara /ama "rich man,
king," Duala mbuan, Bayon mfon, ebon, Mimboma
mfumu, Musentandu mpfudma, Kongo vwama, Kiriman
ufuma, etc. "rich. " Thus we arrive at an ancient form
folom, forom, represented in Soninke fogome, naburu-
gume, which is responsible for Maya p 'olom "merchant"
and Aztec poch in pochteca. If Sahagun's chronology
is at all trustworthy, the Mandingo merchants first
appeared in Mexico in the beginning of the XV. century.
In Sahagun's account of the Mexican merchants we
are constantly reminded that they sold mantles {tilma,
chimalli), waistcloths (maxtli), and chemises {uipilli).
"
It can be shown that all these characteristically Mexi-

can garments are of African origin. Unfortunately
the scanty African vocabularies do not give us any ac-
count of the ancient garments/ and we have to draw our
conclusions chiefly from the Arabic names, which fortu-
nately give us all the data we need. In Molina's
dictionary tilmatli is translated by "manta, " and
chimalU is explained as "shield, buckler," while "man-
ta para combatir" is translated by vapalchimalli or
quauhchimalli. "Manta para combatir" would be a
"battle cloak" rather than a "buckler," and a study
of other Indian languages proves conclusively that there
is a direct relation between "buckler" and "mantle."
In Maya we have chim, chimil " bag, pouch, bird-crop,
chimal "shield," but in Huasteca we have the Nahuatl
quauhchimalli as cuachim, cuachimal "any kind of
garment," hence cuachimzal "to dress." That the
Huasteca has preserved the original meaning follows
from the Arabic, whence the word and the thing were
taken, through the Mandingo. The Arabic cM
Samlah, Simlah, plural Hmal, is "a garment in which
one wraps himself" and J^ iimal is "a sort of bag put
to the raceme of a palm tree, in order that the fruit may
not be shaken off, or to the udder of the ewe or goat,
when the udder is heavy with milk."
In Africa it has generally preserved the meaning of
"wrapper, girdle," as in Berber cemla "girdle, turban."
In Biblical Hebrew we have ^^^V Hmelah "cloak, in
1 "H rfegne une grande confusion dans la terminologie berbire relative
aux vStements. A cSt6 de representants, que I'ancien berbere appliquait h

des parties du vltement dont I'usage a disparu depuis longtemps, existent
des appellations locales, fort nombreuses, et ces expressions 6trangeres,
arabes le plus souvent, d&ignant, parfois m6me, des vStements d'origine
africaine. Le problfeme du vStement se complique en consequence; la
linguistique, en tout cas, ne peut fournir sur la question que des donnfies in-
suffisantes," Laoust, op. cit., p. 123 f.
which one wrapped himself at night," and also ^^/V
ialemah "garment." This latter transposed form is
recorded in Songay tilbi, derbe "garment," which pre-
supposes a form tilme, and this is still further proved
by Soso domma "shirt, blouse, cloak, large wrapping-
cloth. The Soninke irame "shoulder-cloth" is de-
rived from the same, through a form tirme, as in Songay
tilbi, derbe. We find similar forms to the Soso in
various Mande languages, Kono duma, Tene rimo,
Gbandi ndomai, Mende ndoma, Vei doma, duma "an
upper garment of males, of the form of a shirt, without
sleeves and collar, but generally provided with a breast
pocket. " Outside of the Mande languages it is found
in Landoma duma, plural seduma, Baga duma, plural
suma, Temne ruma, plural suma, Bulom lumo, plural
îlumo, and this form shows how all the others are de-
rived from the Arabic Hlmah or §imlah by apocopation.
The latter is also preserved in the Arabic of Adirar as
tasmlr. In a series of languages Songay derbe has be-
come further corrupted to derge, as in Mandingo dur-
ukl, Bambara doloki, dloki, Mandingo dondiko, Soninke
doroke, Gura delegoa, Peul dolokie, dialectically togore,
Dsarawa lugod, Koro loga, Hausa riga "shirt," and this
has become still more corrupted in Bornu kalugu,
kaluru.
Equally interesting is Nahuatl maxtli "waistcloth
to hide the nudity." The early writers on America,
Columbus and Cortes included, have used the Arabico-
Spanish word almaizares with which to designate this
piece of adornment, frequently the only one observed on
women. B ut this Arabic word got to Mexico throu gh
J
the M
andingos before Columbus, f Herodotus tells of the
Arabs as we&rmg^£cpou7 This is the Arabic -''J' 'izdr
"a garment which covers the lower part of the body,
1 VII. 69.
AFRICAN ALMAIZAR, from Freeman's Travels and Life
in Ashanti and Jaman.
from the waist to the thigh, also a woman's veil,"
hence Jj^ ml'zar, mai'zar "a loincloth for girls,
mantle, wrapping-cloth," etc. Both forms are found

in the Mande languages. We have Mandingo sitti
"to tie," Malinke siri, sitti "to tie, attach," Bambara
siri to tie, make a knot," and Malinke mas-
"a bundle,
iti "adornment, jewel, to tie," Bambara masiri "adorn-
ment, to make one's toilet," masirili "ornamentation,
toilet." The derivation of Nahuatl maxtli from this
is obvious.
more interesting is the history of Nahuatl vipilK.
Still
Arabic J^ gafr means "to cover, conceal, hide, to cover

one 's crime, forgive, pardon, " hence J^ gafir "covering
and forgiving the sins, " "J^ gifarah "skullcap, mantle,"
'^ guflrah "small mantle." This is found in the
African languages as the denomination of the waist-
cloth or drawers generally worn by the uncircumcised
boys, to cover the nudity. We have Fulup gabil, plural
obil, Guresa galpali, Bagbalan garpal, Soso kufura
" waistcloth, " and in the latter case the meaning "to
forgive" is preserved in the form khabari, while in
Zenaga we have r'afar "to pardon," r'ufara "a cap."
Similarly we have Arabic (in Adirar) ofara, plural
gofafer, (in Beran) r' ofara, gofara, Soninke kufune,
Peul hufune, Mandingo fula, Asante flla, Aku efila,
Yoruba fila, Mossi fowila, Gurma foalera, Nupe fula,
Timbuktu /tiZa " cap " In the other Mande languages
.
gabil, obviously through a form wabil, has still further

been reduced, and we have Malinke bila, Bambara
bila, bla "waistcloth for uncircumcised children, to
permit, leave alone, put to one's account, put away,"
while in Vei we have bere "a strip of cloth about two
inches broad worn by girls from about their eighth year
up to their marriage, to cover their shame, and hanging
down before and behind to about half a foot from the
ground." It is, therefore, identical with a mark of
and beremo means "a virgin."
virginity, Thus we get
to Nahuatl vipilli "the nether garment of an Indian
I
woman" and "protective garment used in war." In
the Maya language we have this Mexican vipilli in the
form cuyub, no doubt from cuyubil, just as chim has
developed from chimil, chimal.
The early writers, Encizo, Las Casas, Oviedo, Bernal
Diaz^ use naguas as a native name for a woman 's loin-
cloth, which fell from the waist to the middle of the
thigh. This word is from Arabic r^ laâm "bridle,

strap by which a horse is led, menstrual cloth. " This
we find in Berber, Hassania lejam "bridle," Wolof
lakhabe "leather strap by which a horse is led, " Malinke,
Dyula lagba " v§tement in time de femme, " and this pro-
duced Spanish naguas. The change of lagba to nagba
must have already taken place in Mande, for we have
side by side Malinke lamaha, namaha "to shake, " latege,
natege "to cut, " etc., and, as la, na are common prefixes,
they have a tendency to drop off even from words that
are not compounded. Indeed, we have already in
Zenaga and Tuareg ar'ba "bridle," which presupposes
a form agba, and Malinke karafe, Bambara karbe,
karabe, karfe "a bit, bridle" show that agbe led to
karbe, because this guttural Arabic g introduced the let-
ter r after the k. Thus there arose in Mexico the form
cueitl for the island nagua, meaning "a woman's nether
garment." jrhus.JEaJLaY.ei.dentified_^ll_the g^^^
handled_by the early..êxican._ me rchan ts as of Man-
dihgo_orig3^n.
In Nahuatl we have both tilmatli and chimalli, but
the latter is recorded only with the meaning "shield."
It would, therefore, appear that the round object with
' A. Zayas y Alfonso, Lexicografia antillana, Habana 1914, p. 396 f.

JDatte^janjerretter mS âno.
SUDANIC ARMOR, from Frobenius' Th Foice of Africa.

"
a cloth fringe at the bottom, so often represented in

Mexican manuscripts, is and has always been a shield.
But there are a number of disconcerting facts in such an
assumption. In the first place, in the vast majority of
representations of warriors and tributes in Mexican
manuscripts, the garments are the vipilli, and maxtli,
and the chimalli shield, whereas in the vast majority of
cases of the representations of men wearing the tilmatli,
the shield is absent, but the "devisa" is woven or
painted on the mantle itself. It, therefore, appears
that the mantle is, indeed, a quauhchimalli, a "large"
chimalli, whatever the chimalli may have been. In
any case, the mantle and the shield were known in
Nahuatl as tlauiztli, which means "arms, insignia,"
and this shows at once that we are dealing here with
defensive armor, which at the same time had the dis-
tinctive mark, the n'tene of the Mandingos.
We have Arabic w^ labisu "to dress, accoutre,"

o^j^ luhus "cuirass, a kind of linen covering filled with
cotton, which in battle covers the back, flanks, neck
and chest of the horse, and which, they say, is impene-
trable to the lance or sword," u-^y labis "one covered
with a cuirass," tr"^ labus "coat of mail," trr^" talhls
"dressing, investitvire.
In the Mande languages we have traces of this ancient
Arabic defensive armor. We have Malinke labiti,
mabiti "to cover oneself," and, as usual, assuming la,
ma to be a prefix, we get biti "to cover," hence bitlnka
"a cover." Similarly we have Bambara biri, bri,
Mandingo bitta "to cover." In Hausa we get lufudi
"coat of mail put on horses and men, made of cloth
stuffed with cotton, wool, etc., not of iron." In the
Berber language we naturally have the Arabic word
well preserved. Here we have lebsa "garment."
This gets into Songay as dabiri "a cover," dabu "to
dress." All these prove that the Arabic Idbis "one
covered with a cuirass," etc. had an enormous effect
upon the development of protective armor in the West-
ern Sudan. Hence it follows that the African n'tene,
from Arabic niSan, "the mark of distinction," passed
among the Mandingos, no doubt together with the
formation of the Malli state under Arabic influence.
According to Yacoub Artin Pacha, ^ blazonry among
the Arabs had its origin in the activity of the Persian
poet, Firdusi, in his Shah-Nameh, in the beginning of the
XI. century, since he roused with it the historic sense
and a love for heraldry. "In a multitude of verses
he, indeed, makes mention of personal marks of dis-
tinction and of colors, which form true coats of arms,
blazoning the warriors, heroes, kings and nations,
whose great deeds he sings. These coats of arms make
them known at feasts, tourneys, military reviews, and
even on the battlefield. The sovereigns and their mam-
elucks, reading the Shah-Nameh, naturally felt the de-
sire to imitate in everything the costumes and usages
of the kings and heroes, whose valiant deeds were sung
and glorified in this epic. Therefore, they adopted
for themselves, for their great dignitaries and knights,
and even for their armies, symbolical colors and
graphic symbols, like those of the heroes and kings of
the Shah-Nameh. Thus, from the XI. century on, the
mamelucks of Syria and of Egypt, following in this the
example set them by the oriental princes, also adopted
the coats of arms, which they designed on their bucklers,
their pennants, their standards, even on the garments
of their slaves, and sometimes on coins, and monuments
which they constructed."^ As there is no evidence of
the development of heraldry in western Europe before
' Op. cit., p. 8 ff.
2/6id., p. llf.
ARABIC BLAZONRY, from Artin Pacha's Contribution a rhudc du hiason
en Orient.
274 2/5
2jr6 277
It CROISSANT ?.7i- S92 -S96 398
ARABIC BLA70NRY, from Artin Pacha's Contribution a I'kude du blason

en Orient.
Abb. 57. valiente tJac^tiecf.atl. Abb. .58. valiente tUir<nrhf.ah(rtl.
Codci ilenrtoza 68, 20. Codes Moidoaa (,>*, il.
.^bb, 59. valiente uff^tiatutfL Abb. 60. valiente iiôei/uwteatL

Codei MendOM 68, 22. Codei MendMa CS, as.
MEXICAN BLAZONRY, from Seler's Gesammelte Abhandlungen.

the XII. century,^ it follows that the European heraldry

had its origin in the same Persian source. Of course,
banners and special insignia had been in use in the arm-
ies of Rome, Greece, Egypt, etc., but the perpetuation
of personal "devices" on bucklers and garments, so as
to form a heraldic science, is something of which we
have no earlier record.
"In the Orient, the whole heraldic science seems to
have consisted in the emblazoning of knights with coats
of arms which, so to speak, spoke. The design, in fact
indicated in the greatest number of cases the functions
of the knights at the court or in the camps of the sultans
whom they served and whose mamelucks they were.
The coats of arms were generally personal and attached
to the very functions of the knight. When the knight
changed his functions, his coat of arms changed also or
was surcharged with some new design which indicated
his new functions. They were nearly always designed
on the banners and round bucklers which the oriental
knights carried, bucklers that resembled the Roman
clipeus. Also, if they wished to reproduce the coat of
arms of a knight upon monuments, furniture, household
utensils, cloth, arms, etc., which belonged to him, they
represented it as inscribed in a round shield, represent-
"^
ing the round buckler then in use.
The description given above fits the Mexican method
down to the minutest details. Artin goes on to describe
the coats of arms representing the lion, the eagle, the
fish, and a variety of other geometric designs, but the
most interesting to us is the representation of the cres-
cent, because it is by far the most characteristic Moslem
emblem. It is generally represented by one upward
sign, but frequently it has three stars connected with it,
or the crescent is repeated two or three times.^ In the
> Ibid., p. 15 f.
2 Ibid., p. 43.
3 Ibid., p. 165.
great majority of Mexican shields we have the crescent
accompanied at the bottom by three more crescents.
Unfortunately we have no representation of the Malli
coats of arms preserved, except the n'tene, which pictures
animals or some objects and vaguely stands for a
clan or brotherhood, but these are no longer represented
on standards or shields. But the name n'tene, as I
have shown, represents a sign of nobility, and we have
found the word preserved in the same sense in the
Carribean Islands. Thus there cannot be the slightest
doubt that the Mexican coats of arms are evolutions of
the Mandingo n'tene in its original signification.
The Mexicans represented certain dignities or func-
tions on the banners called panitl, pantli, or pamitl.
We have also Maya pan "banner," while in Kiche pan
means "leather or cloth cover." This latter sense
shows that the original meaning of panitl was "a
square piece of cloth." Encizo in his Suma de
geografia says that the women put on a pampaniya, a
leaf tied with a cotton string, if they did not have naguas,
and Oviedo says that a pampaniya is a piece of cloth of
the size of a hand.^ These meanings are nearer to the
Kiche than to the Nahuatl meaning. We find this
word in Soninke fendeli, Wolof mpendel "a woman's
loincloth, petticoat." We have it in Songay bankarey
"cloth," Hausa, Nupa hante "towel, apron, cloth,"
Mandingo fano, Malinke fanu, Bambara fini "square
piece of cloth," the French "pagne," which is not, as gen-
erally supposed, from Spanish " pano " but from Persian-
Arabic •^. band "a knot, tie, ribbon, wrapper," \j^-^.

banddrlyah "drapery, curtain." But the Arabic band
very nearly acquired the meaning "banner," from the
Greek ^dvdov, from the same Persian word, hence it un-
questionably had the same meaning in the African
' Zayas y Alfonso, op. cit., p. 412.

languages, where the word "banner" is not mentioned

in the scanty vocabularies. The Mexican pantli is of
the same Arabic origin. In Molina only, the compound
is recorded, quachpantli, where quachtli means "a large
mantle, cloth," so that the relation of the two is in
Mexico the same as in the Arabic.
This Nahuatl quachtli is the Malinke, Bambara,
Songay kasa, Hassania Arabic kasi "coverlet." It is
the Arabic '^ which is also found in Spanish
kisa,''^
alquicel, alquicer "Moorish garment in the form of a

mantle, table-cloth." Covarrubias says that "it
is a covering for a bench or table, and is woven without
a seam, like a bed-cover." Leo Africanus, speaking
of the Berber dress, says that "their ordinary garment
is an alquicel. It resembles a linen bed-cover in
which one wraps oneself, but it is finer and it is used to
wrap around the body." Cadamosto says that the
Zenaga wear white cloaks which they call alchezeli.
That this is identical with the Nahuatl quachtli is seen
from the fact that the word is used in compounds, where
it means "bed-cover, veil." Thus we get quachicpalli
"pillow," quachpepechtli "mattress, pillow, coverlet,"
quachpanyo "that which has sails." There was also
a small mantle called patolli coachtli, which the
Spaniards called patoles coacheles,^ and here the Spanish
form alquizeles is even more prominent.
While pochtecatl is the usual word for "merchant,"
a more important designation is tiamictli, tianquiztli,
for which there is no Nahuatl etymology. Fray
Toribio de Motohnia calls the Mexican market tiantiztli
or tianquizco.^ It is not likely that tiantiztli is a
blunder for tianquiztli, since it is once more repeated as
^ Dozy, Dictionnaire d&tailU des noms des vitements, p. 383 flf.
2 J. Garcfa Icazbalceta, Memoriales de Fray Toribio de Motolinia, Mljico
1903, p. 330.
> Ibid., p. 321 ff.
tiantiztli,ând since we find it in Aymara as tinta. But
all the words of this group are derived from tangoman,
tangosmaos "the white Negroized trader on the west
coast of Africa."^ Indeed, one of the chief places of
sacrifice was at Tianquizmanalco, and that this orig-
inally meant "marketplace" follows from tianquizman-
aloyan "marketplace, fair." Tianquizman or some
such form for "merchant" preceded the shorter
tianquiztli, which is obviously not a Nahuatl word.
In Central America tianguiz, tiangue is still used col-
loquially for "marketplace, fair."
MoHna^ records toltecatl, plural tolteca, "master
mechanic," hence toltecauia "to produce a work of
mechanical perfection." Since teca is "master," it
follows that tol must mean "mechanic, artisan." The
historical myth which makes of the Toltecs an ancient
race from whom the Aztecs received their culture is not
worth considering seriously, as has already been pointed
out by Brinton. It can easily be shown that the name
arose only after the Mandingos had influenced the
Mexican civilization. I have already pointed out the
development Arabic j:^ dabr in Africa.^ I shall

of
now trace its more important developments. In the
Koran we frequently find the verb dabbar in the sense
of "he devised, planned, or plotted a thing, he managed,
conducted, ordered, or regulated an affair, he acted with
consideration of the issues or results of affairs."
Hence we get Berber debber "to conduct, manage,
influence" and amdebber "counselor, guide, director,"
Hausa dabara "contrivance, skill, counsel," Songay
dabari "counsel, means." In Wolof we get defara
"to make, establish, restore," and the abbreviated
' Ibid., p. 326.

2 See vol. II, p.
112 fi.
'A. de Molina, Vocabulario de la lengua mexicana, Leipzig 1880.
*Vol. I, p. 109 f.
defd "to make, place, construct, create, form, produce,

manufacture, compose." In the Mande languages we
get Soso rafalla "to arrange, make," Bambara dabali
"means, expedient, industry," and the frightfully
apocopated Bambara dala, dla "to arrange, put in order,
build," da "to do, manufacture, put, lay down, add,"
Daba "God," Malinke dala "manufacture, maker,"
da "to make, manufacture, compose, arrange." In
Bambara or Malinke "master mechanic" would be
daltigi, and this is unquestionably the origin of Nahuatl
tolteca "master mechanic." In the Mande languages
this would refer to the Arabic conquerors who brought
the arts to the Malli kingdom, and in the Mexican myth
we have some kind of recollection of this historic fact.
In Nahuatl we have a parallel to toUecatl, the term
patecatl, for "the god of the pulque," which was con-
sidered a medicine, from pai "to drink a medicine,"
paitia "to take a purgative," patli "medicine, plaster,
unguent," which are all related to the bori words
already discussed, hence patecatl is the Mandingo
boritigi. We have another fetish man word, which is
'
'
'
'
far more interesting. In Caraib there is recorded amaoti

"retired, sedentary man," while in Bakairi, in the in-
terior of Brazil, we have omeoto "fetish man." In
Aymara amaotta means "very wise," while among the
Peruvians the amauta was the teacher of the youth, who
instructed them in the wisdom of government and of
religion and was in charge of the quipu records. In
Mexico Molina's dictionary gives for amantecatl the
meaning "master mechanic," as in the case of toltecatl.
Sahagun has the following account of the amanteca:
"According to the accounts of the ancients, in regard to
the name of the amanteca, the first inhabitants of the
country brought with them from the regions whence
they came a god called Coyotlinauatl, whom they never
stopped worshipping. These immigrants were called
— —
icnonitlacapixoani mexiti, which, means 'the first colon-
ists who are called Mexiti, ' whence came the name of
Mexico. The people who settled in this region
multiplied, and their descendants put up a sculptured
wooden statue and erected a temple in a quarter to
which they gave the name of Amantlan. Here they
worshipped the god whom they called Coyotlinauatl,
and made offerings to him, and, on account of the
name of the quarter, which is Amantlan, these in-
habitants assumed the name of Amanteca. On holi-
days they clothed this god in a dressed coyote skin.
The privilege of preparing this sMn was left to the in-
habitants of this quarter of Amantlan. He kept the
coyote head, but was covered by a human mask. His
canine teeth were of gold and all his other teeth were
very long and pointed. In his hand he carried a stick
which was adorned with black itztli stones and on which
he leaned, and he had a buckler made of reeds, on the
border of which was painted a light blue circle. On
his back he carried a pot from which a large number of
quetzalli feathers issued. On his ankles he had a kind
of gaiters with a large number of small white shells
like rattles. He wore cotaras of the leaves of a tree
called yecotl, because this was the footgear that the
immigrants wore when they came to this country.
They always put them on his legs, in order to let it be
known that they were the first Chichimec colonists who
established themselves in this country of Mexico.
"In this quarter of Amantlan they worshipped not
only this god, but also seven other idols, of whom five
were dressed as men and two as women. But Coyotli-
nauatl was none the less the chief god of all. The
second who came after him was Tigaua, the third
Macuilocelotl, the fourth Macuiltochtli; the two women
came in the fifth rank. One of them was called
Xiuhtlati, the oth.er Xilo. The seventli who was

called Tepuztecatl was placed opposite them all.
"This was the way of adorning all these gods: the
male gods wore the costume of Coyotlinauatl, except the
god called Tigaua, who was not dressed in a coyote
skin. But he did have on his back the pot with the
quetzalli feathers, and he carried the stick and buckler,
and wore the white cotaras. The god called Macuil-
ocelotl was dressed in a coyote skin. His head was
covered with it, and he looked out from the jaws of an
animal. He, too, carried on his back the pot with his
quetzalli, and he had the stick, the buckler, and the
white cotaras. The god Macuiltochtli was dressed
up in the same manner. As to the women, the
one who was called Xiuhtlati was dressed in a blue
vipilli; the other, called Xilo, who was the second, was
dressed in a red vipilli dyed with cochineal. Both had
their garments profusely adorned with rich feathers
from all kinds of birds of beautiful plumage; the bor-
ders were formed by beautiful feathers, as we have al-
ready said. The goddesses carried in one hand green
ears of maize in the form of a stick, in the other a fan of
rich plumes and a gold jewel in the shape of a comal.
They had well-poHshed and brilliant gold earrings;
they carried nothing on their backs, and paper bands
took the place of their hair-dress. Their wrists were
adorned with all kinds of rich feathers, and their legs
were similarly garbed from the knees to the ankles.
They, too, wore cotaras of yecotl leaves, in order to show
that they belonged to the Chichimecs, who had settled
this country.
"Twice a year they celebrated these divinities with
a festival, one in the month panquetzaliztli, and the
other in the month called tlazochimaco. In the month
panquetzaliztli they killed the image of Coyotlinauatl.
If no one came forward to offer a slave, called tlaaltiltin,
244 AFEICA AND THE DISCOVERY OF AMERICA
these amanteca united and bought one, in order to kill
him in honor of the God, in exchange for mantles called
quachtli, which were used to pay the tribute with.
But if some amantecatl celebrated for himself a feast and
killed some slaves, one of these was sacrificed in honor of
the god CoyotUnauatl. He was covered with all the
ornaments of the god, as was told before. If the one
who celebrated the feast was a rich man, he killed one,
two, or three slaves, and even more, always in honor of
these divinities. If he was not well-to-do, he was
satisfied with one single victim, in order to honor
CoyotUnauatl. When they celebrated the holiday, all
the old men, men, and women, gathered in the quarter
called Amantlan. There they sang, and compelled all
those who were to be sacrificed to keep awake. They
generally made them take a draught called itzpachtli, so
that they would not fear death. This drink intoxi-
cated them and made them lose consciousness. For
that reason they made use of it, so that they would not
be conscious when their breasts were opened. There
were some slaves who were mad enough to start running
toward the top of the temple, driven by the desire to be
killed forthwith, in order to make an end of life.
'
When the holiday of these divinities was celebrated
'
a second time, in the month of tlaxochimaco, no slaves

were sacrificed. The solemnity was then celebrated
in the name of the above mentioned goddesses, without,
however, forgetting the five other gods. All Amanteca
women then gathered in the Amantlan quarter and
dressed up Like the goddesses, in the manner explained,
whereas the men were satisfied to cover their legs with
red feathers. Then the amanteca offered up their
boys and girls to these divinities, promising to put the
boys in the calmecac, where they would learn the art of
a toltecayotl, while for the girls they invoked the god-
desses to help them become good workers and dyers of
tochomitl of every color, either with, feathers or rabbit

wool.
"The quarters of the amanteca and pochteca were
confused, and the same happened to their gods, of
whom one was called Yacatecutli, god of the merchants,
and the other CoyotUnauatl, who is the god of the
amanteca. For that reason the merchants and feather-
workers honored each other, and when they seated
themselves at a banquet, the merchants placed them-
selves at one side, and the feather-workers at the other.
They were about equal in their wealth and in the man-
ner of celebrating their holidays and banquets. This
was so because the merchants brought feathers from
distant countries, and the amanteca worked them and
manufactured from them coats of arms, such as the
bucklers of which the kings and high personages made
use, which were very numerous in form and bore various
denominations, as had been explained in the text. Be-
fore they became acquainted with the rich plumes,
from which they manufactured their coats of arms,
these tolteca made the dancing-gear from the white and
black feathers of chickens, herons, and ducks. They
did not then as yet know the finer points of their pro-
fession. They limited themselves to fixing the feathers
in a coarse way with itztli knives on aueuetl boards.
The rich feathers were known in the time of King
Auitzotl. They were brought by the tecunenenque
merchants, when they conquered the provinces of
Anahuac, as we have said. It was at that time that the
amanteca began to make fine and delicate pieces of
work."i
In this account we have the same confusion of the
mythical Amanteca with the amanteca artisans that we
had in the case of the Tolteca people and tolteca artisans.
The god of the amanteca is CoyotUnauatl, which Seler
1 Sahagun, op. eit, p. S87 ff. (book I, chaps. XVIII, XIX).
writes coy oil inaual, and says that the god was in the form
of a coyote.^ But as the latter word is derived from
naualli "magician," I shall show that this refers to an
African origin. Brinton very long ago recognized it as
not of native origin: "To illustrate this I shall sub-
join several series of words derived from the same radi-
cal which is at the basis of the word nagual, the series,
three in number, being taken from the three radically
diverse, though geographically contiguous, linguistic
stocks, the Maya, the Zapotec and the Nahuatl.
From the Maya, of Yucatan.
Naual, or nautal, a native dance, forbidden by the

missionaries.
Naatil, talent, skill, ability.
Naat, intelligence, wisdom.
Naatah, to understand, to divine.
Nanaol, to consider, to contemplate, to meditate, to
commune with oneself, to enter into oneself.
Noh, great, skillful; as noh ahceh, a skillful hunter.
From Maya Dialects.
Quiche-Cakchiquel.
Naual, a witch or sorcerer.
Naualin, to tell fortunes, to predict the future.
Qui naualin, to sacrifice, to offer sacrifices.
Na, to feel, to suspect, to divine, to think in one's heart.
Nao, to know, to be alert or expert in something.
Naol, a skillful person, a rhetorician.
Naotizan, to make another intelligent or astute.
Natal, the memory.
Natub, the soul or shadow of a man.
Noh, the god of reason.
Noh, to fecundate, to impregnate {Popol Vuh).
' E. Seler, Gesammelte Abhandlungen zur amerikanischen Sprach- und
Alterihumskunde, Berlin 1904, vol. II, p. 970.
'
Tzental.
X-qna, to know.
X-qnaulai, to know often or thoroughly (frequentative).
Naom, wise, astute {naom vinic, hombre sabio).
Naoghi, art, science.
Naoghibal, memory.
Ghnaoghel, a wise man.
Alaghom naom, the Goddess of "Wisdom.
From the Zapotec, of Oaxaca.
Nana, gana, gona, to know.

Nona, to know thoroughly, to retain in the memory.
Nana ticha, or nona lit, a wise man.
Guela nana, or guela nona, wisdom, knowledge.
Hue gona, or ro gona, a teacher, a master.
Na In, truth;ni na Hi, that which is true.
Nacina, or naciina, skill, dexterity.
Hui naa, a medicine man, a 'nagualist.
Nahaa, to speak pleasantly or agreeably.
Nayaa, or nayapi, to speak easily or fluently.
Rigoo gona, to sacrifice, to offer sacrifice.
Ni nana, the understanding, the intelligence, generally.
Nayanii, the superior reason of man.
Nayaa, 1 superiority, a superior man (gentileza, gen til
Naguii, ) hombre).
From the Nahuatl, of Mexico.
Nana, to dance, holding each other by the hands.

Naualli, a sorcerer, magician, enchanter.
Nauallotl, magic, enchantment, witchcraft.
Nauatl, or nahuatl, skillful, astute, smart; hence,
superior; appUed to language, clear, well-sounding,
whence (perhaps) the name of the tongue.
Nauati, to speak clearly and distinctly.
Nauatlato, an interpreter.
"I believe that no one can carefully examine these
listsof words, all taken from authorities well ac-
quainted with the several tongues, and writing when
they still retained their original purity, without ac-
knowledging that the same radical or syllable underlies
them all; and further, that from the primitive form
and rich development of this radical in the Zapotec, it
looks as if we must turn to it to recognize the origin of
all these expressions, both in the Nahuatl and the Maya
linguistic stocks.
"The root na, to know, is the primitive monosyllabic
stem to which we trace all of them. Nahual means
knowledge, especially mystic knowledge, the Gnosis, the
knowledge of the hidden and secret things of nature;
easily enough confounded in uncultivated minds with
sorcery and magic.
"It is very significant that neither the radical na
nor any of its derivatives are found in the Huasteca
dialect of the Mayan tongue, which was spoken about
Tampico, far removed from other members of the stock.
The inference is that in the southern dialects it was a
borrowed stem.
"Nor in the Nahuatl language —
although its very
name is —
derived from it does the radical na appear in
its simplicity and true significance. To the Nahuas,
also, it must have been a loan.
"It is true that de la Serna derives the Mexican
naualli, a sorcerer, from the verb nahualtia, to mask or
disguise oneself, because a naualli is one who masks or
'
disguises himself under the form of some lower animal,

which is his nagual;' but it is altogether likely that
nahualtia derived its meaning from the custom of the
medicine men to wear masks during their ceremonies.
"Therefore, if the term nagual, and many of its as-
sociates and derivatives, were at first borrowed from the
Zapotec language, a necessary corroUary of this con-
'
elusion is, that along with these terms came most of the
superstitions, rites, and beliefs to which they allude;
which thus became grafted on the general tendency to
such superstitions existing everywhere and at all times
in the human mind.
"Along with the names of the days and the hiero-
glyphs which mark them, and the complicated arith-
metical methods by means of which they were employed,
were carried most of the doctrines of the Nagualists,
and the name by which they in time became known
from central Mexico quite to Nicaragua and beyond.
"The mysterious words have now, indeed, lost much
of their ancient significance. In a recent dictionary of
the Spanish of Mexico nagual is defined as 'a witch; a
word used to frighten children and make them behave,
while in Nicaragua, where the former Nahuatl popula-
tion has left so many traces of its presence in the lan-
guage of to-day, the word nagual no longer means an
actor in the black art, or a knowledge of it, but his or her
armamentarium, or the box, jar or case in which are
kept the professional apparatus, the talismans and
charms, which constitute the stock in trade or outfit
of the necromancer.
"Among the Lacandons, of Mayan stock, who in-
habit the forests on the upper waters of the Usumacinta
river, at the present day the term naguate or nagutlat is
said to be applied to any one who is entitled to respect
'
and obedience by age and merit; but in all probability

'
he is also believed to possess superior and occult

knowledge."^
All these words are from the Arabic, of course,
through the Mandingo. We have Arabic *V naba' "to
proclaim, prophesy," i^' nabl "prophet," V nabah
"intelligent, penetrating, vigilant," and these two roots
1
D. G. Brinton, Nagualism, Philadelphia 1894, p. 56 flf.
have become in Africa welded into one, and have to a
great extent disappeared from the native vocabularies
because of the confusion with nama. And yet, we have
Peul nablu, Dyula nabiu, Soso annabi "prophet,"
Wolof nahlna "prophet, the legislator of a sect."
Among the Mossi naba means "chief, master," and
among the Habbes-Gara the masked young men, who
among the Malinke are known as nama, are called
naba. Here we have a confusion of two Arabic terms,
but the remarkable thing is that the men wearing the
masks among the Malinke are called namakoro, literally
' '
the hyena wise men, ^ that is, an exact translation of
' '
Nahuatl coyotlinauatl, where the American coyote is

substituted for the African hyena. The Nama in
Africa protects the people against the suba, or subaga,
the were-wolf, the hyena,^ and there is a male and a
female Nama, and sacrifices are made twice a year to
them. The resemblance of the Nama worship and
that of the Coyotlinauatl is striking. Even the word
is preserved, for naualli and naba, nama are identical.
The resemblance does not rest here. Let us look at

the celebration of the Nama among the Bambaras:
"The two heads of the male and female Nama seem to
represent two fantastic birds, and when they are pre-
ceded by the darotigi, who shakes the rattle and carries
a burning fagot, and the dyenfa mussu, who run through
the village and around the walls in search of the siri, the
illusion is complete. It is still more so when they put
them on their brows and mix with the dancers, who
shake their bodies which are naked as far as the waist,
and howl at the sound of drums and fifes their sacred
songs He is our host, the nya, great killer of men, he is
:
'
our host. It has blossomed in order to bear its fruits,

this boli has blossomed, it has its fruit. Suba, yo, this
Ji
' See p. 193 for koro "sense, wisdom."
' Henry, op. cit., p. 40.
p. l.Ml.
Fiitiche Natna,
From Henry's Les Bamhara.

was done for shame's sake. If one has been addicted

to the practice of the suha, we have a means of getting
away from it.' The sacrifice of the Noma
not less is
imposing. In the enclosiire where is the beehive that
serves as a tabernacle, only the members of the staff
enter. The mass of the brethren stay outside and
nobody approaches the palisade except when his tm"n
has come to present to the sacriflcer the chicken and the
kola nuts of the sacrifice. While the idol is being
smeared with blood, two men, naked to the waist and
facing the crowd, stand motionless on either side of the
sacrificer, holding in their hands two sticks a meter in
length, upon which three sheep-horns and two other
horns are tied. It is an honor to be allowed to hold
these horned sticks, it is even a dignity, and these
people are called the dyenfa tyeu. .During these
. .
nights the women stay inside the village, and the streets
that abut against the place where they meet are cut off
with kara or rough mats. They hear from time to
time near them the loud sounds of trumpets and horns,
and they see over the mats the outUne of a hideous mask
which they cannot distinguish and even are unable to
say to have seen. When the Nama comes out, the
women are called at day-break, just as the god is put
back in the beehive, and they, at the foot of the tree
where our devotees have been howling and dancing all
night, are shown the siri found by the dyenfa tyeu and
the darotigi or darotala. While these siri are burned,
the women dance and sing praises to the idol. Our
boli have two days of rest during the week, Monday and
Thursday. "1
I have already shown the relation Nama wor-
of the
ship to the Islamic practices. It now can be shown
that here we have, indeed, the aman "the faith," for
the singing of the prayer at day-break, to keep off the
1 Ibid., p. 149 3.
suba, is, of course, the Arabic prayer at day-break, the
C^ subh.The Nama is connected with the Arabic

seven-day week, two of which are used as rest days, that
is, unlucky days. The Namatigi, the priest of the
Nama, that is, here the naba "the prophet," is among
the Malinkes and Bambaras also called fur a tigiba "the
great master of medicine," fur a meaning originally
"leaf of a tree, " hence "medicine whichcomes from the
trees." This is brought out in the prayer where the
boli has blossomed its fruit, with which the suba "the
hyena, were- wolf " can be warded off. As the boli of
the Nama is the chief medicine, he is connected with
the bees, and so resides in a beehive. This is based on
the Koran, XVI. 70, 71: "Thy Lord spake by inspira-
tion unto the bee, saying. Provide thee houses in the
mountains and in the trees, and of those materials
wherewith men build hives for thee: then eat of every
kind of fruit, and walk in the beaten paths of thy Lord.
There proceedeth from their bellies a liquor of various
colour, wherein is a medicine for men. Verily herein is a
sign unto people who consider." To this Sale says:
"The same being not only good food, but a useful
remedy in several distempers, particularly those oc-
casioned by phlegm. There is a story that a man
came once to Mohammad, and told him that his
brother was afflicted with a violent pain in his belly:
upon which the Prophet bade him give him some honey.
The fellow took his advice but soon after coming again,
;
told him that the medicine had done his brother no

manner of service: Mohammad answered, 'Go and
give him more honey, for God speaks truth, and thy
brother's belly lies.' And the dose being repeated,
the man, by God's mercy was immediately cured."
Among the Mexicans the " medicine " god has split off
from the "hyena" god, and we find him as Ixtlilton
" the one with, the black face. " " They built for this god
an oratory from painted boards, a kind of tabernacle in
which his image was placed. In this oratory or temple
there were a large number of bowls and jars filled with
water and covered with boards or comalli. This water
was called tlilatl, which means 'black water.' When
a child fell ill, they took it to the temple of this god
Ixtlilton, opened a jar, made him drink this water, and
he was cured. If one wanted to celebrate the feast of
this god with personal devotion, its image was taken to
the house. It was then neither painted nor sculptured,
for a satrap just put on the ornaments of this divinity.
During the transportation they burned copal before it,
until the image came to the house where it was to be
celebrated with dances and songs, as was their custom,
for their manner of dancing is very diffOTent from ours.
I shall describe here the one which we cajl areyto, and
which they denominate in their language as maceualiztli.
They came together in large numbers, by twos or by
threes, and formed a more or less large circle, according
to their numbers. They carried flowers in their hands
and were adorned with all kinds of feathers. They
produced all together a uniform motion with their
bodies and with their feet and hands, a thing well done
and well worth seeing. All the movements were in
harmony with the music of the drum and teponaztli.
They accompanied the instruments with their voices,
singing in unison the praises of the god whose feast
they were celebrating. Even nowadays they give
theniselves over to the same exercises, although for a
different purpose. They regulate their movements and
adornments according to the nature of their songs, for
their dances and their intonations vary considerably,
without ever ceasing to be very charming and even full
of devotion. The forest of their idolatry has not yet
been rooted up.
" The image of the god having arrived at the house of
him who was celebrating the occasion, they at first set
out to eat and drink, after which began the dances and
songs with which they honored the divinity. The
god himself having danced for a long time, descended to
the cave where the pulque had been kept in various jars
covered for four days with varnished boards and comalli.
He opened one or more, an operation which was called
tlayacaxapotla, which means the wine is new. Then
'
'
he and those who accompanied him drank this wine;

then they went out and repaired to the yard of the
house where were the jars full of black water which was
dedicated to him and which had been covered for four
days. The one who played the part of the god opened
them, and if, having opened them, he found in any of
them any impurity, such as a piece of straw or hairs or
coal, they at once said that the man who gave the feast
was a bad man, an adulterer, a thief, or a libertine, and
everybody insulted him, imputing to him some of these
vices, and pretending that he was only a sower of discord
and trouble, and these offenses were addressed to him
publicly in the presence of all. When the man who was
the image of the god left the house, they gave him stufis,
called for that reason izquen, which means 'face cover,'
to allude to the shame of the celebrator of the feast,
when the waters were altered."^
From this description we see that Ixtlilton was a
dance god, just like a "medicine" griot. That he split
off from the Nama worship follows from his use of the
"black water," which, as we shall later see, refers to a
honeyed drink, used as medicine. Just as in the
Nama prayer the words are, "This was done for
shame's sake," so the Ixtlilton priest receives a face
cloth, apparently to be used as a veil. But in the
Nama ritual the women were not allowed to see the
' Sahagun, op. cit., p. 34 f. (book I, chap. XVI).
Nama, and, obviously, "for shame's sake," the streets

were cut off by mats to serve the same purpose. Seler
has pointed out that Ixtlilton, as a dance god, paints
his face black and is related to the Ueuecoyotl "the old
coyote."^ The painting of the face black, just as the
feather adornment and dance, is obviously of griot
origin, that is, points to the "tarring and feathering."
Ueuecoyotl is a translation of Malinke, Bambara
namakoro "the old hyena," which is a synonym of
nama "hyena, fetish." This Ueuecoyotl is represented
as the regent of the fourth Tonalamatl division, and
is clearly represented as a dance god,^ and is identical
with the Coyotlinauatl of the amanteca}
Among the Mayas the African Nama
represented is
by the long-nosed black god Ekchuah: "God L's
features are those of an old man with sunken, toothless
mouth. His hieroglyph is Fig. 44, which is character-
ized by the black face.
"God who is also black, must hot be confounded
L,
with M
whose description follows. L is represented
and designated by his hieroglyph in the accompanying
text, in Dr. 14b and 14e and Dr. 46b; the figure has the
characteristic black face. He appears entirely black
in Dr. 7a. The hieroglyph alone occurs in Dr. 21b
and 24 (third vertical line in the first passage) with a
variation, namely without the Ymix-sign before the
head. This deity does not occur in the Madrid and
Paris manuscripts.
"The significance of god L does not appear from the
few pictures, which are given of him. In Dr. 46b the
god is pictured armed and in warlike attitude. Both
in Dr. 14b and 14c he wears a bird on his head and has
a Kan in his hand.
1 Op. dt., p. 462.
* E. Seler, Codex Borgia, Berlin 1904, vol. I, p. 98.
' Ibid., p. 99.
"According to Forstemann, Ms day is Akbal, dark-
ness, niglit.
"Cyrus Thomas (Aids to the Study of the Maya
Codices, in the 6th Annual Report of the Bureau of
Ethnology, "Washington, 1888, p. 358) thinks he is the
god Ekchuah, who has come down to us as a black deity.
God M seems, however, to correspond to Ekchuah (see
the description of M).
"God M's hieroglyph is Figs. 45,46; it seems to rep-
resent an eye rimmed with black, though the figure of
the god himself displays an entirely different drawing
of the eye (see Fig. 47).
"The god is found in the Dresden manuscript only
three times, namely in Dr. 16b (with a bone in his
hand) in picture and sign, in Dr. 13e grouped with an
animal, without the hieroglyph, and in Dr. 43a (with
his sign) while finally his hieroglyph alone appears in
Dr. 56 (top, left) in a group and of a somewhat different
form.
"On the other hand, god M appears with special
frequency in the Madrid manuscript, which treats of
this deity with great fullness of detail. While he is
represented in the Dresden manuscript (16b) with his
body striped black and white, and on p. 43a entirely
white, he is always entirely black in the Codex Troano.
His other distinguishing marks are the following:
"1. The mouth encircled by a red-brown border.
"2. The large, drooping under lip. By this he can
be recognized with certainty also in Dr. 43a.
"3. The two curved lines at the right of the eye.
"His significance can be conjectured. He seems to
be of a warlike nature, for he is almost always repre-
sented armed with the lance and also as engaged in
combat, and in some instances, pierced by the lance of
his opponent, god F, for example in Tro. 3c, 7a, 29*a.
The peculiar object with parallel stripes, which he
wears on his head is a rope from which a package fre-

quently hangs. By means of a rope placed around his
head the god frequently carries a bale of merchandise,
as is the custom today among the aborigines in different
parts of America. On 4b and 5a in the Cod. Tro.
this can plainly be seen. All these pictures lead us to
conclude, that we have here to do with a god of
travelling merchants. A deity of this character called
Ekchuah has been handed down to us, who is designated
explicitly as a black god. In favor of this is also the
fact, that he is represented fighting with F and pierced
by the latter. For the travelling merchant must, of
course, be armed to ward off hostile attacks and these
are admirably symbolized by god F, for he is the god of
death in war and of the killing of the captured enemy.
The god is found in the Codex y^'^p-^n in the following
places and on many pages two or three times: pp. 2, 3,
4,5, always with the hieroglyph, then without it on pp. 6,
7, 19, 4*c, 14*b, 17*a, 18*b and again with the hieroglyph
on pp. 22*a, 23*a, 25*a; finally it is found again without
the hieroglyph on pp. 29*a, 30*a, 31*, 32*, 33*, 34*. In
the Codex Cortesianus god M occurs in the following
places: p. 15, where he strikes the sky with the axe and
thus causes rain, p. 19 (bottom), 28 (bottom, second
figure), 34 (bottom) and 36 (top). M is always to be
recognized by the encircled mouth and the drooping
under-lip figures without these marks are not identical
;
with M, thus for example in Tro. 23, 24, 25, 21*. Tro.
34*a shows what is apparently a variant of M with the
face of an old man, the scorpion's tail and the vertebrae
of the death-god, a figure which in its turn bears on its
breast the plainly recognizable head of M. God Mis also
represented elsewhere many times with the scorpion's

tail, thus for example on Tro. 30*a, 31*a.
"Besides his hieroglyph mentioned above. Figs. 45

and 46, another sign seems to refer to god M, namely
Fig. 48 (compare for example Tro. 5a and Cort. 28,
bottom). The head in this sign has the same curved
lines at the corner of the eye as appear on the deity him-
self. Forstemann mentions this sign in his Comment-
ary on the Paris Manuscript, p. 15, and in his Comment-
ary on the Dresden Manuscript, p. 56. He thinks the
hieroglyph has relation to the revolution of Venus,
which is performed in 584 days. A relation of this kind
is, I think, very possible, if we bear in mind that all the
god-flgures of the manuscripts have more or less of a

calendric and chronologic significance in their chief or in
their secondary function.
"It should be mentioned that god M
is represented
as a rule as an old man with toothless jaw or the
characteristic solitary tooth. That he is also related to
bee-culture is shown by his presence on p. 4*c of the
Codex Troano, in the section on bees.
"Besides gods L and M, a few quite isolated black
figures occur in the Codex Troano, who, apparently, are
identical with neither of these two deities, but are
evidently of slight importance and perhaps are only
variants of other deities. Similar figures of black
deities are found in the Codex Tro. 23, 24 and 25 (per-
haps this is a black variant of B as god of the storm?)
and on 21*c we twice see a black form with the aged
face and the solitary tooth in the under jaw (perhaps
only a variant of M). In the Codex Cortesianus and in
the Dresden manuscript no other black deities occur,
but in the Paris manuscript a black deity seems to be
pictured once (p. 21, bottom). '"^
That this Ekchuah is identical with Nama follows
from his being represented as old and as connected with
bee-culture. That he is identical with the "old coyote"
follows from his being represented as the god of the
'p. Schellhas, Representation of Deities of the Maya Manuscripts, in Papers
of the Peabody Miiseum of American Archaeology and Ethnology, Harvard
University, vol. IV, No. I, p. 34 S.
05
Q
o
o
Q
w
o
2;
6
2;
o
w
traveling meroliaiits, with whom, as we are informed by

Sahagun, the god of the amanteca was generally confused.
The Mayas celebrated his holiday together with Hobnil,
Uhe god of the bee-keepers,^ and during the feast they
drank only three bowls of honey wine.^ The long nose
of the Ekchuah is due to the fact that the idol of the
Nama, called the Kungolo Nama "head of the Nama,"
is represented by a fantastic bird, that is, with a beak.
The Ekchuah becomes among the Mexicans the
Yacatecutli "the lord of the nose," the god of the mer-
chants, with whom the Ixtlilton of the Amanteca is
confused.
It is not more difficult to identify the Mexican god
Quetzalcouatl, even though we observe here the same
tendency to split the original god into a number of
subsidiary forms, each accentuating a special attribute,
and though the Mexicans often confused the god with
the priest. The Mexican Quetzalcouatl, literally "the
Fine Feather Snake," "originally, no doubt, was the
symbol of water or the moisture produced by rain,
which, after a long drouth, awakens vegetation to new
life. Even thus the Chiapanec priests explain Cuchul-
chan as the feather snake which goes into the water.
'
It is the regent of the seventh sign, that is, as we shall

soon see, these priests identified this feather snake with
the rain god Tlaloc. Similarly K'ucumatz of the
Guatemalan myth appears entirely as the principle that
lives in the water. U c'ux cho u c'ux palo 'heart of the
sea, heart of the water' he is called in the Popol Vuh:
xa pa ya xu col vi ri 'in the water is his realm of activity'
say of him the Cakchiquel annals. And when Saha-
gun tells us that the sacrifices brought by the Mexicans
in the beginning of their year, according to the state-
ment of some are dedicated to Tlaloqu^, the rain gods,
1 Seler, Codex Borgia, vol. I, p. 321.
» Ibid., p. 322.
according to others to Chalchiuhtlicue, the water god-
dess, and according to others again to the chief priest and
wind-god Quetzalcouatl, and when in the Codex Borbon-
icus the sixth annual holiday Etzalqualiztli, which was a
great rain ceremonial and was dedicated to the rain
gods, was represented by the image of Quetzalcouatl and
his twin brother Xolotl, the same basic conception seems
to be represented even here. And yet it must be some-
what baffling to find the god everywhere in Mexican
"^
tradition represented as the Eecatl, the wind god.
"A peculiarity of Quetzalcouatl is the priestly char-
acter attached to him, and that to him are ascribed the
invention and conscientious execution of penance and
chastisements, of blood-letting and offering of one's own
blood, which was one of the most common ritual per-
formances of the old tribes of Mexico and Central
America. This is related to his role as master and king
of the Toltecs, since the Toltecs were supposed to have
been the inventors of all civilization and with it of the
ritual and priesthood. . . The most plausible and
.
simplest assumption is this, that in the remarkable

figure of this god, the rain god was combined with the
rain magician, who, with his prayer and his practices,
insured to his people the rain needed for their crops.
No doubt it was a later interpretation which added to
him the nature of a wind god, since the wind god was
explained as the forerunner and road-cleaner of the
rain god. Before the rains begin there are great winds
and dust storms, and so they said that Quetzalcouatl,
the wind god, swept the roads for the rain gods, so that
they could rain. "^
From all the accounts of Quetzalcouatl it appears that
he was a tutelar deity, from whom prosperity could be
received through his grant of rain necessary for the
> Ibid., p. 83.
2 Ibid., p. 85.
crops. Wehave in Mm. obviously the Bambara bene-

ficent dasiri, the protector of the village, the favorable
înn, whose sacred animal is a snake, a rat, a lizard,
etc., who is addressed with the words: "Preserve us
from evil-doers, from discord, quarrels, and brawls,
from the fury of women who easily succumb to the
power of the devil, from disease; above all, give us rain,
without which the harvest is impossible/'^ The feast of
the dasiri takes place once a year,^ just as that of
Quetzalcouatl. But the worship of the dasiri is closely
connected with the activity of the Mandingo kuare,
kore, or kote, a semi-secret society, with which we must
now become acquainted. "The fundamental basis of
power and authority, the cult of the dasiri, may, with
good right, be considered as the first and oldest of all.
That of the kote or kore seems to me equally respectable
from this standpoint and equally old. As the first of
all, it is held in a small grove near the village, and the
baobab, the khay, the nettletree are the three trees he
likes to live in. The cult of the kore is especially hon-
ored in the Bani. Although not all the villages now
possess the fetish, all at least have adepts who belong to
the sect. This semi-occult sect, which counts women
in its midst, is divided into eight groups, each ruled by
the oldest under the authority of the koretigi, highest
chief of the association. In the yearly sacrifices and
especially in those which take place every seven years
on the initiation days (and one has to be a kore de, in
order to participate in the inner circle of the mysteries
which take place in the sacred grove), these groups
present to us the most biirlesque assemblage of the
circle of Segu. The kara mau carry a perforated board,
painted red and white, and about two meters long;
the suruku (hyenas) put on a mask resembling that of
' Henry, op. cit, p. 120.
2 Ibid., p. 101.
the hyena and walk leaning on two canes; and the
koroduga (buffoons) are dressed up in an indescribable
manner. The ta tugula (fire-burners) dance with a
burning torch in each hand; the n'goni sama (thorn-
attraetors) wrap themselves in thorns or lacerate their
breasts and armpits, and the bisatyila (flagellants)
strike themselves with long, flexible scourges. There
are also the dyara (lions), who wear masks resembling
those of the hyena, but of a larger shape, and the sula
(monkeys), who walk about in a hideous, grimacing
mask and wear in the back tails made of plaited grasses.
"The sect of the kore is semi-occult, for if the adepts
alone are admitted to the sacrifices and to penetrate the
sacred grove, the dances take place in the village, and
everybody, men, women, and children, take part in
them. The wives of the koroduga belong to the
brotherhood by right, but they do not know the pass-
word dyantema (men of the dyante), and so are refused
admission to the sacred grove. It is their great privi-
lege to wear the livery of their husbands, a long necklace
of red beans, to be able to sit down with the men to the
beer calabashes, and, finally, to dance the indecent,
disgusting dance of the koroduga without blushing.
The fetish kore guards the crops and gives its adepts the
only pleasure which they know and long for, eating and
drinking, and the enjoyment of women, and all this, as
one may convince himself, follows from their dances
and orgies. . . The initiation or affiliation with the
.
sect of the kore takes place once every seven years. It

is long and painful; for two weeks our young initiates
live in the sacred grove and may not enter the village.
As they need a little training for their future dances,
the old people come every day at noon-time, during the
greatest heat, to torture them, and every adept, ac-
cording to the group to which he belongs, suffers a few
moments of torture by fire, thorns, and the rod. . . .
Independently of the yearly sacrifice, they sacrifice

to the ginn every time an adept passes away. They
also sacrifice at the anniversary of his death and during
the dry season, and our kore de run from one feast to an-
other, passing their nights in drinking bouts and in-
expressible orgies. "^ "After the initiation, the koretigi
(priest of the kore) proceeds to the sacrifices. Every-
thing is as with the dasiri; after the libations of flour
and beer, the trunk of the sacred tree is smeared with
blood, and all about it are tufts of hair and feathers
stuck to the bark by bits of chewed kola nuts. They
smear with blood a hyena mask called kore kungolo
(head of the kore), then the kara, emblem of the ginn,
which serves as an altar for the deceased adepts. A
part of the viands is taken to the village, and the re-
mainder, roasted on the spot, is eaten without any
condiments. In these sacrifices they ask the kore
and those who form his court to watch over the crops,
so that they should be abundant, to watch over the
flocks, so that they should prosper, to have all their
lives enough to eat and drink, to have a lusty old age,
etc., and all this is asked with the only purpose in view
'of being able to enjoy women and giving themselves
over to carnal pleasures. "^
'
The very name of the society, kuare, kore, kote, shows

that we are deaUng here with an aberration of an Is-
lamic brotherhood, no doubt of Sufi origin, known by
the Arabic name of V^- haunlyah, 'j^' ahuwwat, and
preserved in Berber hauni, plural huan, "member of the
pious brotherhood of the Khuan." As the kore is the
society of the koroduga, the buffoon, it is not unlikely
that the obscene practices in these societies and the
masquerading are of Asiatic origin. In any case, it is
clear that the practices of the kore are subsidiary to
1 Ibid., p. 102 ff.
2 Ibid., p. 124.
those of the tutelar divinity, the dasiri. But in the
Quetzalcouatl myth we have the same reference to the
chastisement and use of the thorn-pricking that we have
in the case of the kore. Finally, Quetzalcouatl is repre-
sented with a beard, ^ and so is the sun god Tonacotecutli.
But the beard appears again in the case of an unnamed
old god, whose mask resembles that of the feathered
snake and of the water goddess Chalchiuhtlicue} There
can be no doubt that the beard is to represent an old
man, and this is precisely what we find in Africa in
connection with the dasiri worship, in which all the
house chiefs, the sotigi,^ are old men, and the religious
priest, who presides over them, the nyenansonaba, is
an old man* and, as Henry says, a barbe blanche.
As the dasiri and kore festivals take place annually,
the god, or spirit, becomes the measurer of time.
Similarly, Quetzalcouatl is considered to be the inventor
of the tonalamatl, the calendar, and it is significant that
in this case he is represented in connection with a tree,
even as the dasiri and kore are inseparable from a tree.
The dasiri is supposed to live on a specific tree, which
is sacred;^ even thus Quetzalcouatl is represented as a
humming-bird kneeling on the top of a tree.^ The
dasiri is also worshipped on an altar of a conical or
truncated form, which supports a clay bowl. If this
altar is not near a cross-road, or in a public place, but in
the house, it will be of any simple shape, a stake with
three-coriiered prongs on which to place the bowl.^
Otherwise the bowl which receives the libations is placed
under the tree or on the first branches of the tree.^ We
' Seler, Codex Borgia, vol. I, p. 87 et passim.
2 Ibid., p. 243.
' Henry, op. cit., p. 100.
'Ibid., p. 116.
' Delafosse, Haut-Senigal-Niger, vol. Ill, p. 168.
" Seler,Codex Borgia, vol. II, p. 66.
' Delafosse, Haut-Senegal-Niger, vol. Ill, p. 169.
«
Ibid., p. 168.
M> -
f Ml.
Fete d« A 'inmo. Ki) Itatil: Fete du \^toiitft.'

Fln.'.'lliili-.ri .!.< .'i.fjinl-;.
AFRICAN CHASTISEMENT, from Hcnr>-'s L-s Bamhara.

Arbre f6tiche tlanh-i (de Songobougou).
From Henrv's Les Bamhara.

have the representation of this tree in America: "One

of these is the central design in the Chilan Balam, or
Sacred Book, of Mani. It was copied by Father Cogol-
ludo in 1640, and inserted in his History of Yucatan,
with a totally false interpretation which the natives
designedly gave him.
"The lettering in the above figure is by the late Dr.
C. H. Berendt, and was obtained by him from other
books of Chilan Balam, and native sources. In
CogoUudo 's work, this design is surrounded by thirteen
heads which signify the thirteen ahau katuns, or greater
cycles of years, as I have explained elsewhere. The
number thirteen in American mythology symbolizes the
Ithirteen possible directions of space. The border,
therefore, expresses the totality of Space and Time;
and the design itself symbolizes Life within Space and
Time. This is shown as follows: At the bottom of the
field lies a cubical block, which represents the earth,
always conceived of this shape in Mayan mythology.
It bears, however, not the lettering, lum, the Earth, as
we might expect, but, significantly, tern, the Altar.
The Earth is the great altar of the Gods, and the offering
upon it is Life.
"Above the earth-cube, supported on four legs which

rest upon the four quarters of the mundane plane, is the
celestial vase, cum, which contains the heavenly waters,
the rains and showers, on which depends the life of
vegetation, and therefore that of the animal world as
well. Above it hang the heavy rain clouds, muyal,
ready to fill it; within it grows the yax che, the Tree of
Life, spreading its branches far upward, on their ex-
tremities the flowers or fruit of life, the soul or immortal
principle of man, called ol or yol."^
'
D. G. Brinton, A Primer of Mayan Hieroglyphics, [Philadelphia] 1895,
p. 47 f.
Here Brinton equally missed the explanation, for the
picture exactly represents the Bambara altar, the bowl
on the lower branches, the rain which conies from the
clouds. We
shall return to the thirteen heads later.
The Codex Cortesianus has the representation of
Quetzalcouatl under a tree: " Turning now to the central
design of what has been called the 'Tableau of the
Bacabs, in the Codex Cortesianus, Fig. 10, we can
'
readily see in the light of the above explanation that its

lesson is the same. The design is surrounded by the
signs of the twenty days, beyond which the field (not
shown in this cut) is apportioned to the four cardinal
points and the deities and time-cycles connected with
them.
"Again it is Life within Space and Time which the
artist presents. The earth is not represented; but
we readily recognize in conventionalized form the great
Tree of Life, across it the celestial Vase, and above it the
cloud-masses. On the right sits Cuculcan, on the left
Xmucane, the divine pair called in the Popol Vuh the '
Creator and the Former, Grandfather and Grandmother

of the race, who give Life, who give Reproduction. ' In
his right hand Cuculcan holds three glyphs, each con-
taining the sign of Life, ik. Xmucane has before her
one with the sign of union (sexual) ; above it, one
containing the life-sign (product of union); and these
are surmounted by the head of a fish, symbolizing the
fructifying and motherly waters.
"The total extension of the field in these designs re-
sembles the glyph a in Fig. 6. It is found in both
Mayan and Mexican MSS., and expresses the concep-
tion these peoples had of the Universe. Hence I give it
'"^
the name of the 'cosmic sign.
Here again we have the dasiri tree, bowl, rain, and
god, although a second, a female divinity, is added.
' Ibid., p. 48 flf.
'•••« .;c
It
I .^î^^
r,„l„> r,,ri,.« II. I-.'
Die viiT Vif-rti'l dfs TMiialainad't; uml rtir- vioi' Himijn.'i:-:ri<-!ittjii
PLATE OF THE BACABS, from Sclcr's Gesammelte Abhandlungen.

This is only natural, since the dasiri is equally used as
the tutelary god of women: "If the women will give

birth to a baby, they vow to call it dyiriba (great tree),
nyenamake (male înn); they even take pleasure in
adding the name of the village where the beneficent
dasiri is found: Kolotomo nyena ma, boy, girl of the
'
ginn Dasiri of Kolotomo. The name nyenama is, in-

'
deed, given only to girls, and for the boys they add the
appellation ke 'male. '"^ But this drawing, generally
known as the "Plate of the Bacabs," has a far greater
significance. Its full form is described by Cyrus
Thomas as follows: "This page consists of three di-
visions: First, an inner quadrilateral space, in which
there are a kind of cross or sacred tree; two sitting
figures, one of which is a female, and six characters.
Second, a narrow space or belt forming a border to the
inner area, from which it is separated by a single line;
it is separated from the outer space by a double line.
This space contains the characters for the twenty days
of the Maya month, but not arranged in consecutive
order. Third, an outer and larger space containing
several figures and numerous characters, the latter
chiefly those representing the Maya days. This area
consists of two distinct parts, one part containing day
characters, grouped together at the four corners, and
connected by rows of dots running from one group to
the other along the outer border; the other part con-
sisting of four groups of figures, one group opposite each
of the four sides. In each of the four compartments
containing these last-mentioned groups, there is one of
the four characters shown in Fig. 1 (abed), which, in
my 'Study of the Manuscript Troano, I have con-
'
cluded represent the four cardinal points, a conclusion
' Henry, op. cit., p. 101 f.

'

also readied by Rosny and Sctultz Sellack. "^ We
shall now begin ab ovo and will show the evolution of
certain aspects of astrology and magic from the Arabs
through the Mandingos to the American Indians, es-
pecially of Mexico and Central America.
Among the Arabic talismans the most highly prized
are those which are written on paper and are called Jj-*»-
gadwal: " Djadwal (PI. djaddwil) means firstly, 'brook,'

'watercourse'; it further means 'table, plan' (in this
meaning derived from schedula?). It thus becomes a
special technical term in sorcery, synonymous with
khatim; here means quadrangular or polygonal,
it
sometimes also circular figures, into which names and
signs possessing secret magic powers are inserted in the
most varied fashion. These are usually certain mys-
terious characters, Arabic letters and numerals, magic
words, the names of God, the angels and demons, as
well as of the planets, the days of the week, and the
elements, and lastly pieces from the Ko'ran, like the
Fdtiha, the Surat Yasln, the so-called throne- verse '
etc. The application of these figures is manifold;

frequently the paper on which one has been drawn is
burnt to smoke some one with its smoke or the writing ;
may be washed off in water and drunk; along with the

da'wa (conjuration) and often also the kasam (oath) the
djadwal forms the contents of a hirz (amulet). The
very popular da' wat al-Shams is, for example, prepared
as follows it is quadrangular, is divided into 49 sections
:
by six lines drawn lengthwise and six drawn across its

breadth and contains: 1. The sab' a khawdtim, i. e.
Solomon's seal and other peculiar figures. 2. The
seven sawakit or consonants which are not found in
Sura I. 3. The names of God, Fard, Djabbar, Shakur,
^ Notes on Certain Maya and Mexican Mamiscripts, in Third Annual
Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institu-

tion, 1881-82, Washington 1884, p. 7.
|<^. U^Ss ^fôr' 5^t-«
^M
II serait facile de multiplier ind^flniment ces exem-

ples. Nous nous bornerons & donner encore deux
icl
amulettes. La premiere est destinee h procurer h son

beneflciaire la bienveillance, ['affection, I'amour de telle
ou telle personne et aussi un bon accueil de la part des
personnages puissants
ARABIC GADWAL, from Doutte's Magie et religion dans I'AJrique du Nord.

MEXICAN GADWAL, from Codex Fejenmry-Mayer.
Thdbit, Zahlr, Khabir and Zaki. 4. The names of the

seven RuhiyaHl, Djabriya'il, Samsama'il,
'spirits':
Mlka'il, SarfiydHl, Aniyd'il and Kasfiya'il.
' 5. The
names of the seven kings of the djinns: Mudhhib, Marra,
Ahmar, Burkdn, Shamhurash, Abyad and Mlmun. 6.
The names of the days of the week. 7. Those of the
planets. The underlying notion is that secret relation-
ships exist between those various components and the
djadwal is therefore made to obtain definite results from
the correlations of the heterogeneous elements compos-
ing it. In this way new djadwals for particular purposes
come to be made: these are also made by using the
above mentioned seven seals. The extremely compli-
cated system of mystic letters, which is based on the
numeral values of Arabic letters, is very frequently
used for the djadwal. A
special class is formed by the
squares called wifk, in the fields of which certain figures
are so arranged that the addition of the horizontal and
perpendicular lines, as well as that of the diagonals
gives the same total (e. g. 34 or 15). The quadrilateral
containing the celebrated magic name budiih is derived
from such an arrangement."^
To us the most interesting gadwal is the one which is
the basis of the "Plate of the Bacabs" in Central
America. Like that one, it consists of a central circle
with four radiating demi-diagonals of a square surround-
ing it. In the center are the words, "Such a one, son
of such a woman. " The diagonals bear as inscriptions
four verses or parts of verses from the Koran, and parts
of these verses are given in the corners, while the rest of
the square is filled with Arabic letters making no sense
whatsoever. The form is identical in a general way
with the American "Plate," but the American "Plate"
belongs to one of the more complicated âdwals, as de-
» M. Th. Houtsma, etc., The Encyclopaedia of Islam, Leyden, London 1913,
p. 992.
scribed above. We have no means of ascertaining
whether the Mandingos possessed the âdwal from
which the American "Plate " and its Uke were produced,
since nothing of documentary antiquity has come down
to us, but we have still ample documentary evidence to
prove that the American "Plate" has gone through a
Mandingo redaction.
In the the central square contains the
first place,
Mandingo tutelary god with his attributes and appur-
tenances. The numerical calculation based on 20 and
13, which is the essence of the American calendars, is
surely built on African models. Here again we possess
but the scantiest material for verification, but just
enough to be startling and unique. Travelers have
taken no trouble to ascertain African calendars and
chronologies. The following few facts are about all we
know. The Habbes have a lunar month of five weeks
of six days each.-^ The Tchi tribes have a seven-day
week varying length,^ while the Yorubas have a week
of
of five days, six of them making a lunar month.^ The
Islamic week has everywhere else taken the place of the
native time reckoning, but as the numeration of the
Mandes, Uke that of most Sudanese people, is based on
"five,"* the original week was unquestionably the
same as that of the Yorubas. For astrological purposes
there was in use a division of the zodiac in thirteen parts,
such as has been found on three calabashes in western
Africa,^ and it is a curious fact that a similar division
into thirteen is recorded only among the Kirghizes and
in America. The division of the year into thirteen
parts would demand a twenty-eight day month, but, in
1Desplagnes, op. cit. p. 377.
2A. B. Ellis, The Yoruba-Speaking Peoples of the Slave Coast of West Africa,
London 1894, p. 142 f.
' Ibid., p. 143.
* Delafosse, HauirSinSgal-Niger, vol. I, p. 404.
' F. Bork, Tierkreise auf westafrikanischen Kalebassen, in Mitteilungen der
vorderasiatischen Gesellschaft, vol. XXI, p. 266 ff.
•"ig. 8:1. — Tv|.(- il'in.li-riv; II.-iI,Ih--K,i^-.\iiiIi;i .Irvniil <în5 ru[ê>l,ri^s
« -••"^
Fig. S6. — r»(>ssins nipestre.-* 11 tS kil. riu N.-O. ilc Handififtara.
SUDANIC ROCK INSCRIPTIONS, from Desplagnes' Le plakau

central nigenen.
SUDANIC ROCK INSCRIPTIONS, from Desplagncs'
Le plateau cc7itral 7iigerien.
)>'{ VL.
/>(U ^3 f^" •./""'•-.

\
"^
•/
'.
^^ u
-a. •-;••
NEPHRITE STATUETTE FROM SAN ANDRES TUXTtA, VERA CRUZ

SIDE VIEW
From American Anthropologist, Vol. IX.

,
reality, the order is reversed, for we still have among

the Berbers a division of the year into twenty-eight
parts, of thirteen days each,^ which is based on the
astronomical or astrological calculations of the Arabs,
whose twenty-eight lunar mansions of thirteen days
each were, in the IX. century or later, adopted from the
Hindus,^ who had by that time arranged the twenty-
eight nakshatras, or constellations, into equally spaced
divisions of the zodiac, which naturally led to the
thirteen days unit of time.
The Arabic âdwal has also the meaning of vertical
'
'
column, table of the zodiac, talismanic signs written in

columns. " Precisely such âdwals have been found in
the region of the Mandingos,* and these have glyphs
that bear an amazing resemblance to the Central Amer-
ican glyphs, especially those of the Tuxtla statuette,
where we find similar signs encysted in squares and
parallelograms.* Unfortunately we possess only the
three photographs of the African inscriptions which
Desplagnes has reproduced. Most of the columns in
two of these begin with the forms of animals, the spider
and lizard, which are also scattered throughout the
columns. The spider is identical in form with the one
given on Mound-builders' gorgets,^ where the cross in
the center indicates that it is related to the tonalamatl
of the Mexicans and the Maya calendar. In Africa the
• "The solar year is divided into twenty-eight manazil (sing, minzla)
each containing thirteen days, with the exception of the iebha (18th-31st
July), which contains fourteen," Westermarck, Ceremonies and Beliefs,
p. 73.
2 C. A. Nallino, in History of Arabic Astronomy (in Arabic), in MUteilwn-
gen zur Geschiehte der Medizin, vol. X, p. 552.
' Desplagnes, op. cit., p. 78 £f.
* S. G. Morley, The Inscriptions at Copan, Washington 1920, p. 403. See

also W. H. Holmes, On a Nephrite Statuette from San Andres Tuxtla, Mexico,
in American Anthropologist, N. S., vol. IX, p. 691 ff.
' W. H. Holmes, Art in Shell of the Ancient Americans, in Second Annual
Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institu-

tion, 1880-'81, Washington 1883, p. 286 flf.
"

spider is connected witli an enormous number of tales/
and among the Hausas the rainbow is called bakan
gizzo, literally "the spider's bow," which indicates the
relation which the divinities that came down from
heaven bear to the spider. Indeed, " since the spider is
the king of cunning and craftiness, all fables are told in
his name. "^ Among the Hausas 'Gajjimare is the god '
of rain and storms, which has the shape of a snake, and

IS double-gendered, the male part being red, the female
blue. It Uves in the storm-clouds (same name),
but is supposed to come out at night, and it is also said
to inhabit walls, and in fact all watering-places, so a
pot is kept full in every house. Gajjimare (rainbow)
may be represented by the water-serpent killed in the
legend of Daura before referred to, but sometimes it is
said to be the husband of Uwardowa, and the father of
Kuri. Other names of the rainbow are Masharua,
'water drinker,' and Bakkan gizzo, 'spider's bow. '"^
In Tchi "legend" is called anansesem, which means
"story of the spider."* That the Mandes similarly
connected their stories with the spider follows from
Bambara n'tale, which means both "proverb, parable,
and "spider." The African connection of the two
concepts is, however, not of native origin, but is due
to Arabic homonyms. Bambara n'tale, Malinke tali
"fable, story," talln "spider" are all derived from
Arabic "^ J ratlah, *^~J rutailah, Hassania "^"J retla
"spider." But among the Arabs there are a number
of words from the same root which mean "to recite in
a leisurely manner, and distinctly, to chant the Koran,
' C. Spiess, Fabeln iiber die Spinne bei den Ewe am
Unterlauf des Volta in
Westafrika, in Mitteilungen des Seminars filr orientalische Spra^hen, Berlin
1918, vol. XXI, part III, p. 101 S., and Fortsetzung der Fabeln iiber die Spinne
bei den Ewe am Unterlauf des Volta in Westafrika, ibid., vol. XXII, part III,
p. IB.
^Tremearne, Hausa Superstitions and Customs, p. 10.
'/6id., p. 112.
* Perregaux, op. cit., p. 186.
j//^^^fi-
Fig. 26. — Pattern on Vxiot similar to fig. 25.
GADWAL DESIGN IN AFRICA, from Trc-

mearne's Ilausa Superstitions and Cusloms.
Fig. 38. — Pattern oa knob of fig. 35.
Gy\DWAL DESIGN IN AFRICA, from Trcmcarnc's

llausa Superstiiiovs and Customs.
', /
^m"^- \ " ^--^1 •
GADWAL DESIGN IN MOUND-BUILDER GORGETS,

Ameneans.
in Shell of the Ancient
'
from Holmes' //r^

%fF#
V V
(^^â^
i
1, ajfoU t'.iri^ct fr.iat -itc-ui^ Kruve, T'-i
3, Shall ôryol froiu rtloao Kfjvc. 'J.'.'i
laE BIRIi.
GADWAL DESIGN IN A'lOUND-BUILDER GORGETS,

from Holmes' ^-Irt iti Shell of the Ancient Jmericans.
Kl
o
o
s
z
a
w
Q
Q
<;
e>
o
u
w
o
W
Q
<
mmt0
and this became contaminated, by Arabic ^" iala "to

recite tbe Koran," hence the spider has become closely
related to a well-told story.
In some places in Africa the scorpion has taken the
place of the spider, for in Songay don "to sing" also
means "scorpion," which can only be the result of a
translation of Arabic ^J ratlah. As the spider is con-
nected with the rainbow, it connects heaven with earth,
and the spirits of heaven, the stars, the constellations,
are thus brought in contact with the spider; whence it
is only natural that the âdwal, which deals with
astrology or astronomy, should be connected with the
spider. For this reason the spider of the Mound-
builders has in its middle the cross, which is the simplest
representation of the fourfold division of the âdwal.
This ornamentation is in constant use in the Western
Sudan. It forms the central design of circular ob-
jects,^ is done in square patterns with looped ends, ex-
actly as in bird âdwals in Mound-builders' gorgets,^
and is worked in dotted line form upon knobs.^ The
cross so often found in these designs has nothing what-
soever to do with the Christian cross.
In the Mound-builders gorgets we not only have the
'
spider with its âdwal, but the looped-end gadwal,

with a cross in the middle, all placed within a circle, is
the most striking object found in the mounds. That
we have in these cases a development and simplification
of the gadwal with the Mandingo dasiri, the tutelar
god and rain giver, who lives upon a tree, is brought out
in a number of cases, where each side of the square has
the representation of a bird 's head.* Thus the Mound-
builders gorgets are mere modifications of the "Plate
'
' Tremearne, Hausa Superstitions and Customs, p. 29.

2 Ibid., p. 97.
' Ibid., p. 145.
' Holmes, Art in Shell of the Ancient Americans, p. 282 ff.
of the Bacabs, " where the bird god is represented as the
long-nosed god. In the Codex Cortesianus (plates XLI-
II) we have an interesting evolution of the tonalamatl.
The four squares are filled respectively with figures of
Tlaloc, Macuilxochitl, Quetzalcouatl, Tlagolteotl, while
the middle of the gadwal is occupied by a convention-
alized spider, representing the setting sun. Before
dealing with this new aspect of the gadwal, which in-
troduces the sun as the central figure, it is necessary to
investigate the close relationship which exists between
certain games of chance and the astrological gadwal in
Africa and in America.
Ferrand^ has given an exhaustive treatment of the
Malagassy divination called sikidy, which is nothing
but the Arabic geomancy, found, with certain varia-
tions, wherever Islamic influence has been exerted.
Burton has described the geomantic board of the
Yorubas: "TheBuko-no ignored the Yoruban triad,
Shango, Oro, and Obatala; but he agreed with the
Egbas about Afa. Seeing that I had some knowledge
of the craft, he produced from a calico bag his 'book,'
a board, like that used by Moslem writing-masters, but
two feet long by eight inches, and provided with a dove-
tail handle. One side of this abacus contained what
are called the sixteen 'mothers,' or primary, the other
showed as many children, or secondary, figures. Each
was in an oblong of cut and blackened lines, whilst at the
—
top were arbitrary marks circles, squares, and others,
to connect the sign with the day. It began with the
Bwe-Megi, the figure, assigned to Vodun-be fetish —
—
day, or Sunday, whose mnemonic symbol was six
dots in a circle; whilst Monday had a sphere within a
sphere. It was a palpable derivation from the geom-
ancy of the Greeks, much cultivated by the Arabs under
the name of El Rami J* J' ,
'
The sand, ' because the
' G. Ferrand, Les Musulmans d Madagascar, Paris 1891, vol. I, p. 73 ff.
—
figures were east upon the desert floor. 'Napoleon's

Book of Fate' is a notable specimen of European and
modern vulgarisation. The African Afa is not, as in
Asia, complicated with astrology; and no regard being
paid to the relative position of figures, it is compara-
tively unartful. Two details proved to me its Moslem
origin: the reading of the figures is from right to left,
and there are seven days, whereas the hebdomadal
week is beyond the negro's organisation." "The
following note will explain the use of the palm-nuts, and
the names of the figures:
"In throwing Afa, the reverend man, or the scholar,
if sufficiently advanced, takes 16 of the fleshy nuts of a
palm, resembhng the cocoa- trees; these are cleared of
sarcocarp, and are marked with certain Afa-du, or Afa
strokes.
"When Fate is consulted, the 16 nuts are thrown
from the right hand to the left; if one is left behind, the
priest marks two; if two, one (the contrary may be the
case, as in European and Asiatic geomancy) and thus
;
the 16 parents are formed.

"The 16 are thus named and made:
Called Bw6 Megi: it is the Mother of
all.
Yeku Megi.
Wudde, or Od6-Megi.
DI-Megi.
5. Losu Megi.
Urdn Megi: an inversion of No. 5.
7. Called Abla Megi.
8. Akla Megi; or Abla inverted.
9. Sa Megi.
10. Guda Megi: an inversion of No. 9.
11. Turupwen Megi.
12. Tula Megi.

13. Lete Megi ; or Tula inverted.
14. KS, Megi.
15. Ch6 Megi.
16. Fli Megi: considered the Father of all.
"These 16 parents may have many children. Nos.

13 and 2, for instance, make
—and"^ so on, showing an infinite power of combina-

tion. Becker has shown the fallacy of Frobenius'
work on the Yorubas, because of his neglect to connect
their Ifa worship with the Arabic geomancy, and has
explained some of the terms of Yoruba divination and
fetishism as translations from the Arabic.^
Some of the terms of the chief geomantic figures are
found in Hausa in the sense of "story," while one of
these, from its association with the zodiac, also means
' R. Burton, A Mission to Gelele, King ofDahome, London 1864, p. 332 ff.
2 C H. Becker, Der Islam, Strassburg 1913, vol. IV, p. 303 ff.
)
"star." Thus the first sign, ^ ; ;

*^^ âma' ah "oon-
*
gregation" produces Hausa aljama, aljima; the eleventh,

.' . '-r^ humrah "red" leads to Hausa almara; the
third, !
"Î i^timdh "gathering" leads to Hausa
gatana; another Hausa term, tatsuniyu, tasunia, tasinia
"story, star," I am unable to identify as to origin, but
the two connotations show that they similarly arise
from the astrological âdwal.
An Englishman, upon reading an account of the
Malagassy geomantic table, devised a game of SMddy,
"played with boards of 8 squares, markers, counters,
and dice. "^
In this he simply duplicated what had
long been observed by the users of the geomantic gadwal,
for forms of pachisi, as found over an enormous terri-
tory, are nothing but âdwals used for games of chance.
In Arabic J^ qimdr refers to any game of chance. The

Spanish-Arabic dictionary in the beginning of the XVI.
century translates Spanish "dados" and "naypes" by
quimar, which shows that even at that late date "dice"
and "cards" were not yet fully distinguished. But
"cards" were called naipes in Spanish from Arabic v'^"
naib "lieutenant, " and the first fundamental row of the
geomantic gadwal is called alanaua,^ unquestionably
from naib "lieutenant, regent, " for we find this word as
laibe "story" in Wolof, which indicates that in the
Western Sudan the game was closely related to the
gadwal. Cards seem not to have been known before
the end of the XIV. century, and it is significant that,
^ M. Steinschneider, Die
„ Skidy " oder geomantischen Figuren, in Zeitschrift
der deuischen morgenldndischen Gesellschaft, vol. XXXI, p. 762.
^ J. Sibree and R. Baron, The Antananarivo Annual and Madagascar
Magazine, Antananarivo 1885, vol. IX, p. 324.

' Steinschneider, op. cit., p. 763.
.M.h. i:!. Ii-i 'li.ti ,U.
J-iiNl r.,M.-rli„.-,|i„,in
-M'li II. M<h<n! ,„rl,.Jl. (:,,() ,|,,y ALL. L', li.iC.li .I/,,.,,,/ /,„./,f/, j.'h, It ii:i^:'T v..i- d.'rn .^.tkitiji
Spi.-I" Salun-im-M^ Ilibl, dv! Kauiu. li.ir. S.tli:,uiiii-Ms j;ii,i ,i..i r,/_-,/„y„„-/,//-, -J.'^ I i..tl|..s
r.ilaiio. il|!„lu(l.-| i'iil..n... i.M:nlri.|.i S:.liat'iin-.Ms. P.ilil. J. 1 r«lafi.)

(M.ntPlclj
0©©©
AM.. II .1/./.-..,; .,-,,.;„<;, fioti j,..- ,„,(r.m,S|,iL.i..-.

Cml.., Mashaljotiu,,,,,, xtn, :;. f.,1. rj-j.
A1EXICAN GAME OF PATOLLI, from Seler's Codex Borgia

although the original deck of cards had 4X18 and more

cards, it soon developed into a deck of 4X13 cards, in
which 13 is identical with the calabash zodiacs of
western Africa. It, therefore, follows from this that in
western Africa there was, for reasons which we do not
at present know, in vogue the 4X13 astrological cycle,
which forms the basis of the same cycle in Mexico and
Central America.
It is significant that in Mexico the game of patolli is
under the supervision of the god Macuilxochitl, who
represents one form of the sun god, and this god, in his
turn, is related to and confused with Xolotl. Here we
are once more on safe ground, and the Mandingo origin
of the sun myth becomes clear. The Mexicans placed
their terrestrial paradise in the west, but also thought of
it as connected with the heavens, where the gods were
born. Its name, for which the Nahuatl furnishes no
explanation,^ is Tamoanchan. This, of course, is
Malinke duna do arjana "paradise." The heaven
divinities of the Mexicans are nothing but the zodiacal
signs of the nocturnal sky, "the descenders from
heaven. " Xolotl is distinctly the gemini of the zodiac.
The word means "twin, "^ in various connotations. It
is used for "twin, house-slave, dog," that is, "com-
panion." This at once identifies it as Arabic 'j-^
'uSdrah "to accompany one," ''v^ û§ara\ plural of

.jy^ "aSlr, "associate, friend, relation."This word is
found throughout the Sudan. We have Songay tjero
"friend," tjere, djere "half, comrade," Malinke teri,
Bambara teri, terike, terke, nterge "comrade, friend,"
etc. The twin god was the god of twins among the
Bambaras: "This fetish protects the mother of twins
' Sahagun's derivation from tictemoa tochan "we seek our home" is useless.
Op. cit, p. 9.
' Seler, Codex Borgia, vol. II, p. 37.
and unites them in so close an affection that one never
receives a present without sharing it with the twin.
If one of the twins dies young, and this is generally the
ease, the surviving child receives a small statue which
he always keeps with jealous care, and to which he
gives the name of the deceased child. He dresses it in
the best manner possible, often covers it with beads,
pearls, bits of amber, rings, etc., and refined people never
give him a present without adding at least five cowries
for the statue. The sinsin gratifies the twins with a
peculiar privilege. Not only can no scorpion sting
them, but it is always at their service, and at their
command will sting a companion with whom they are
dissatisfied."^
Xolotl is the accompanying god in the tonalamatl,
where otherwise Quetzalcouatl or Ueuecoyotl are rep-
resented. The center of the tonalamatl thus fre-
quently represents the sun, which is only natural, since
the Arabic qimdr, the game of chance, is also applied to
the sun and moon, as the chief representatives of the
âdwal zodiac. In Arabic we have ^' al-qamaru "the
moon," ly'-f^' al-qamarani "the sun and moon," but
Xolotl is for some reason connected with a cooking-pot.
This is due to an Arabic philological calculation, be-
cause J^' aSdr means "a cooking-pot that boils the
ten portions of a camel slaughtered for sacrifice."
Xolotl is generally confused with or turned into
Nanauatzin, whose glyph is the same cooking-pot with
human bodies boiling in it.^ Nanauatzin is supposed
to be the god of syphilis, his name being derived from
nanauatl "syTphilis." But this is a mistake: nanauatl
is translated by the early writers as "bubas," and this
means "pustules, itch, " which shows that we are dealing
' Henry, op. dt., p. 98.
^ Seler, Codex Borgia, vol. I, p. 287.
in tlie early Mexican accounts of nanauatl with the

same disease, which in Hispaniola was given as cara-
caracol,^ although there can be no doubt that during the
Spanish occupation nanauatl was also the name of
syphilis. It must be observed that Bambara many a
means not only "the itch," but also "syphilis,"^ hence
it may be that syphilis was actually introduced into
America by the Negroes before the landing of Columbus.
Sahagun tells the following story of Nanauatzin:
" This is the way the moon began to shed light upon the
world. It is said that before day existed the gods
united in the place called Teotiuacan, and said to each
other, 'Who will undertake to enlighten the world?'
To which a god called Tecuciztecatl replied, 'I will do
so.' The gods spoke a second time and said, 'And
who will be the second?' They looked at each other,
trying to find out who it would be, and no one wanted
to offer himself for the undertaking; they were all
afraid and excused themselves. On of them, whom
they did not consider and who had the bubas, did not
speak and listened to what the others were saying.
They finally turned to him and said, 'You be it,
bubosito. He listened gladly to what they ordered him
'
to do and answered, I receive your order as an act of

'
grace, —
be it so. The two chosen ones immediately
'
began a four days penance. Then they lit a fire in the

'
hearth made in a rock, which even now has the name of

Teotexcalli, rock of the gods. The god called Tecuciz-
tecatl offered up only precious things, for in place of
bouquets he made offerings of rich feathers called
quetzalli; instead of straw wisps he offered gold balls;
thorns made from precious stones in place of maguey
thorns, and thorns of red coral in place of bloody thorns.
Besides, the copal used in the offering was of the best.
1 See my Africa, vol. I, pp. 76, 78, 160.

2 Henry, op. cit., p. 47.
The buboso, called Nanauatzin, offered nine green reeds
tied in threes, in place of the usual branches. He
offered straw wisps and maguey thorns reddened with
his own blood, and in place of copal he offered scabs of
his own bubas.
"They built a tower in the form of a hill for each of
the two gods. There they did penance for four days
and four nights. These hills are at present called
Tzaqualli,— they are found near the village of San-
Juan, called Teotiuacan. After the four nights of
penance were passed, they threw all about the place
the branches, bouquets, and all other objects which
they had used. The next night, a little after mid-
night, when the service was to begin, they brought
the ornaments to Tecuciztecatl; they consisted of the
feathers called aztacomitl and a jacket of light cloth,
while the head of Nanauatzin, the buboso, was covered
with a paper cap called amatzontli, and he was dressed
in a vestment and girdle also of paper. When mid-
night had come, all the gods placed themselves around
the hearth called teotexcalli, where the fire had been
burning for four days. They divided themselves in
two rows and placed themselves on both sides of the
fire.' The two chosen ones seated themselves in front
of the hearth, with their faces turned to the fire, be-
tween the two rows of the gods, who stood up and, ad-
dressing themselves to Tecuciztecatl, said to him, 'Come
now, Tecuciztecatl, jump into the fire.' He tried, in-
deed, to do so, but, as the hearth was very large and
hot, he became frightened, as soon as he felt the heat,
and withdrew. For a second time he took courage and
was about to leap into the hearth, but as he approached
it, he stopped and did not dare to do so. In vain did he
make four separate attempts, but it was ordered that
no one could try more than four times. When these
four attempts had been made, the gods turned to
Nanauatzin and said to him, 'Come now, Nanauatzin,

it is your turn to try. 'Scarcely had they said this to
him, when he collected his strength, closed his eyes,
leaped forward, and jumped into the fire. He began
at once to crackle, like a thing roasting. Tecuciztecatl,
seeing that he jumped into the fire and was burning,
also dashed forward and threw himself upon the coals.
It is said that an eagle that entered the fire at the same
time burned himself, and so this bird now has blackish
wings. A tiger followed him, but he did not burn him-
self; he only singed himself, and so he remained
speckled white and black. From this legend arose the
custom of calling men expert in the use of arms quauhtli,
ocelotl. They say quauhtli, because the eagle was the
first to enter, then ocelotl, because the tiger was the
second to rush into it after the eagle.
"After the two divinities had thrown themselves
into the fire, and were consumed in it, the other gods
sat down, thinking that Nanauatzin would not fail to
arise. They had waited for a long time, when the sun
began to glow, and the dawn was seen to arise. The
gods, they say, fell upon their knees to wait for Nanau-
atzin to turn into the sun, without knowing where it
would arise. They turned their glances in all direc-
tions, but they could not tell where its rising would
take place. Some thought that it would be to the
north, and they turned their eyes there. Others
thought that it would be to the south. In fact, their
suspicions were directed everywhere, because the dawn
glowed in all directions. Some fixed their attention
upon the east, and insisted that the sun would rise
there. This was the correct view. Those who waited
for that region, they say, were Quetzalcouatl, also called
Ecatl; Totec, who also bears the names of Anaoatlytecu
and Tlatlauic Tezcatlipoca; others, who are called
Mimizcoa, who are innumerable, and four women, of
whom the first is called Tiacapan, the second Teicu, the
third Tlacoeua, When the
and the fourth Xocoyotl.
sun was about to looked very red, tottering from
rise, it
side to side, and nobody could fix his gaze upon it,
because it blinded him, so bright was it with its rays
that escaped from it and spread in all directions. The
moon arose at the same time, and also in the east, that is,
at first the sun and then the moon, in the same order as
before at the hearth. Those who tell these stories say
that the sun and moon had then the same light, and
that the gods, perceiving this equality of splendor,
conversed with each other once more and said, 'O gods,
how can this be? Will it be all right if they are both
equal and enlighten the world in the same manner?'
And then they uttered words and said, 'Let it be as
it may. ' . .And immediately one of them began to
. .
run and struck with a rabbit the face of Tecuciztecatl,

who at once became brown, lost his splendor, and as-
sumed the form which we all know to-day. When the
sun and moon had risen upon the earth, they both re-
mained motionless. Then the gods spoke once more to
each other and said, How can we live this way? The
'
sun does not move. Are we to pass all our lives among
unworthy mortals? Let us all die and let our death
give life to the luminaries. The wind then undertook
'
to cause the death of the gods and killed them. The

one called Xolotl refused, they say, to die, and he said
to the gods, 'Gods, I do not want to die!' And he cried
so that his eyes were swollen. When the one who
caused the massacre came to him, he fled and hid in a
field of maize, where he was changed into a stalk of this
plant with two ears, which the farmers call xolotl. But,
as he was recognized among the maize, he fled a second
time and hid in the magueys, where he was changed
into a double maguey, which is called mexolotl. When
he was once more discovered, he fled again and threw
.Mil.. r,T .V,/„«/.-, ,-.:.„. <t.r >v|.liiljiik.-!-. .Ii. .Mill, IV I-,.ii,i Xoh.if^. ,U~. (;,,n<-
.lii y.uitlirn;'- iiivi .li-i- Mi<-L'''i'iii-ti,'ri ('..li.x l!iiii:ia In : : F\itiu'~li"r'i'iL:h l".'i.
NANAUATZIN, THE MEXICAN GOD OF SYPHILIS,

from Selcr's Codex Borgia.
himself into the water, where he was changed into a fish

called axolotl.
"There they seized him and killed him. But, al-
though all the gods were killed, the sun did not get into
motion. Then the wind began to blow and to rush
violently, which caused the luminary to move and start
upon its journey; but the moon stayed still where it
was. It began to move only after the sun had done so.
Then only they separated from each other and acquired
the habit of rising at different hours. The sun lasts a
whole day, and the moon shines and works during the
night. They said justly that Tecuciztecatl would have
been the sun, if he had been the first to throw himself
into the fixe, because he was the first to be shaped and
because his offering had consisted of precious things."^
I have given the whole cosmogonic story in order to
show the conflate manner of a Mexican myth. The
constituent parts can be easily reconstructed with the
aid of the Mandingo myths, even though in this case the
information is unusually scanty. The Malinkes call
the itch liana, and a person suffering from the itch is
called nanato. This nana, that is, nganya, is unques-
tionably derived from Arabic lH nagal "corrupted,
irritated,
'
' ^ nagila
'
' the itch,
'
' which is very likely also
the origin of French gale " the itch. " In Bambara this
has become manya. But we have also Malinke maria,
Bambara many an "black, stinging ant, whose stings
are painful," Mandingo menango, merango, melango
"ant." These "ant" words are most likely related to
the "itch" words, since the Arabic word also means
"creeping worms." In any case, in the Bambara the
two concepts are connected, as meaning "something
that itches, stings. " But Bambara many an also means
"fetish whose pyramidal altar may be seen near the
» Sahagun, op. dt., p. 478 B.
villages."^ Unfortunately we have but the scantiest
reference to this fetish by Henry,^ where he speaks of a
round tower used as an altar and surmounted by a bowl
to receive the victims' blood, which is dedicated to
Manya and is always placed under a ceiba tree. Among
the Mayas the ceiba, called yaxche, was the tree which
represented heaven and was the abode of the blessed.^
The philological and semantic identity of Malinke
nanato with Nahuatl Nanauatzin "the buboso" is
obvious. Both have a pyramidal altar, called in
Nahuatl tzaqualli, in the Western Sudan, in Hausa
dakali "pagan altar, raised clay bed, seat," from Arabic
cP^ dakl "to knead clay." Not only do we have

Nanauatzin as jumping into the fire, but in the Codex
Borgia Xolotl is represented as Nanauatzin sitting in the
cooking-pot.* This makes it necessary to connect
Nanauatzin, on the one hand with ants, and on the other
with twins. Here again the fragmentary Sudanic
tradition solves the question completely.
In southern Nigeria twins are in many locaUties
killed. "Although the destruction of twins is not, in
the strict sense of the word, a sacrifice, it is all the
same a sacrificial offering, very much in the same light
as the purification ceremonies which have been just
described. For the custom, based on the identical
spiritual principle of an evil react, is treated as one of
offence against the ancestral gods that must of necessity
be removed, along with the offending cause the woman. —
"As I have already pointed out with regardto human
sacrifice, this a purely religious custom, the origin
too is
of which is lost in antiquity, and due apparently to the
conception that one birth at a time is the distinguishing
' p. Sauvant, Grammaire bambara, Maison-Carr6e (Alger) 1913, p. 134.
2 Op. dt., p. 148.
^ Seler, Codex Borgia, vol. I, p. 316.
* Ibid., vol. II, p. 53.
feature between man and all other creation, and there-

fore the birth of twins was regarded as an unnatural
event, to be ascribed solely to the influence of malign
spirits, acting in conjunction with the power of evil.
And the custom has been tenaciously adhered to, in
spite of the fact that every child born into a family,
apart from all other human considerations, has a
monetary and a practical value attached to it.
"Indeed, according to their ancient faith, although
two energies are requisite to produce a unit, the pro-
duction of two such units is out of the common groove,
therefore unnatural, because it implies at once a spirit
duality, or enforced possession by some intruding and
malignant demon, in the yielding and offending person
of a member of the household, consequently an outrage
committed upon the domestic sanctity. For in their
opinion, the natural product of two human energies, as
a single unit, is only endowed, or provided with, one
soul-spirit. The custom that prevails among the Ibo
—
and Brassmen of allowing one always the first-born of
—
the twins to live, is a practical admission of this
conception.
" The custom is universal throughout the Delta, and is
only dying out in those few localities in which the people
are actually in touch with civilisation. The advent of
twins is looked on in every home of the Delta not only
with horror and detestation, but as an evil and a curse
that is bound to provoke the domestic gods to anger
and retribution. In order, therefore, to avert the
expected vengeance, it is the standing law of the priests
that no time is to be lost in at once removing the un-
fortunate infants. This is generally done by throwing
them into the bush, to be devoured by wild animals, or
the equally ferocious driver ants, or sometimes, as is
done by the Ibibio, Ijo, and other coast tribes, by setting
them adrift in the rivers and creeks in roughly made
baskets of reeds and bulrushes, when they are soon
drowned, or swallowed by sharks or crocodiles.
"In most cases the mothers, who are looked on as
unclean, are driven out of the town and into the bush,
and unless given protection by the people of another
community, or surreptitiously fed by some old crony,
they often fare as badly as their offspring, whom they
look upon as the work of evil spirits.
"In some cases, however, humaner treatment is
accorded to them. In Ibani, for instance, it was
customary, as it now is among the other middlemen, to
quarantine the unclean mothers in an out-of-the-way
hut for a period of sixteen days. Here neither man,
woman, nor child dare visit them, with the exception of
certain old women who were specially set apart to tend
and provide them with food, water, and other necessary
requirements. At the end of this time they were
brought out and obliged to undergo the ceremony of
purification, at the hands of the priests, which, in addi-
tion to washing off the chalk that had previously been
smeared all over their bodies, consisted of the sacrifice
of a chicken, or a new-born pup. Besides this, the
father, or in the case of a slave or poor member of a
family, the head of the house, was also obliged to avert
the wrath of the enraged deity and the consequences
that were to be expected, by offering special sacrifices
—
and presenting gifts to the priests an undertaking
which, as a rule, implied a minimum outlay of at least
1600 manillas, equal in those days to about £6: 13s., or
less or more, according to existing rates. Purified and
once more clean and free from evil, the women were
received back into the family circle, and the threatened
evils were considered to be averted.
"In the Ibibio country, and formerly among the
Efik, the regulations with regard to the women are
much more elaborate, and in a certain sense humane.
" Here, as invariably in all similar cases, the ancestral

gods are propitiated by gifts and sacrifices, but the
women, looked on as unclean for the rest of their lives,
are obliged to reside in villages, which are known as
Twin Towns, or the habitations of defiled wom.en, ap-
pointed for that particular purpose. From this time
forth the husband, whether he be head of the house or
not, is obliged to maintain a wife who has been so de-
filed; although at the same time he is strictly forbidden
to cohabit or to have any dealings with her, being, as
he is in every religious and personal sense, human and
spiritual, divorced from her. But in spite of the fact
that to him, as well as to all the members of his or her
community, the woman is unclean and therefore tabu,
the penalty of death being inflicted on both in the event
of their breaking the law in this direction, she is allowed
to form connections, but on no account to marry with
strangers, or men belonging to outside communities, and
the offspring resulting from such intercourse becomes,
as a matter of course, the property of her husband, or
the head of the house.
"In order to remove the child from the defiled locality,
which cannot, however, be done until it is weaned, i.e.
when from two to three years old, a special sacrifice of
chickens and fowls must first be made. Sacrifice, in
fact, is imperative and inevitable in all cases in which
intercommunication is necessary, and an interchange of
visits is made between all members of the households in
question and the defiled women. Thus, for example,
when it is obligatory on certain occasions for any near
relatives or others of either sex to visit one of these
women, the visitors are compelled to sacrifice fowls or
goats to the domestic deities, so that the act of contact
may not be productive of the evil effect of twins, in any
subsequent issues of children, on the part of female
visitors; and on exactly the same conditions defiled
women are permitted to visit relatives, also to work for
their husbands.
"But in the event of the defiled woman herself
bearing twins again, these must be destroyed unknown
to any one. For, if known, the probabilities are that
the death of the mother would be demanded by the
household and the community as well. Or if not killed,
she would be driven into the bush and left to die, al-
though, if discovered by a stranger, he is at liberty to
claim her as his own property, that is, at least, if he
feels inclined to run the risk of a venture so truly pro-
vocative of offence.
"Among the various clans of the Ibo, when the birth
of twins takes place, the people belonging to the
quarter in which the mother resides are obliged to
throw away all the half-burnt firewood, the food cooked,
and the water brought in the previous night every- —
thing, in a word, in the shape of nourishment, solid or
liquid, because the advent of the unholy twins defiles the
house and practically all its contents. To purify the
place from this unwelcome pollution, the inevitable
sacrifice, consisting in this case of fowls and goats, is
there and then performed, and the unclean mother is
at once removed from the house and town. Indeed,
as soon as a pregnant woman is delivered of a child, and
it is known that another is to follow, she is instantly
carried into the bush, and when the second is born it
is immediately thrown away, while the first-born is re-
tained, and named M 'meabo, which means two people.
"If it happens also that during childbirth the infant
—
comes out of the womb feet foremost the event which
is referred to as Mkporo-oko, i.e. bad or evil feet it —
is regarded in the same light as twin-birth, and the un-
fortunate mother is accorded exactly the same treat-
ment, her eventual destination in either case being a
Twin Town.
"The Ibo customs are, however, practically identical

with those of the Ibibio, Ijo, and other tribes, except
for one or two trifling differences. For example, in the
event of the defiled woman bearing issue by a stranger,
the children, although the property of the husband,
must be maintained by the natural father, who is
obliged to pay over the legitimate expenses to the
former. The women and the children are, however,
placed under certain laws or restrictions, the use of
certain trade markets and roads being prohibited to
them, but they are permitted to have a market of their
own."i
"All deformed children are not human; they are
devils, and their influence is evil. They must be killed.
These devils' usual dwelling-place is in the bush and
they annoy travellers by night by 'throwing stones at
them.' Some men assert that often a woman who
gathers herself a new dress, i.e., leaves, from a bush
which is inhabited by a devil seizes the devil together
with the leaves, and in this way he has intercourse with
her, with the result that a devil-child is born, i.e., a
deformed baby. Apparently the women do not be-
lieve this —
they say that some men are devils them-
selves.
" As to whether one 's child is a devil or a human being
only the sorcerer can say. I will explain later how the
information is imparted. Almost invariably a de-
formed child is a devil; twins sometimes are, and
now and then a quite healthy child is so proclaimed.
Once the father is certain that his wife has brought
forth a devil, he proceeds to the devil-killer, who returns
with him to the house where the child is. There he
receives a red-and-black hen and a goat, and gives in
return the devil-killing medicine to the child and ties
round its neck a ram's horn flUed with a powder of
1 A. G. Leonard, The Lower Niger and Us Tribes, London 1906, p. 458 ff.
earth, shea-butter and ashes. The child soon after
dies and the killer is called back to bury the corpse.
This he places in a large water-pot, and the father
carries it into the bush, where, finding an ant-heap,
he buries the pot and its contents."^
"A former resident of Onitsha on the Lower Niger,
informs me 'Twins are objected to and both are killed
in the Ibo country. The killing is brought about by
placing the children in a large earthen pot which is
then carried into a part of the bush which is tabued
and called tonton. The children are soon killed by the
ants and other flesh devouring insects. Passing Euro-
peans hearing the wailing of the children have carried
them off, but in most cases the exposure had already
been too much and they succumbed. A child so res-
cued and surviving would, if a girl, experience difficulty
in getting married for fear she have twins. Bishop
Hill in traversing a piece of bush which the natives had
made over to him for missionary purposes, found a
tonton with over 200 pots which had contained babies.
The natives would not assist to clear the spot.' "^
"Infanticide of a peculiar nature likewise prevails
among them twins are never allowed to live. As soon
:
as they are born, they are put into two earthen pots,
and exposed to the beasts of the forest; and the un-
fortunate mother ever afterward endures great trouble
and hardships. A small tent is built for her in the
forest, in which she is obliged to dwell, and to undergo
many ceremonies for her purification. She is separated
from all society for a considerable time; her conjugal
alliance with her husband is for ever dissolved; and she
is never again permitted to sit down with other women
in the same market or in the same house. To give
birth to twins is, therefore, considered to be the greatest
'A. W. Cardinall, The Natives of the Northern Territories of the Gold Coast.,
London, New York, p. 27 f.
2 H. L. Roth, Great Benin, its Customs, etc., Halifax 1903, p. 36.
misfortune that can befall a woman of the Ibo nation.

If any person wishes to annoy an Ibo woman, he lifts up
two fingers, and says, You gave birth to twins;' which
'
is sure to make her almost mad.

"^
The specific exposure of twins in a cooking-pot can
only be due to the same philological connection which
brought about the representation of Xolotl with a cook-
ing-pot, namely the Arabic homonyms which mean both
"comrade" and "cooking-pot," and this led in Mexico
to the burning of Nanauatzin and Tecuciztecatl in the
fire, while otherwise Nanauatzin is represented in a
cooking-pot. The Mande homonyms for "ant" and
"itch" led to the correlation of Xolotl with the buboso.
The Arabic conception of qamaran as "sun and moon"
led to the Mexican correlation of Xolotl with Tecucizte-
catl, and the usual connection of the "ant-god" with
the pyramidal altar among the Mandes led to the same
pyramidal altar to Nanauatzin and Tecuciztecatl in
Mexico.
We are unfortunately badly informed as to the in-
visible triune gods of the Mandingos. Delafosse says:
"The spirits often form a veritable mythological
family, at the base of which one generally meets Heaven,
male spirit and fecundating principle, and Earth,
female spirit and fecundated, generating principle.
The Heaven, at times identified with the Sun, has
married the Earth, at times identified with the Moon,
and from their union or from the union of their children
have arisen all the chief spirits which direct the world
and there dispense life and death, happiness and un-
happiness in forms.
all their
"
Heaven and Earth, parents of the spirits, are often
called upon as witnesses in oaths or invoked in wishes;
1
J F. Schon and Crowther, Journals of the Expedition up the Niger,
S.
London See also W. Allen and T. R. H. Thomson, A Narrative
1842, p. 49 f .
of the Expedition to the River Niger, London 1848, vol. I, p. 243, and J. R.
Harris, The Cult of the Heavenly Twins, Cambridge 1906, chaps. I and II.
but the cult directly rendered them is mucli less wide-
spread than the one rendered to their eldest child, who,
endowed both with the male and female virtues of its
progenitors, is the true intermediary between the
mysterious power of God and the timid weakness of man.
"Generally Heaven and Earth remain entities, if not
abstract, at least without any palpable representation;
sometimes, however, the first is invoked under the form
of a man provided with an enormous phallus, or even
under the form of an isolated phallus, while Earth, or
the female spirit, is represented by a woman with large
breasts or simply by a pair of breasts.
"As to the eldest son of Earth, he is represented under
various aspects, sometimes as a hermaphrodite, more
often as an animal figure, as the head of an ox, crocodile,
fish, or serpent. The cult of this spirit, under the var-
ious aspects of his external representation, is, I believe,
common to all the non-Islamized populations of West
Africa, and it exists, in a reduced form, even among
several Islamized peoples. It is found among all the
Mande tribes, under the name of Koma or Komo, and,
under the name of Do, among the Senufu, the Agni-
Asante, etc.;elsewhere he is known under different
names, but, whatever may be the name given to this
spirit, his cult is found everywhere, from the Senegal to
the Congo and, no doubt, beyond, with quite analogous
external ceremonies. Everywhere, too, these cere-
monies are forbidden to women, and certain of their
rites are hidden also from non-initiated men; the ini-
tiation to the cult demands a whole series of tests which
are surrounded with mystery and which they do not
like to reveal to strangers; the religious association
which has for its principal aim the cult of this spirit is
one of the most widely spread and most firmly consti-
tuted in western Africa."^
' Delafosse, Haut-Sinegal-Niger, vol. Ill, p. 173 ff.
—
It seems that Delafosse has confused two separate

worships, that of the Komo with that of the invisible
triune divinities. We are better informed on the latter
in the Mossi country: " The Mossi have the idea of the
One God. One finds among them two beliefs accord- :
ing to one God is the Sun; according to the other, he is

a material being which they cannot represent. They
have borrowed from the Moslems the idea that God has
created everything, even the Sun, which is only fire.
This fire, if it were left by God constantly at liberty,
would consume everything, so he built a house where
—
nine of his children (malekdamba, this is the Arabic
word for 'angel' malakum) keep it shut up at night,
thus producing night. This god is well materialized,
since he has a wife {tinga, which also means earth) and a
child, and eats and drinks (but not dolo, which is re-
served for his followers). This infinite being, all-
powerful Naba, is obviously surrounded by a consider-
able number of pages or good spirits, but there are also
others, bad ones, who run away from him {djidamba, in
Arabic djinun) and are the cause of all bad acts. We
find in all this exposition a reduced and naive Moslem
theology."^
Among the Habbes the highest triune divinity is
called Ammo or Amma: "To this triune divinity they
raise three-pointed altars from cut stone, upon which
the religious chiefs, called Hogon, offer their sacrifices."^
"The Divine Force which they invoke is called, as at
Timbuktu, Harkoy or Herkoy, 'Chief of the Males, and '
the altars raised to this celestial fecundating power are

built of clay, in the form of cones and painted red.
These are sometimes nothing but conical stones or
monolithic pieces put up vertically in the yard of every
1 E. Ruelle, Notes anthropologiques, ethnographiqiies et sociologiques sur
quelques populations noires du 2e ierritoire militaire de I'Afrique Occidentale

Franeaise, in U Anthropologie, vol. XV, p. 689 f.
2 Desplagnes, op. cit., p. 269.
family house. In all these families the decoration of
the pillars supporting the verandas of public places and
of beams sustaining the ceilings of the chambers, and
the ornamentation of the front walls of houses agree
with the conventionalized designs of the tana animals of
the confederacy, of conical pilasters and innumerable
phallic emblems like the decorative motifs which crown
the door-fronts in the Djenne houses."^
The coincidence of the corresponding Mexican belief
with the African triune divinity is most striking. "The
god of fire, called Xiuhtecutli (master of the stars), has
also two other names, one of which is Ixcogauhqui, which
means 'the one with the yellow face,' and the other
Cuegaltzin, which signifies flame. '
He was also called
'
Ueueteotl the old god, and everybody considered Fire

'
'
to be his father, in consideration of the effects produced

by him, because he burns, and because his flame shines
and burns. "^ He is also identified with Tonacatecutli,
"the lord of our flesh, "^ "God, lord, creator, governor
of everything, —
all these names were referred to him,
since he was the god of whom it was said that he had
created the world, and so he was represented with a
royal crown upon his head. "* No sacrifices were made
to this god, because he did not want them.^ "The
same interpreter says in the notes to Codex Vaticanus A
(No. 3738) Tonacatlecotle, which means lord of ouj*
:
'
bodies, while others say that he was called the first man,
and possibly it is intended that the first man be so
called. This is the image of the first lord that the
. . .
world has had and who, when it pleased him, blew and
separated the waters from the heavens and earth, which
formerly had been mixed up, and it is he who put them
1 270 f.
Ibid., p.
" Sahagun, op. cit., p. 27 f.
' Seler, Codex Borgia, vol. I, p. 116f.
* Codex Telleriano Remensis, vol. VIII, in E. K. Kingsborough, Antiquities
of Mexico, London 1830, vol. II, p. 1.

s
Ibid.
Fig. tlB. — Aulcl il ti-llis poinlcs rlr\i: ,i l;i Tll.l'lr Cilr^lc
ilii villaRO (I'KnKiMii-fiijiinini.
AFRICAN THREE-POINTED ALTAR, from Desplagnes'

Le plati-au cenirul ni'âit'n.
AFRICAN THREE-POINTED ALTAR, from Desplagnes'
Le plateau central iiigerieî.
. "
in order, as they are now, and so they called him the

lord of our bodies and lord of abundance, who gave
them all things, and so they represented him alone with
the royal crown. He was also called Seven Flowers
(Chicome xochitl), because they said that he divided the
principalities of the world. He had no temple whatso-
ever, nor were any sacrifices made to him, because they
say that he did not want them, as it were for his greater
glory. He was called Tonacatecotle and otherwise
. . .
Citallatonali, and they say that he is the sign that ap-

pears at night upon the sky, which by the people is
called Via San Giacomo or The Milky Way.'"^ Ton-
acatecutli is frequently represented with his wife Ton-
acaciuatl,^ and the two are frequently called Ometecutli,
Omeciuatl, who live in the thirteenth heaven and are
the gods of generation.^ Ometecutli means "the lord of
two, " but this is mere popular etymology, since we have
the same Life-giving principle in Africa as Ammo, and in
the Mandingo language similar words refer to the life-
giving principle. We
have Malinke lamo, namo "to
ripen, bring up, raise, educate, " mo " to be ripe, brought
up, raised, educated, " Bambara mo "ripe, to grow up,
lamo "to ripen, raise, nourish, educate, take care of,"
Soso md "to grow, to grow old," mdkhi "ripe." No
doubt, all these are derived from Arabic >- namu "to
produce, procreate, give birth," and we have also V^'
(al-nafs) al-ndmiyah "the vegetative soul, which makes
the plants grow."
While Ruelle may be right that the idea of the One
God, creator of all things, was by the Mossi, hence by
the other Sudanese people, derived from the Arabs, the
triune divinity, to whom no temples are raised and for
whom there is no visible worship, is unquestionably of
> Seler, Codex Borgia, vol. I, p. 78 f
2 IMd., vol. II, p. 41.
' Ibid., vol. I, pp. 32, 55 et passim.
Christian origin that has percolated through an Arabic
source into the Sudan. This is again made clear from
the worship of the god of fire, that is, the triune divinity,
by three hearth-stones,^ just as in Africa we have the
three-pronged stone altar. Among the Mayas the re-
lation of the divinity to the Christian religion was ob-
vious to the priest, Hernandez, who tells of the one god
in heaven, who was father, son, and holy ghost. The
father was called Iqona (Itzamna), and he created men
and all things; the son was called Bacab, who was born
of a virgin called Chibirias (Ix chebel yax), who lived in
heaven; the holy ghost they called Echuac (Ekchuuah).^
If the Mexicans and Mayas received their triune di-
vinity and the sun god from the Mandingos, they also
received from them such notions as are perpetuated in
their calendars. We find in Mexico and Central Amer-
ica, as among the Mossi in Africa, the Nine Lords of the
Night, that is, the regents of the hours of the night
mentioned upon their calendars, and the presiding divin-
ities, of which the first is Xiuhtecutli, the fire god, are
identical with nine of the day divinities, and the night
hieroglyphs are the same as those of the day.^ This
puts the whole Mexican calendar, in so far as the hoxirs
are concerned, into a very recent period, and there is no
reason for assuming that the calculation of the days and
the divisions into years are any older.
It has sometimes been supposed that Mexican teotl
may have something to do with the Latin or Spanish
-word for "god," but such is not the case. To begin
with, Seler has shown that teotl, in the narrower sense, is
equal to Tezcatlipoca, the evening sun god: "In Mexico,
according to Duran, there was a confraternity consisting
exclusively of people of noble or royal origin, who con-
secrated a special cult to the sun. But among the
^ Ibid., pp. 30, 116, 124, 322.
2 Ibid., p. 322.
5 Ibid., p. 218 B.
historical tribes there in fact, no mention of any that

is,
worshipped the sun as a tribal or national god. Only

of the old Teotiuacan, the cultural locality long deserted
in historical times, do we hear that the old tribes there
had built a pyramid each for the sun and moon. And
—
the name Teotiuacan which, it is true, in the text is ex-
plained differently from 'place where the dead kings
—
were turned into gods' may, perhaps, be translated as
'place of the sun- worshippers,' for under teotl, 'god'
par excellence, is to be understood the sun. Teotl ac
'god has entered the earth,' the Mexicans said for
'sundown.' And wherever in city glyphs the syllable
teo- was to be brought out, the Mexicans represented a
whole or a half of a sun disc. Of late there have been
those who went so far as to express the belief that or-
iginally the sun played no part in the Mexican cult, and
that only the custom of sacrificing a demon of growth at
an important place of the solar course, in order that he
'
may provoke the heat necessary for the year's crops,'

has led to the development of the vegetation demon
into a sun god, and that the latter ultimately gained the
upper hand over the terrestrial and vegetation demons,
the types of primitive culture.
"Against such an explanation is the circumstance
that, as I have already indicated, among the Mexicans
the concepts of teotl 'god' and 'sun' are interchange-
able, and that correspondingly the dead kings and war-
riors, who went to the sun, themselves became suns
{in aquin oonmic oteut). And the remarks, such as that
of the interpreter of the Codex Telleriano Remensis,
—
'
they said that it was the sun which produced all things,
and, therefore, the maize was called tonacayotl, that is,
it was produced by the sun (todas las cosas dizen que la
produce el sol y ansi dizen al tlaule tonacayotl que quiere
dezir, ya se criava del sol)
'
—
in spite of the mistake
which is obviously contained in the etymological in-
terpretation, point to a definite conception of depend-
ence on this celestial luminary, and in this dependence
and fear does the essence of religion consist. Then the
other remark of the interpreter of Codex Vaticanus A
in another place, where the picture writing represents
the sun god opposite the death god, that this juxtaposi-
tion meant, winter is so disagreeable on account of the
'
absence of the sun, and summer so agreeable on account

of the presence of the sun, and the return of the sun to
the zenith means that this their god comes to show them
a favor,' and referred to a definite annual festival, the
feast of Toxcatl as a sun festival."^ Sahagun says:
"The god called TezcatUpoca (smoking mirror) was
considered a true and invisible god who penetrated all
places, heaven, earth, and hell. "^ There can, therefore,
be no doubt that we are here dealing with a triune god,
but he represents more nearly the setting sun or the
rising moon.^
Before identifying him with a Mandingo conception,
we shall begin ab ovo, with the Arabic conception, from
which the Mexican religious ideas and the very name
ultimately are derived. We have Arabic J^ zil" shade,
i.e.,the light of the sun without the rays; anything
that shades one; an apparition, phantom, or a thing
that one sees Uke a shadow, the jinn; God's means
of protection; a state of life ample in its means or
circumstances, unstraightened or plentiful and easy,
;
pleasant, soft, delicate; paradise." But we have also

"^ "a thing that shades one from the sun, a
zullah
cloud," and J:^ zalil "a collection of trees tangled,
or luxuriant, or abundant and dense." This root has
entered into the Berber languages, where we have Beni,
1 Ibid., p. 205.
2 Op. cit, p. 14.
» Seler, Codex Borgia, vol. II, pp. 12, 23, 32, 35, 57, 63, 298.
Menacer, Bougie thili, Bot'iua thiri, Zenaga tiji

"shade, image. " This has led to Bambara dya "spirit,
sense, image, tangled woods, obstruction, trap, net,
dryness," diya, dia "pleasant, good, agreeable, for-
tunate, successful, hospitable," and Malinke dya
"spirit, faculty, memory, image, hospitality, to become
warm or dry," dia "ease, contentment, joy."
"Dya is a difficult word to translate. We translate
by dya the words 'phantoms, images, photographs,
shades, reflections in the water, mirror.' But it has
also other acceptations, and in the phrases a dya tikera,
a dya ulila, misiu dya ulila it is not a question of the
shade or image, for they mean : 'He is afraid, he is mad,
the cows are mad.' In the eyes of a Bambara, the
dya is not only the shade, image, contour of an object,
it is also that which produces it. It is a force, a power
common to beings from all kingdoms, and although it
cannot be compared to a jinn, the soul, there is none
the less in his mind a kind of being which he cannot de-
fine, but which exists and can act by itself. The
Bambaras are convinced that a man can take away his
dya, and then it will not cast a shadow. If the people,
out of fear of the European, and several, because they
have been for a long time in contact with him, readily
pose before the camera, we may none the less say
they in general do not like to be photographed.
Especially the old people make difficulties and are recal-
citrant the moment they see a lens. They are obsessed
by the idea that their dya will escape from them, in
order to impress itself upon the plate and leave there
its form, the portrait, and that thus a part of their being
will be in the hands of the photographer, and that they
will be at his mercy. They are afraid that in the
photographer's retaining them they will lose their
shadow and will henceforth be unable to cast it, hence
become idiots."^
will
If we now turn to Tezcatlipoca, we see that he is "the
smoking mirror," that is, the mirror that gives shade,
but tezcatl "mirror" is unquestionably the Arabic

C^-J zgag, Jt^J zagangal "glass, mirror," which we
find in Hausa as tsokachi "to look, gaze," matsokachi,
Songay didji, digi " glass, mirror, " and we have Malinke
dungari, Bambara dungare "glass, mirror."
dugare,
Whether the Nahuatl the Arabic word or not,
tezcatl is
there cannot be the slightest doubt that in Tezcatlipoca
we have the setting sun, teotl, which is our Mandingo
dya, hence the relation that Tezcatlipoca bears to the
underworld, the realm of the dead. Hence teotl is
etymologically derived from Mande dya.
The investigation of the Mandingo influence among
the Mayas and other non-Aztec tribes becomes in-
creasingly more difficult because of the unsatisfactory
information we have of the religious development out-
side the direct Aztec influence. In Peru the Mandingo
influence is as obvious as elsewhere, but here the
documentary evidence we have upon religious practices
is negligible, while the absence of written records of a
type contained in the Mexican and Mayan manuscripts
makes it impossible ever to reconstruct in full the
Peruvian antiquity.
Like the Mexicans and Mandingos, the Peruvians
worshipped the invisible Trinity. The monks who in
1550 began the investigation of the Peruvian idolatry
wrote as follows: "We asked the priests what they
thought of God, whom they worshipped, and they
said, Ataguju, creator of all things, whom they con-
sidered to be the principal end, according to their re-
ligion. And they said that he was in heaven and did
^ Henry, op. eit., p. 41 f.
not move away from there, but that he governed all

the things from there and created them, and they say
that he created heaven and earth, and rules them from
there, and, seeing himself alone, he created two others
(they say cross or to do' for this verb, ruram, which
'
'
'
means 'to do') so that there should be three, and all

should have one will and power, and they had no wives
and were equal in all things. The devil, who is like
the xamua of God, told them this and this most false
Trinity."^ From other sources we learn that they
called this invisible chief god Ilia Tecce. For a long
time no temple was bxiilt to him, and later only one
temple was erected to him. The eldest son of this first
cause was Inti, the sun god, and his sister and wife was
Quilla, the moon goddess. There were a number of
secondary gods attached to these, and many of them
bear the name of ilia, such as Illjapa or Catoylla, the
god of lightning and thunder, also called Chucuylla;
then there was Auki-ylla, the prince of glory, Llari Ylla,
and others.
We
shall try to ascertain the exact meaning of Ilia.
This was also the name of an idol or fetish: "When
they capture some wild animal, they look into its
stomach and if they find in it a stone or hardened ex-
crement, they keep it in a pouch and worship it and
sacrifice to it the blood of the coy. Only lately the devil
has taught them, since the arrival of the Spaniards in
Peru and since they have goats, to look into their
stomachs, when they kill them, for a small ball of dry
herbs, and this they keep in their houses and worship it,
so that their goats should multiply. This they call
Illa-cabra, and offer to it the blood of the coy, and
'"^
celebrate its holidays like other holidays.
' Coleccidn de documentos in6ditos relativos al descubrimiento, conûista y
colonieaddn de las posesiones espanolas en AmMca y Occeania, M&dnd 1865,
vol. Ill, p. 13 f.
2 lUd., p. 50.
Here we see that the bezoar-stone of the Arabic medi-
cine has become a fetish. The bezoar-goat is called Ji'
âll, and Al-Damiri says :
'
'When it is bitten by a snake
it eats the crab (as an antidote). It associates with
fish on terms of friendship, and walks to the sea-coast to
see them upon which they also approach the shore
to see it. Fishermen know this and with that view
put on its skin for the purpose of meeting fish which
they then catch. It is addicted to eating snakes which
it seeks in places where it can find them; if it is bitten
by any of them, tears flow down from its eyes to the
hollows which are under the sockets of the eyes, and
which are deep enough to admit a finger. The tears get
congealed and become (lustrous) like the sun, in which
state they are used as an antidote for the snake-poison,
and are known as the animal bezoar-stone. The best
kind of it (bezoar-stone) is yellow, and the places in
which it is found are India, Sind, and Persia. If it is
placed on a snake-bite or the sting of a scorpion it is
beneficial and if a person who has drunk a poison holds
;
it in his mouth, he too will be benefited by it. It has a

wonderful property in warding off the effects of pois-
ons. "^ In reality, the bezoar-stone is a globular con-
cretion in the large gut of certain ruminants, chiefly
goats.'' The Peruvians wore the ilia as a talisman to
ensure good luck, hence ilia also means "that which
attracts good luck, good luck." In Aymara, ilia
means "bezoar-stone, anything kept for a future day,
provisions, clothes, jewels," etc. The Kechua has
similarly ilia "kept for a long time," ilia kollke "old
piece of silver, generally with a hole through it and
worn on a string around the neck as an amulet and
supposed to attract silver." In spite of the close re-
' A. S. G. Jayakar, AdrDamtri's I}ay&t Al-Hayawdn, London 1906, vol. I,
p. 222.
' H. Fiihner, Bezoarsleine, in Janus, vol. VI, p. 318 flf.
semblance of KecliuaiZZa and Arabic 'ail, we cannot con-

nect the two, because we have no proof of the survival
of tbe latter word, in tbe sense of "bezoar, " in Africa.
But this much is certain, ilia has reference to something
that is old and precious as an amulet. It is precisely
what il means in Nahuatl and the Maya languages.
We have Nahuatl Ilamatecutli "the old goddess,"
that is, the goddess of the Trinity, the mother of the
later gods, wife of the fire god Ueueteotl, the old god.
llama has in Nahuatl assumed the meaning "old
woman," but Ilamatecutli shows that the reference is
to the "old goddess," that is, that ilama must mean
"the old god." Indeed, we have Bambara alia, ngala
"god" and allama "divine," while in Mandingo both
alia and allama mean "god." This allama is from
Arabic c^ âldmu, r^ "the omniscient; he who
"âllamu
knows what has been and what will be, from whom
nothing is concealed in earth nor in the heaven, whose
knowledge comprehends all things, the covert thereof
and the overt, the small thereof and the great, in the
most complete manner." The confusion of this with
Arabic '^î •"^' Illah, Allah " god," and of both, in Africa,
with the "old god," has produced Nahuatl ilama "the
old (goddess)" and il, in compounds, with reference to
heaven or knowledge. This Arabic r^ 'alamu has
been understood as aUdmu, al-namiyah "the vegetative
soul, " and from the confusion with "the old god, " that
is, the Moslem god, has been superimposed upon the
Habbes' Ammo, and has led in America to Omeciuatl, by

the side of Ilamatecutli. In Kiche we have il "to ob-
tain, see, much, great; guilt, misfortune,
preserve;
evil." would seem impossible to connect all these
It
meanings, but if one keeps in mind that the original
meaning is "noteworthy and remarkable," as shown in
the Peruvian, one sees how it means "to preserve, ob-
. "

tain," and "great," and "a noteworthy, but unfortun-
ate, event," that is, "misfortune." We shall meet
with this in the other languages. Pokonohi has il,
ilvuic "to see," Kekchi has il "to see," ilbal "to take
care of, watch, " Maya has ilah "to see, watch, observe,
ilil "vicious thing." The Nahuatl ilhuia means "to
imagine, invent, do a thing with the use of all one's
powers," hence "to increase." Here we not only see
what seems to be a borrowing from a Maya language,
but we get the fundamental meaning "to do something
noteworthy or remarkable, " and pass over to the mean-
ing "much," hence ilhuice, ilhuiz "much more, above
all." Hence we get ilhuitli "merit, recompense" and
ilhuitl "celebration day," that is, "a noteworthy day,"
and then "day in general." We have also the com-
pound ilnamiqui " to remember, imagine, think, reflect,"
from namiqui "to meet, grasp," that is, in ilnamiqui we
have the connotation "to grasp a noteworthy thing."
In Nahuatl ilhuicatl "heaven" we have, no doubt,
il-uica "god government, the reign of the old gods."
Among the Peruvians ilia is similarly applied to the
celestial divinities, hence illay "to shine," illaj "light,"
illapa "lightning and thunder, misfortune." In the
latter case we have the identical meaning as in Kekchi
and Maya. "After Viracocha and the Sun, the third
huaca, and the one most venerated, was the thunder,
which they called by three names, Chuqui-illa, Catu-illa,
Intu-illapa, imagining that it was a man in heaven with
a shng and club in his hand, in whose power is rain, hail,
thunder, and everything which belongs to the region of
the air where the clouds are created. This is the gen-
eral huaca of all the Indians, and different sacrifices are
offered to him, and in Cuzco they also used to sacrifice
children, as to the Sun. ""^ We have too scanty infor-
' Coleccion de libros
y documevtos referentes d la historia del Peril, Lima
1916, vol. Ill, p. 6 f
Abb rym. Thc>nfi>;iii
Forr,! .-ic.."' Krugps, mil Alili, ;it)l Til ni. fltw nut 3 m (i.-L..Ul-.- <.W
HOik.iti.'i'-'ti-iK.
.1,-ni IJ.-i.-liK' doH R.-ii'-ii
V, niitlirliL'lKT (;î—
'^t,Ui'> Tlalof KMiih^l. >riw iin tilr \ r.lkcrkimdf Tlcilin-
';, until I'lii^her ttrO»ii-.

\VilI<.-rkiiiul>., fi'Tlin.
Kilnigl. MiwMim flU

Vl''lkprkiin*3c. Berlin.
THE RAINGOD TLALOC, from Seler's Codex Borgia.

mation here, but the use of the club and the sacrifice of
children at once identify the Peruvian thunder god with
the Mexican rain god, Tlaloc. The Mexican Tlaloc is
hard to define etymologically, but we can locate his
prototype in Africa.
"The god which bears this name of Tlaloc was one
of the most popular figures of ancient Mexico, and rep-
resentations of this god belong to the most common
finds from pagan times .... The crest of mountains
over which the road goes from Tetzcoco to Uexotzinco
and Tlaxcallan was the region to which the name of
Tlaloc or Tlalocan was more particularly applied and
which was considered to be the seat of the rain god and
where there was an ancient idol of this god, made from
lava, with its face turned to the east, and carrying upon
its head a dish, into which every year was placed every
kind of reaped edible seeds. The god, who was
. . .
considered by the inhabitants of the Mexican plateau to

dwell on these heights, and was at the same time wor-
shipped throughout the land, is everywhere represented
in a fairly uniform and very remarkable way. His face,
as may very well be seen in a stone figure of the Uhle
collection in the Royal Museum for Ethnology at
Berlin, consists entirely of the windings of two serpents,
which, spirally winding about each other in the middle
of the countenance, form a kind of nose, then on each
side twisting into a circular knot encircle the eyes, and,
finally, meeting each other with their muzzles, represent
the upper lip and the long, descending fangs of the god.
This fundamental form has in a continuous simplifica-
tion of the design assumed the usual shape of the stone
figures and manuscripts."^
Thus we arrive at the identification of Tlaloc with
serpents. At the same time the syncretized forms of
Tlaloc represent him as a fire god or as Xolotl, hence we
1 Seler, Codex Borgia, vol. I, p. 108 f.
"

have the constant equation of "serpent-heaven-rain."
In the Maya documents Tlaloc is called Chac. He is
painted red, is represented with a skeleton body, and
carries on his head a bowl with kernels or ears.^ In
Maya chac means "red, " but we have chacal "cooking,"
chacah " to boil in water, " and chaac "rain, downpour,
so that here we have the "rain" word connected with
"red, ripe," as in the case of the Mexican rain god.
Similarly we have Kiche chag "ripe," Pokonchi chak
"cooked," chakij "dry, hot," chaklaj "ripe," and
similarly in the other Maya languages, so that "rain"
is brought in contact with the sun and flre.
Here, similarly, "sky" and "serpent" are correlated.

In all the Maya languages a root can means " serpent,
but in Maya we have caan " sky, " caanal, canal "high,
where "sky" has arisen from the meaning "serpent."
In Nahuatl couatl means not only "serpent," but also
"twin," so that we are once more brought back to the
twin god Xolotl, with whom Tlaloc is confounded. In
Maya we have also Ah bolon tz'acab taking the place of
Chac,^ but the sentence means "he of the nine geneal-
ogies," so that we are once more brought back to "the
nine guardians of the night," with which the sun god
and the fire god are connected.
It would seem maddening and impossible to connect
heaven, serpent, and rain, if we did not possess a
complete solution of the matter among the Bambaras.
"I have always been surprised by the facility with
which the black man designates a divine favor by the
name of its author, and when in the rainy season he re-
ceives the chief favor, the rain, which makes everything
germinate and the crops grow, this is the way he ex-
presses himself: 'Alia bi fign, Alia bi na, Alia bi kulu-
kulu, Alia bi yege yege, Alio do, etc. God is growing dark,
' Ibid., p. 335.
* Ibid., p. 217 and vol. II, p. 154.
God comes, God rumbles, God makes lightning, it is

God, ' etc. These are
purely metaphorical expres-
all
sions, to say the clouds are gathering, it is going to rain,
'
thunder is rumbling, it is lightning, it is raining,' etc.

"These same metaphors I find, in an absolutely iden-
tical sense, but this time in Bambara terms, 'Sa hifign,
sa hi na, sa hi kulukulu, sa hi yege yege, sa do, which also '
mean 'the clouds are gathering, it is going to rain,

thunder is rumbling, it is lightning, it is raining. The '
word sa being exclusively Bambara, it cannot reason-

ably be considered as posterior to the Arabic word,
hence the Negro possessed it before knowing the word
Alia. This word did not mean 'cloud, thunder, light-
ning, rain, and there is reason to believe that there was
'
a time when, as to-day, it awakened in his mind an iden-

tical idea as does now the word Alia, and by this word
Alia he expresses merely a concept long possessed by
him.
"What does this word sa mean? Could it be the
primitive name of the divinity? Since certain Bantu
tribes had much respect for serpents and worshipped
them, the Europeans wanted to see in it a translation of
the word 'serpent.' But that is a gross error. The
word sa serpent and the word
'
' sa in the phrases sa hi
fign, sa hi na, etc. may have thesame orthography, but
the pronunciation is different. The hard s of sa 'ser-
pent is pronounced by slightly opening the teeth and
'
puckering the lips, while the word sa in sa hi fign, etc. is

pronounced by compressing the teeth and opening the
lips. Therefore these are two different words. It
seems to me
a bit of stupidity to consider the serpent as
being or having been a Bambara divinity. If even with
his tabu taken from among the animals he does not look
for any physical and real tie of relationship, one must a
fortiori assume that he cannot consider as creator and
sovereign master an animal, even if it were a serpent.
'

Certain Bambaras have it for a family tabu, and a
branch of the Mande race has so great a respect for the
python Minia that it can be recognized only under the
name Minianka, the people of the serpent Minia. But
have the Minianka ever worshipped the serpent as their
god, their supreme divinity? They are less stupid. The
serpent is a common tabu, the residence of a jinn, an
infernal spirit, a demon, and nothing more.
"The word sa in the phrases sa hi fign, sa do, etc.
'
the clouds are gathering, it is raining, etc. has a broad
'
meaning. It signifies everything that is upon high,

everything which is above us. Has the word itself ever
been the name of a divinity, as the word Alia actually
stUl is? I think not. The word sa was a veiled ex-
—
pression for divinity, it expressed essentially every-
thing that the word Alia did as a principle, an idea, and
not 'a being.
"God being according to those concepts a spirit re-

siding above us with his court and angels, these ex-
pressions sa bi na, sa hi fign, sa hi kulukulu, sa dyi, etc.
should be translated, to have any meaning, by Min be
sa-n-fe bi na, Min he sa-n-fe hi fign, Min he sa-n-fe hi
kulukulu, Sa-n-fe ta dyi, He who (the spirit that) is above
comes (will do us the favor of giving us rain) He who is
,
above gets black (gathers the clouds). He who is above

rumbles (makes his voice heard). The water of him who
is on high.
"The favor par excellence granted to human beings by

the divinity, the supreme being, is rain, and the Bam-
bara, as I shall show presently, not being allowed to
pronounce the name of this god, has in his gratitude
used a metaphor. This metaphor, Alia hi na, Alia hi
fign, identical with the one into which the word sa
enters, namely sa hi na, sa bi fign, etc., leads me to be-
lieve, nay, compels me to believe that this word Alia
originally expressed less an individual, a being, than an

idea, a concept."^
One must be grateful to Henry for the very explicit
account of this matter, though his explanation is ob-
viously wrong. I have already shown that in the
Sudanio languages God and heaven and rain, that is,
Allah and sama', are interchangeable,^ and that the
latter has led to Mande sama and sa "heaven, rain,"
hence the Bambaras are logical when for Allah they sub-
stitute the more common sa, which means so much the
more to them, since the chief benefaction of the invisible
divinity is rain. Just as in Hausa rain is interchange-
'
'
'
'
ably man sama or man allah, literally "sky water" or

"god water, " so in Malinke and Bambara sa takes the
place of Allah. But this sa, which also means "year,
age," was naturally confused with Malinke, Bambara sa
"time, moment, end, cessation, death," from Arabic
^^ saah "hour, instant, resurrection, death, distress."
But Malinke, Bambara sa also means "serpent."
this
The origin of the word is not certain, but Soninke
samaqhe, samake, Soso sanyina indicate that we
have here a frightfully reduced Arabic samak "fish, eel,"
but the verb '^^ samaA;a also means " to elevate, raise "
and refers to heaven, so that the confusion may already
have occurred in Arabic, especially since the serpent was
considered to be a jinn and messenger of God. "The
true form of the jinn is a serpent, with which, however,
other reptiles and vermin are connected under the com-
mon name of lianaS. The serpent is gann and gul;
in every serpent there is a jinn. Noldeke has collected
several examples. During the war of ditches, in the
fifth year of the flight, a man from Medina ran his lance
through the body of a serpent which he found lying
• Henry, op. eit., p. 76 flf.
2 See p. 174 ff.

upon his bed, and planted the lance in the yard, with
the trembling serpent on its point, but the moment the
serpent died he himself fell dead. Mohammed was
asked to invoke God, to bring him back to life, but he
said: 'Ask forgiveness for him, for it is a devil.'
When the prophet was on his way to Tahuk, a large, fat,
male serpent crawled up to him and remained long in
one place, while he stopped with his camel, then he
moved aside and stood up straight. 'Do you know,'
said Mohammed, 'what this is? It is one of the eight
jinn who wants to hear the Koran. He must be greeted
since he is visiting the messenger of God in his country.
He himself is greeting you, so you must return the
greeting!' And the people did so, but the prophet
said, 'Love God's servants, whoever they be!'"^
The confusion of "serpent," "rain," and "God" was
inevitable in Mande, and has similarly passed over to
America. But the parallelism does not stop here.
The modern Malinke, Bambara sa "serpent" is not the
original form, because we have in the other Mande
languages a form ka, kan, kali, such as Kono, Vei ka,
Gbondi, Landoro, Mende, Gbese, Toma kali "serpent,"
and this appears in the neighboring Bornu as kadi, gadi,
and here Wolof dyan shows that these words are most
from Arabic >J^ gdn "snake." In Bam-
likely derived
bara and Malinke ka, ka, kan means "above, " and just
as Mandingo santo, Malinke safe, Bambara sanfe
"above, upon high" are derived from Arabic sama'
"heaven, " so this may have arisen from Mande ar-dyan
"heaven." But whether this etymology is correct or
not, there can be no doubt as to the homophony with
the "snake" word. Again, Malinke ka, Bambara ka,
kan, kana mean " to cut with a sickle, " and in the other
Mande languages we have Vei kali, Soso keri, Gbandi kail
' Wellhausen, op. cit., p. 137 f.
"hoe. " This latter word is of very wide distribution in

Africa. In the Arabic oases we have Wadai dsarai,
Adirar d§alo, Beram keri, and both occur in Peul, Fil-
ham ebara, plural tibara, Baga daba, plural saba, and
kdra, plural tsdra, Timne ketsala, plural tsetsala, Bulona
kara, plural sikara, all show that we have in all these
various corruptions of Arabic Jj^ sauqar "a pickaxe,"
and Malinke, Bambara daba "native hoe" is apparently
borrowed from a language like Baga, where daba occurs
as the singular of saba, and this for kara, tsara side by
side with it, from sikara in Bulom.
However this may be, we have the remarkable fact
that in Malinke and Bambara we have the homonyms
ka, kan "to cut with a sickle" and "above, upon high,"
while the other Mande languages have ka "serpent."
It is, therefore, hardly a coincidence that we have Maya
can "serpent," caan "heaven," but also Nahuatl coatl
"serpent, hoe," the latter sense being also recorded by
Las Casas as coa for the Caraibs. Coa in Nahuatl also
means "to buy," and in coaca "to invite to a repast,"
tecoani "merchant who invites to a banquet" it ob-
viously has something to do with merchants. Now
Malinke s and k are interchangeable in many cases, and
the ka, ka, kan words are thus brought into intimate re-
lation with the sa, sa, san words, but here we have sa
"to buy, sell," sani "purchase, sale," sanila "buyer,
seller, " Bambara sa "market, " san "to buy, exchange,"
samba "buyer, peddler, " sani "purchase, sale, " sanikela
"buyer." Just as Nahuatl coa "to buy" is related to
coatl "snake, " so we have Maya con "to sell" related to
can "serpent, " and in both the merchant is brought in
contact with the serpent, that is, with Quetzalcouatl, the
feathered snake, with whom Yacatecutli, the merchant
god, is syncretically connected. Before giving the
precise meaning of the bird snake, as which Quetzalcouatl
.

is to be understood, we must analyze tlie uactli of the
Mexican religion, to which Quetzalcouatl is related.^
Among the Aztecs the uactli, identified as the Falco
cachinnans, was considered as a bird of omen, more
generally of good than of bad omen, and sometimes con-
nected with Quetzalcouatl. Among the Mayas, another
bird, the moan, apparently an eagle or vulture, is a
mythical bird connected with Quetzalcouatl and repre-
sented as a bird with a snake body and often a snake
tail. There can be little doubt that we have in these
birds a graphic representation of the god, the feathered
snake. The relation to the god becomes clear only
when we consider the African bird of omen, described by
de Marees.^ One need only compare this description
with Sahagun's description of the uactli. "Its augury
was indifferent, since it as frequently announced good as
evil. It was considered good when its laughter simu-
lated the ordinary laughing, because it then seemed to
say yeccan, yeccan, which means 'good weather, good
weather.' When the bird uttered such a sound they
had no suspicion of any evil. On the contrary, they
were glad to hear it, because they expected good luck.
But when this bird in singing seemed to imitate laughter
in a high voice that seemed to come from the very depth
of the chest, as happens upon the occasion of a great
joy, those who heard it lost their voice and strength.
They stopped talking among themselves."^ Then Sa-
hagun gives a lengthy account of the merchants plight '
when they heard the uactli' s unfavorable laugh. The

pitoir of de Marees is the Spanish "buitre, " the vulture,
who is in Malinke and Bambara called duga, but this
word also means "imprecation, benediction, curse, " and
is the Arabic '^-^ dua' "a prayer, supplication, in-

' See p. 120 f
' Op. cit., p. 295 f.
vocation of good, a blessing, benediction, a calling or

crying for aid or succor." Here we have a purely lin-
guistic reason for selecting the vulture as the bird of
omen, and Allah, the giver of a blessing, who is invoked
with a prayer, has thus changed into a vulture, even as
through the sa and ka words he became identified as a
serpent. Thus God became the bird and snake par
excellence, the Quetzalcouatl of the Mexicans, the Kuk-
ulcan of the Mayas. Indeed, we hear that Quetzalcouatl
was the god of the Tolteca, that is, the race that civilized
Mexico, but since in Africa dala refers to the "makers,
manufacturers," that is, to Arabs, we have in Quetzal-
couatl the god of the Arabs, that is, Allah, as has already
appeared from previous considerations. No wonder,
then, that the uactli, the bird of omen, should also ap-
pear as Toluactli, the bird of the Tolteca. But the
linguistic homonyms of coatl also lead to the identifi-
cation of Quetzalcouatl with the hoe, which he is repre-
sented as holding in his hand.^
It is more likely, however, that Toluactli is a mere
popular etymology of a word which should be touactU,
since the Malinke duga stands for Arabic t-J^ duwd "a
certain night bird, a species of owl," from the verb
which means "it frightened him," that is, the bird of
omen par excellence. The confusion of the bird of omen
with " divine interference, blessing, curse" was perfectly
natural. The Maya moan bird looks much more like an
owl than an eagle, and has thus been identified by
Seler.^ The owl, tecolotl, isamong the Aztecs a symbol
of death,' and among the Bambaras death is frequently
applied to the fetish Duga.* There can be little doubt
that tecolotl stands for tocolotl, and, in the first part,
1 Seler, Codex Borgia, vol. II, p. 187.
2 Ibid., vol. I, pp. 13, 180, etc.
5 Ibid., p. 68.
* Henry, op. eit., p. 48 ff.
represents the Malinke duga, for we also have Maya
tunculuchu "owl," apparently borrowed from the
Aztec, or both froip a third source. In its turn tecolotli
is but a different form of toluactli. The apparent dis-
crepancy, in applying one to the owl, the other to the
falcon, while duga itself in Malinke refers to the vulture,
is due to the fact that the Arabic name duwa is " shared
in common with it by other birds, that is to say, is ap-
plied to all nocturnal birds, which come forth from their
nests at night. "^ "Al Mas'tidi has copied from Al-
Jahid, that the owl does not show itself in the day out of
fear of the influence of the evil eye affecting it, on ac-
count of its beauty and handsomeness; and because it
considers itself the handsomest of all animals, it does
not show itself excepting at night. "^
This leads us at once to the Mexican quetzalli, the
beautiful feathers found in Tecolotlan, literally "the
country of the owl," "a province situated near Hondu-
ras, where formerly the feathers of the Quetzaltototl were
found." Thus once more Quetzalcouatl is the bird
snake, the bird represented by feathers from Owl-Land.
But the Nahuatl quetzalli is an African name. For
Quetzalcouatl the Mayas say Kukulcan, in which, as we
have seen, can means "serpent," hence kukul should
mean something like "owl-feather." Indeed, Maya
kukum, kukumel means "feather." But both Nahuatl
quetzal and Maya kukum are represented in Wolof
khergej, Hassania Arabic kujil, Soninke gugute, Serere
lukukuk, Malinke klkln, Bambara gingi, guelu "owl,"
while in Songay gaga is vulture." The handsome
quetzal bird took the place of the African owl, which,
according to Arabic tradition, was the handsomest bird,
and was applied to the chief divinity, the Quetzalcouatl,
the god of the Toltecs.
1 Jayakar, op. cit., p. 346.
2 Ibid., p. 347.
"
We have already seen that the social and religious

orders are of African origin. It now remains to show
that the political order of the Mexican state is nothing
but Mandingo in every detail. Nahuatl tepetl means
"country, locality, mountain," while altepetl means
"town, state, king, sovereign." This Nahuatl word
has entered many American languages. We have
Maya tepal "king, majesty, highness," Kiche tepeu
"greatness, glory," Huasteca alte "woods, monte,
Tarascan tepacuaro "city," tepani "to be large,"
tepamani "to rise." I have already pointed out that
the mounds of the Mound-builders were town sites, with
the hill for the cacique's residence and temple, just
as in the Sudan, and that the North American stockade
is identical with the one in West Africa. The close re-
lationship between the art of the Mound-builders and
that of Mexico has been shown by Holmes. It now can
be shown that all that is a development of the Bedouin
encampment in the Mandingo country.
Arabic ->'» ddr, plural adwar, adwirah, etc., "habita-
tion, house, residence," J^^ da'irah, plural dawa'ir,

"suite of apartments, convent" have had peculiar de-
velopments in western Africa. We not only have at
Timbuktu the duaria "houses with one wall open and
supported by columns," but Ibn-Batutah and Idrisi
report there Jh^ dawwar "encampment of the Bedouins
with the tents in a circle and the cattle within the circle,"
which is obviously the simplest form of a stockade. In
the Hassania Arabic we have both diar "house" and
the much corrupted daSera "village," from Arabic da'-
irah, though dasera is supposed to be of a different
origin. We have similarly, in Africa, Biafada dare,
Padsade ydr, Bulom ter, Gura dsawa. To us the most
interesting development of this Arabic word is in the
Mande languages. We have Malinke, Bambara dugu
"

"village, earth," which has been shown to be derived
from Arabic and which crowded out the
>=-»"'
tujiut,
Arabic J^J^ dawwar, preserved in Soninke. We have

Soninke debe, which is the older form, and is obviously
derived from the Arabic. This is independently proved
by Malinke debe "to curl or braid the hair," which is
also from Arabic »J^^ da'irah "the circular or spiral
curl of hair upon the crown of a man's head, " which is
also preserved in Hausa dauri twisted locks of hair ar-
'
'
ranged on either side of woman's face." By the side

of J^^ dar the Arabs use al-dar "the City of the
-''•^'^
Prophet," and similar expressions are found in Africa.

Thus, by the side of Hausa gari, kauye, Timbuktu
koyra "town," from Hassania Arabic garlatun, Arabic
\J qaryah "village," we have also Hausa alkaria
"large city, capital of a chief, town placed on the com-
mercial highways.
Even thus Mande debe produces Nahuatl tepetl
"town with its hill" and altepetl "large city, royal
power."
It can now be shown that the very crown of the
Mexican kings was derived from the Mandingos. I
have already given a part of the vast number of
"enclosure, cover" words derived from a root karpar,^
which led to the "cotton, garment" words. I will now
treat with that development of the group which led to
the perpetuation of the conical hat with a neck-flap as a
distinctive head-gear of kings and priests. We have in
Assyrian karballatu "name of a garment, cap." We
can tell what kind of cap was meant from the occurrence
of the word in the other Semitic languages, for we have
Syriac karbdltd "cockscomb," and while Hebrew ^31?

' See vol. II, p. 11 f.
kirbelmeans "to be dressed in a cloak," Talmudic

«Pl?515 karhalta means "covering of the head, helmet;
cockscomb." By the side of karballatu the Assyrian
has also kuMu "head-gear, cover," so that both represent
an original karpar, the latter having passed through a
form kurpaS, that is, the Sanskrit kurpaSa "coa,t of mail."
That this Assyrian kubiu was originally kurbSu follows
from the Greek xijpâ(rta,reGorded. by Dionysius and Hero-
dotus as meaning "high, conical hat," by Aristophanes
as "cockscomb" and "hat of the great Persian king."
But Pollux also writes xuâpaia, which is not a mere
blunder, since similar forms, including Assyrian kubSu,
occur. A form kurbaS is responsible for kalpak, the
name of the pointed hat over an enormous territory
among the Tatar tribes, while in Tibetan we have kebs
"cap, hood."
The phonetic form of the original word deteriorated
very early. A garment kubbu is already mentioned in
Assyrian, but the exact meaning is not ascertainable.
However, we have Talmudic ^513 kobd "high cap,
turban, priestly hat, Adam's apple," n^313 kub'dh
"sheaves tied together over the ear of corn," ^^^'^p
qoVd "cap, turban," Hebrew f^VI^P migbadh "cap or
fillet of the priest," Syriac ^zso* qui) a, Ethiopic qobue
"cap. " The development of this from a longer form is

seen in Persian kulbak "a covering for a stack of grain,"
kulbe, kurbe "hut." It is this shorter form which
entered Europe in the form of cappa "hood," but I
must forego here the extremely interesting develop-
ment of this word, and confine myself exclusively to
the conical hat.
The Assyrian conical hat, which was also the royal
hat of Persia, played an important part among the
Magi and entered into the Mithra worship. It gener-
a

ally was the representationof the visible heaven, and so
was painted to represent stars, ^ and St. Augustine
called Mithra " deus pileatus. " It is this starred hat of
the Magi that we find in the Middle Ages upon the head
of a magician, and the Arabic magic unquestionably
perpetuated this hat together with the eastern magic.
Indeed, we find the same Semitic word in Arabic,
namely quU "cap, conical hat," but more com-
i;;^
monly the form ^-Jjl 'uqruf "conical hat," "J^ gifdrah

"cap."
I have already summarily treated the hat in Africa,''
and now we can study it in greater detail. The latter
word is recorded in the Arabic oases as ofdra, rofdra,
gofdra "hat." But we have also Arabic J^ migfar
"what iswore beneath the helmet, a piece of mail, —
man throws it upon his head, and it reaches to the coat
of mail, —sometimes they make above it a tapering top
of silver; the term is also applied to the helmet itself."
This leads to Hausa malafa "a large hat made of
plaited straw," hence Peul malafdre, Afudu mfoar,
Kamuku malafa, Esitako mardfa, Musu marafua,
Bornu maldwa, Pika malufa "hat." Arabic 'uqruf is
found in Soninke kurufe, Bornu gurumbd, Karekare
gurumpa, while Arabic gifdrah is found in Boritsu glbur,
Gurma kapira, Legba gboro, Yula ylpura, Timne alapra,
Bulom lapora, and in the Mande languages in Gio gbira,
Mano gbola, Gbese, Toma gbara, Mende gbawere,
Gbandi gbaralei, Tene lavdra, Vei gbdra, Bambara
liblri, Malinke libldi, libri, sibri, Soninke kufune.
Malinke gabd, Peul hufune, plural kufune, bear witness
to the fact that the Mande forms gbara, gbira go back
to Arabic gifdrah, and the Mande form fula "cap,"
' F. Cumont, Textes et monuments figures relatifs aux mysUres de Mithra,
Bruxelles 1899, vol. I, p. 115 f.

2 See p. 233.
®
.r«
Ie o <^
a
Abb. 284. Quetzalcouatl, der Biisser von ToUan.
Codex Vaticanus A (Nr. 37.38) fol. 7 verso
(= Kingsborough 11).
CONICAL HAT, from Scler's Codex Borgia.

found over an enormous territory, is an abbreviated

gufula. We have Gio gbilen, Dewoi fule, Gbe file,
Hwida fila, Aku efila, Wossi fowila, Gurma foalerah,
Kupa efula, Dsarawa /oZa "cap," and all these point to
an older form gofila, kofila, by the side of Arabic gifarah,
and this is actually preserved in Swahili kofia "cap."
We also have Arabic '^ qabllah, J^ qabllar "hood,"
which is a contamination with Spanish capilla, itself of
Asiatic origin, like the Arabic qub\^
Clavigero says that "the crown which was called by
the Mexicans copilli, was a sort of small mitre, the
fore-part of which was raised up, and terminated in a
point, and the part behind was lowered down, and hung
over the neck. ... It was composed of different
materials, according to the pleasure of the kings; some-
times made of thin plates of gold, sometimes wove with
golden thread, and figured with beautiful feathers."^
The copilli was a specifically Huastecan head-dress,^ and
was frequently made of an ocelotl skin, so as to represent
a lot of dots, that is, the stars of the magician's cap.
That the kingly and priestly cap of the Magi should
have been preserved in America in the identical form,
with the identical decoration, and should, besides, have
kept the name current for it among the Mandingo
people, makes it impossible to admit any other solution
than the one that the Mandingos established the royal
and sacerdotal of&ces in Mexico. Ultimately this
external representation of the royal power goes back to
Persia, even as a vast amount of the Arabic magic has
its origin there, and many of the religious motifs in the
Mexican religion, through the Mandingo prototype,
go back to Persian conceptions, but only as they seeped
into Africa through Arabic magic. The universal
• D. F. S. Clavigero, The History of Mexico, trans, by C. CuUen, London
1787, book VII, sect. VIII.

2 Seler, Codex Borgia, vol. III. See under copilli in the vocabulary.
Arabic vocabulary and the popular Arabic etymologies
on which the religious speculations are often based make
this conclusion positive and final.
Further analogies in America and Africa must be left
for a future time. Here I wish only to do homage to
the man who long ago had suspected the African in-
fluence in Mexican civilization, though he could give no
analyses of special cases or define the time at which the
transference of African fetishism had taken place. It is
no mere accident that in 1862 a colossal granite head,
representing a Negro, was found in the canton of Tuxtla,
that is, near the place where the most ancient statuette
was discovered, and that at Teotihuacan, that is, the
oldest Mexican city of temples, a number of Negro
heads in stone have been discovered. Orozco y Berra
was forced to the conclusion from these finds that in the
past a relation had existed between the Mexicans and
the Negroes.^ It has been my good fortune to establish
this relation on positive and irrefutable evidence.
' M. Orozco y Berra, Historia antigua y de la conquista de Mexico, Mexico
1880, vol. I, p. 109 e.
CHAPTER XIII.
The Mexican New Year.^
The 62. canon of the Council at TruUo prohibited the
celebration of the calends of January and the "vota
brumalia, " which took place on March 1; at least, men
were not permitted to dress in women's clothes, and
women in men's clothes, nor to put on any masks.
This legislation against the pagan celebrations is only
natural, and no specific importance is to be attached to
the demand that men abstain from donning women's
clothes, and women men's clothes, for these acts rep-
resent forms of the pagan mummery scorned by the
church. But, at even an earlier date we find a frequent-
ly repeated prohibition against acting as an old woman,
which cannot be explained from any classical source,
to which it may refer.
In the first canon of Auxerre we read that no one is
allowed on the first of January to act the old woman.^
Nothing definite is known about the date of this
council, which is variously dated from 573-603. That
the passage in question is, to say the least, an interpo-
lation follows from the bad Latinity "vetolo, " "cer-
volo, " which is out of keeping with the rest of the text.
The canon itself is based on the 23. canon of the
Council of Tours (A.D. 567), which objected to the
celebration of the calends of January, but where there
is no reference to acting as an old woman. In the
1 After the book was finished the startling discovery here discussed was
made. It is given without the pregnant consequences to which it leads.
2 "Non licet kalendis lanuarii vetolo aut cervolo facere vel streneas
diabolicas observare, sed in ipsa die sic omnia beneficia tribuantur, sicut at
reliquis diebus," Monumenta Germaniae historica, Concilia, vol. I, p. 179.
second book of the Vita EUgii, which is of late origin,
that is, belongs to the VIII. century, we have the same
reference to the calends of January,^ which, with the
interpolation in the Auxerre canon, is taken out of
some late source.
The earliest reference historically to this word, from
the VIII. century, is found in Dicta abbatis Priminii,
de singulis libris canonicis scarapsus,^ namely: " ceruulos
et uetulas in Kalendas uel aliud tempus nolite anbulare."
This prohibition is also found in Cummianus, Theodor
of Canterbury, Halitgarius, Bmrchard of Worms, ^ all
of which are of later origin, and need not detain us.
More interesting and more important is the treatment
of the matter in two pseudo-Augustinian sermons:
"Who can be considered to be wise or of sound mind
who 'making the stag' (cervulum facientes) wish to
transform themselves by donning the vestments of
beasts? Some put on animal skins; others put on the
heads of beasts, and rejoice when they have thus
transformed themselves into wild animals and no longer
appear as men. By this they show and prove that
they have, not so much the vestments, as the intellect,
of beasts. For while they wish to express in themselves
the similitude of animals, it is certain that there is in
them more the heart of beasts than their forms. Again,
how stupid it is for those who are born men to put on
women's garments, and by an indecent transformation
to effeminate their virile strength with a woman's form,
without a blush putting their soldierly arms through
women 's tunics they show their bearded faces and want
:
to appear as women. Indeed, they no longer have their

virile valor who have assumed the garments of women,
and so it is to be considered a just judgment of God that
' "NuUus in Kalendas lanuarii nefanda et ridiculosa, vetulas aut cervulos
vel iotticos faciat," Scriptores rerum merovingicarum, vol. IV, p. 706.

^ C. P. Caspari, Kirchenhistorische Anecdota, Christiania 1883, vol. I, p. 175.
' Ibid.
THE MEXICAN NEW YEAR 325
those wlio appear in the form of women have lost their

military valor. " ^ " Some commit that vile turpitude of
'making the stag' (de hinnicula vel cervula exercere). "^
It is clear that these sermons distinctly refer to two
separate mummeries, that of an animal, and that of an
old woman. In the latter case we read annicula,
anula, agricula in some manuscripts, and this is for
anicula "old woman," that is, an equivalent of
"vetula." This is taken out of some Graeco-Latin
gloss, for we have in Placidus " grauescella. grauesidus
anni, " "grauascela graues id est anni, "^ which ob-
viously should be "xpauz id est anicula," and all the
interpolations and forgeries read the first as "cervus"
and produced the "cervula aut anicula," "cervula aut
vetula" of the texts. Indeed, "anicula ypaix:" is a
very common gloss. The "anni" of the Placidus gloss
is due to "anui anicule" found there elsewhere.*
It is now possible to show that the whole prohibition
is directed toward the mummeries of the Arabs, in-
troduced by them into Europe from the practices among
the Christians in Egypt. The Egyptian solar year
consisted of twelve months of thirty days and a
thirteenth, five-day month, called the "little" month.
The Christian Copts did not disturb the Egyptian
calendar, and called the thirteenth month pi-abot
n-huzi " the little month, " which was considered worth-
less, and during which time no important work was
undertaken. This month fell in the end of February,
and the old Greek and Roman custom of celebrating the
calends of January was also applied to this period of
the year. The Arabs transformed the Coptic word
kuzi "little" into Arabic Jj^ "aguz "old woman," and
1 Sernu) CXXIX.
2 Sermo CCLXV.
5 Goetz, op. (At., vol. V, pp. 24, 72, 107.
< Ihid., p. 167.
"

a mass of folklore with this philological blunder,
let loose
which has swept over Europe, Africa, and America.
The Arabic Jj^ âguz is recorded in the Spanish-
Arabic vocabulary for vetula, the word which we have
already found in a mass of forgeries. The idea that
there were seven last days is already found in the
Koran: "Thamtid was destroyed by a terrible noise,
and Ad was destroyed by a roaring and furious wind,
which God caused to assail them for seven nights and
eight days successively. " It does not appear that these
referred to the intercalary days, which is the case with
"the days of the old woman.
Among the Berbers^ the old Roman calendar was not
entirely superseded by the Arabic nomenclature.
Among them the New Year is called yennair, inneir,
nnair, etc., from "Januarius, " and byenni, bydnnu,
apparently from "bonus annus." At Fez the New
Year's day and the following day are together named
haguza, elsewhere l-haguza, haiduza, which is repre-
sented as a female spirit of an old and hideous appear-
ance, which is obviously derived from the Arabic, by a
misinterpretation of the Coptic kuzi "little." At
Tlemcen they say that Ennayer once came in the form
of an old woman to ask for alms. At Tangier, Rabat,
and Fez "parents press their children to eat much of the
New Year's food, telling them that otherwise Haguza
will come and fill their stomachs with straw. A small
portion of it is sometimes left for Haguza in a covered
plate, and if any hair is found there the next morning it
is said that she has been there and partaken of the
food. "^ "In the Garb haguza is an unlucky day, when
1 E. Westermarck, Ceremonies and Beliefs Connected with Agriculture,

Certain Dates of the Solar Year, and the Weather, in Morocco, Helsingfors
1913, p. 56 ff.
2 Ibid., p. 63 f.
no ploughing is done but the people hunt and play at

ball."i
"From 25th February to 4th March (Old Style)
there a period, lasting for eight days and seven nights,
is
which is called la-hsum (Fez) or more commonly
hdyyan. The Braber of the Ait Warain call it tdmgart,
'the old woman,' probably because the winter is then
coming to an end. I was told that Tamgart was an old
woman living at the foot of BHiblan, the highest moun-
tain in the district of Ait Warain. Once when it was
raining during the three first days of the said period the
calves in her yard took refuge in her tent, but she drove
them away telling them not to be afraid of a little rain.
Then HSyyan said to Marssu (March): 'O March, —
lend me an evil day, I shall Mil by it the bad old woman.'
March, who then had thirty-two days, lent one of them
to Brayer (February) so that only thirty-one remained.
Now there came much rain and cold and snow. Tam-
gart with her tent and all her animals were transformed
into stone, and are still to be seen at the foot of Buiblan
where there is a large stone which from a distance looks
like a woman at a churn, another having the shape of a
tent, a third looking like a shepherd leaning on his staff,
and a collection of smaller stones resembling sheep.
In the Hiaina, where the second day of hdyyan is called
nhar la-' guz, the day of the old woman, the following
'
'
story is told. There was an old woman who went out

on the pasture with the sheep and goats. As the
ground was very dry and the crops were suffering from
drought, she asked Hdyyan to send rain. Hayyan in
his turn asked March to lend him one day; this he did,
and rain fell so heavily that the old woman was killed,
whereas the animals escaped to the village.
"Hdyyan is represented as a bitterly cold time of the
year, known for its rain, wind, and snow, which are
1 Ibid., p. 67.
considered very dangerous for people, animals, and
crops. It is called bu-tlilz, 'the master of snow,' in the
saying, Hdyyan hu-tluz, lulu bdida u aheru '
asluz
(Hiaina), which means that on its first day the part-
ridges begin to lay eggs and on its last day the young
sprouts of various wild herbs are big enough to be used
for food. Nobody likes to travel during this period,
hence all necessaries have to be provided in advance
(Garb, Hiaina, At Ub^hti). A Berber from the At
Ubahti told me that when he and some relatives once
during hdyyan went to fetch dates from a neighbouring
Arab tribe, two of their donkeys died on the road in
consequence of the rain; but he said that people, also,
may die if they expose themselves to the rain by travel-
ling in hdyyan. In the same tribe it is the custom to
keep the sheep inside the tents during a rainy hdyyan,
but even then they are supposed to be in danger owing
to the cold. There is a saying, '
Don 't separate your
kids from the flock till the nights of hdyyan have passed,'
or, 'till hdyyan has passed;' they, as also the lambs,
are then only too liable to be killed by the rough
weather. Especially the second day, nhdr, la-' g-dz or,
as it is called in Andjra, nhar l-md' za u r-ra' i, 'the day
of the she-goat and shepherd, is considered to be full
'
of danger; the shepherd must then be thickly clad and

eat well, and, at least if he is a young boy, somebody
must accompany him to look after him and the flock
(Hiaina, Andjra). But rain in hdyyan is considered
equally injurious to the grass, crops, vegetables, and
fruit-trees, its water being salt (Hiaina). Among the
Ait WarS,in and Ait Sadden nobody must go in the
fields during the three first days of this period; should
anybody go, the crops would get dry or be beaten down
by a thunderstorm, and even the owner of the field
might be personally affected by it. Nothing can be
worse than a thunderstorm in hdyyan: it hurts the
little children,animals, and bees, and makes milk and

honey The Arabs of the Hiaina therefore
say:
— scarce.
'May God save us from the thunder of hdyyan.'
On the other hand, if an east- wind blows in hdyyan the
'
durra will have a bath in the nlsan, and the year will
turn out good without scarcity'. For there is no rain
while an east-wind is blowing."^
That this is all an Arabic, and not a Berber, concep-
tion, follows from the Arabic word, which is due to a
Coptic homonym, and because the legend is equally
familiar among the Arabs in the east: " Jj=r^ also called
J^' because they come in the latter part ( 'j^ )

C^J ,
of winter; but the former is the correct appellation;

accord, to the usage of the Arabs, Five days, the names
of which are dr* and jr-^ and j.i and ,/^' 'J^ and
i>^' *LS°^ by Ibn-Kunaseh to be of the 'y of
; said
^-r-J' [by which is meant the auroral setting of the
Twelfth Mansion of the Moon, which, in Central
Arabia, about the commencement of the era of the
Flight, happened on the 9th of March O.S.: in the
modern Egyptian Almanacs, the Jyy^^ c^.' are said to
commence now on the 9th of March N. S., which is now
the 26th of February O. S.]: or, accord, to Abu-1-
Ghowth, they are seven days, named O^ and j-r^ and
y.i and y'^'^ and _r-"*>J' and tP*^' and ,^1 'J^ ,
or 0*^' and some reckon a*^' 'J^ an eighth:

'J^ :
but authors hold these names to be post-

most
classical: accord, to Esh-Shereeshee, they are seven
days; four of the last [days] of February, and three of
the first [days] of March: during these days blew the
wind by which the tribe of 'Ad was destroyed: and
1 lUd., p. 70 ff.
they are thus called because they are [in] the latter part
(jr>^ ) of winter; or because an old woman Oy^) of
'Ad concealed herself in a subterranean excavation,
from which the wind dragged her forth on the eighth
day, and destroyed her or y^ and j^'y are the names
:
of the last two days; the former being the sixth, and
the latter the seventh. Ibn-Ahmar says. The winter is
driven away, or is closed, by seven dusty (days), our
old woman's days of the month; and when her days
come to an end, and Sinn and Sinnabr, with El-Webr,
and with Amir and his little brother Mu-temir, and
Mo'allil, and with Mutfi-el-Jemr, pass, the winter goes
away, retiring quickly, and a burning wind comes to
thee from the first day of the ensuing month. "^
At a later time other Arabic names were used for the
February intercalation,^ which itself is known under
the name of ^^ sab' at, and begins on February 24
and lasts to March 4. The Arabs brought the super-
stition to Spain, where the witches' night is conse-
quently still known as "la noche del sahado," perpetu-
ated in English as the witches' sabbath, and by Goethe
as the Hexen-sabbat. But the German Hexe is itself
the Arabic Jj^ 'aguz, Berber haguza. The early Anglo-
Saxon vocabularies translate haegtis, haegtes, haehtis,
hegitisse by "Eumenides, Erenis, furia, " and similarly
we have the OHGerman hazus, hazes, hazis with similar
meanings. Unfortunately no earlier text contains any
explanation as to the nature of such a hag, but in the
Ancren Riwle the seven capital sins are called the
seven hags (seouen heggen),^ which agrees with the
conception of seven hags among the Arabs.
' E. W. Lane, An Arabic-English Lexicon, p. 1961.
2 E. Destaing, Files el coulumes saisonniires chez les Beni Snoits, in Revue
Africaine, vol. L, p. 244 flf.
2 J. Morton, The Ancren Riwle, London 1853, p. 216.
While the intercalary days were considered to be

useless,they were celebrated with good cheer, and a
simple porridge, having some religious significance, is
considered indispensable.^ In some places they eat
barley porridge with oil poured over it, or gruel made
with milk, to which salt butter or oil has been added, or
pounded wheat boiled in water. The Arabs of the
Hiaina eat on New Year's Eve UiSa made of wheaten
meal boiled in water to which, while still boiling, are
added salt, milk, and salt butter. This tSiSa had origi-
nally medicinal properties. Ibn-al-Baitar mentions it
as <Jr~i- gaSlS, coarse flour made of wheat or spelt, and
refers to Galan and Dioscorides as to his authority that
it is very nourishing and easily digested.^ It is also
known as d§l§, and the Lri~* hasiS mentioned by me
before^ are other forms of the same word. They are all
from Coptic oouS "porridge," which is frequently re-
ferred to by the Greeks under the name of ddapa or
&d-/]pa, and mentioned by Pliny as of Egyptian origin,
but it is really a Semitic root, found also in Arabic as

jr^ "to crush (grain)." This very ancient
'a§ara
custom of eating porridge on a religious festival is
universal in the Western Sudan; among the Bambaras
it is known as dege. But the non-Islamic Bambaras
have also the fermented grain, dlo, the maize beer,
which, as has already been pointed out,* is frequently
produced by mastication.
In America where this religious custom has led to the
use of the phonetically identical chicha, masticated and
fermented mash, and to the use of pulque in Mexico, we
have thus a direct reminiscence of the African habit,
1 Westermarck, Ceremonies and Beliefs, p. 57 ff.
2 L. Leclerc, Traite dessimples par Ibn El-Beithar, in Notices et extraits des
mamiscrits de la BibliotMque Nationale, vol. XXIII, No. 485.
5 Vol. II, p. 114 f.
* Ihid., p. 115.
which itself, through the Berbers and Arabs, goes
back to Egypt. The purpose of eating the porridge
and generally gorging oneself with food during the in-
tercalary days is due to the desire to produce a baraka
"a blessing," a return of fortune for the ensuing year.
Thus "the Arabs of the Hiîna on the day in question
take some barley to the field, put it into the keskas, or
steamer used for the making of t'am (seksu), leave it
there for a while over the fire, then dry it in the sun,
roast it in an earthenware pan, grind and sift it, and at
last mix it with fresh milk or buttermilk together with
the root of a plant called huzeffur. This is eaten to
destroy the has (evil), it makes the people strong as
—
there is much baraka in it but only on the condition
that the rainbow is seen on that day; otherwise, the
baraka in it is slight, and if it thunders then there is
none. "^ As there is baraka in New Year's Day, all
observances at that time have a religious significance.
Among the Mandes, as elsewhere, barika, barka means
"force, vigor, energy, benediction, gratitude." But
more usually another Arabic word has taken its place
through the Sudan, as it has among the Berbers and in
Europe.^ The Arabic ^ na'am "to live in ease and
affluence, enjoy life, grant, rejoice, overwhelm with
riches," nu' m
"luxurious life," naîma "well, excellent,"
ni'am "benefit, bounty, favor, mercy, kindness" have
left a long trail behind them in western Africa. We
have the Berber nam "to approve, accede to the
desires, obey," nama "grace, favor, benefit, prosperity,
abundance." These lead to Peul nammude "to be
abundant, much," nammaduru "great abundance,"
hence Nammadlri "the Futa, the country of abun-
dance;" Hausa naima, nema "kindness," nima "pleas-
^ Westermarck, Ceremonies and Beliefs, p. 77.
^ The European development of this word and idea I shall treat at another
time.
ure," niam "prosperous;" Timne namfa "to thrive,"

namra "to satiate, be satisfied," namsarne "overeat;"
Dahtome neme "good;" Asante nem "to be diligent,
assiduous, sedulous, careful;" Bambara nema "God's
gift, fortune, ease, satisfaction, luck;" Bambara,
Malinke nyuma "good, benefaction, generous," hence
a Bambara says "barka da ma ye a ka nyuman na, to
thank one for favors bestowed" and " Alia m'i nyuman
segira, may God grant you a safe return." We know
nothing of the Mande calendar, but there can be no
doubt that many of the holidays or religious festiv-
ities have their origin in the solar or lunar events,
and that the celebration with bountiful food and dege,
and with masquerades is identical with the similar re-
ligious observances among the Berbers, that is, that
the intercalary days, though "useless" for the counting,
were a source of baraka and led to "abundance, increase
of flocks," etc., if properly celebrated. Ultimately, no
doubt, the origin of these festivities was forgotten in the
Sudan as much as among the Berbers and Arabs, and
the celebration was transferred to other occasions as
well.
In the Mexican calendar the eighteen months of
twenty days were followed by five intercalary days,^
called nemontemi "superfluous," which were considered
useless and to which no divinities were attached. On
these days people did not quarrel, for fear of laying a
foundation for a year of dissent, and, in general, kept
from doing that which they did not want repeated
during the year. We have here exactly the same days
as in Africa, during which that had to be done which
would bring nama "grace, favor, prosperity, ^.bund-
ance. " There can be no doubt as to the derivation of
Aztec nemontemi from a word which also led to nen
1 Seler, Gesammelte Abhandlungen zur amerikanischen Sprach- und
Alterihumskunde, vol. I, p. 510 ff.
"superfluous, vain." The precise meaning can be
ascertained from a study of the root in the languages
bordering on the Nahuatl. In the Maya languages,
Huasteca does not record any word from this root, and
Tzendal and Choi have no such word for "large,"
while all the other languages have some derivatives
from it. Maya has num. as an ending of cardinal
numbers meaning "times;" in composition it means
"very, too much, greatly;" it does not exist as an indi-
vidual word. In the other Maya languages nim means
"much," hence Pokonchi nim "large," nimaj "to
obey," nim kij "holiday," Kekchi niman "to grow,
become large," nink "large, to provide with things, be
rich," but here we also have nume "to surpass, over-
take," numta "to surpass, too much, be overripe,"
Kiche nim "impulse, large, fat," nimah "to obey,
revere, extol," nimar "to grow." Similarly we have
Tarascan nimani "to pass (the time), " nimaqua "later,"
namucheni "to be numerous," nimatehpei "to be a
grandfather," ninini "to mature." In all of these
cases we proceed from the meaning "abundant," as in
the African words from the Arabic nama. In Nahuatl
we have nemi "to live," that is, "to grow, feed," as
may be seen from nemitia "to live, nourish oneself,"
nemilia "to live from one's labor, to think, consider;"
but it also has the connotation "too much," as in the
Maya languages, hence nen "superfluous, gratuitous,
useless." The root in these languages is distinctly an
intrusive one, and is unquestionably due to the haraka
of the intercalary days.
The most extraordinary object for a haraka among the

Berbers consists in the game of ball, which is connected
with the intercalary days and with other religious
observances. We have already seen that ball is played
during the haûza days. Similarly "games of ball are
frequently played with a view to obtaining rain.
Among the Ait Warîn two or four naked women for

this purpose play a kind of hockey, not, like the
women of the Tsui, with ladles but with sticks. Among
the Ulad Bu-'Asiz some good old women play at ball
when rain is wanted, whilst in the Hiîna under similar
circumstances the men of two neighbouring villages
have a football match in the afternoon, after which
they drape a ewe with a woman's shawl, as has been
said above. Among the At Ub5,hti men and youths
in spring play at ball with sticks, as a means of pro-
ducing rain. It may be asked why games of ball are
supposed to have a rain-producing effect. An ex-
planation given me by an old Arab was that the
ball is dark like a rain-cloud, but the accuracy of
this statement is doubtful. Among the Tsui the men
play at ball to put a stop to a long-continued rain,
the scribes and students playing with the feet and
other men with sticks, and I was told that in Andjra
one game of ball is played to obtain rain and another
to obtain dry weather. From these facts we may con-
clude that the essential function of playing at ball as a
weather-charm is to bring about a change in the
weather through the movements and changing fortune
of the game. "^ "It should be added, however, that
both the tug-of-war and games of ball are believed to
have a strengthening or purifying effect and are also
practised on that account. So far as the former is
concerned, I have given instances of this in my article
on the popular ritual of the Great Feast in Morocco and
above, whilst games of ball are reported to be played
both at the Great and Little Feast (Ait Warain) or on
the last day of the year (Garb), and are in certain
cases expressly said to remove evil influences. The
ball —
in Arabic called l-kora, in Berber taSurt (Ait
—
Warain) or taSurt (At Ub^hti) is considered haunted;
1 Westermarck, Ceremonies and Beliefs, p. 121 f.
in the Hi&ina it must not be taken into a house, nor
must a game of ball be played in the yard or close to the
domestic animals, but it should be played on the waste
"^
land so that all the has, or evil, goes there.
Doutt6 has given an elaborate account of the games of
ball in northern Africa: "A very popular sport among
the Rehamna, as elsewhere in the whole of northern
Africa, is the game of ball or kurah. It is played there
in three different ways, as we shall describe one after
the other. In the first, the players are divided in two
camps and each camp attempts to throw the ball into
the adversary's camp. This ball, which is made of
wool and is covered with leather, is thrown between
the two camps, and is kicked off by him who can
reach it with his foot. When the ball has reached
the camp, in spite of the camp 's efforts to push it back,
the camp loses the game. One recognizes in this the
game of soule au pied of our old France, which is in
vogue especially in Normandy and Brittany. The
English are said to have borrowed it from us during the
Hundred Years' War, and we have received it back
from them as a novelty under the name of football.
"At other times the kurah is played, not by kicking it
with the foot, but by striking it with a stick, aqfah,
'
which is bent at the end. No one has a right to touch

it and only the stick must be used. The attempt is
made to ward off with the stick the adversary's blow
about to be given to the kurah, in order to send it to
his own camp, for in the kurah played with the stick
each camp tries to get the ball to its own side, and
whoever succeeds in doing so wins the game. This
game is no other than our old soule d, la crosse, which,
carried by our Norman and Breton colonists to Canada,
has there become the national game; up to the XIX. cen-
tury our game of billiards was played with lacrosse bats.
> Ibid., p. 122 f.
Fiu. 85. — Lc jell tic 1,M konm on Algc'ric, a El Miliii
Clirlir de »t. Mt'nrlrel:
From Doutte's Merrakech.

"The third way of playing is much more brutal:

the ball is thrown up, and he who catches it must throw
himself upon the ground, turn a somersault with his
hands (iUeqleh), hit his nearest neighbor, and then in
his turn throw up the ball. He is not allowed to throw
it before making the somersault and giving a blow.
One must marvel at the dexterity with which the
natives execute this complicated rule. The blows,
which are generally given with the feet, are very violent
and make the game very brutal. The people are not
divided into cainps, and the most resistant fellow is the
victor. This form of the game has naturally a more
popular character than the other games. In some re-
gions of Morocco they play this kind of game with
a helra, or slipper, in the form of a kurah. This is the
case among the Siadma. Here the players place them-
selves in a circle; one of them holds the helra and throws
it to another player at his will. This one catches the
helra, turns a somersault on his hands, at the same time
trying to hit another, who tries to evade the blow, and
he throws the helra to a third person, who proceeds with
the game. If the person to whom the helra is thrown
is unable to catch it in its flight, he picks it up and has
to pass it to his neighbor without turning a somersault,
whereas he who gets it in the regular course may throw
it to whomsoever he pleases.
"As we have said, the game of the kurah is played in

the whole of northern Africa; it is especially popular in
all of Algeria, where it is generally played with a stick.
The players form two camps, and each of them is pro-
vided with a bat. A level piece of ground is usually
chosen; when everybody is gathered, one of the players
throws the kurah up into the air; around him the
players are waiting, with bat in hand, trying to strike
the ball in such a way as to bring it into their camp.
The m^lee never ceases, all the players throwing them-
selves upon the ball, some to drive it to one side, others
to the other side. Often they make use of their bats
to keep off the adversaries; veritable hand-to-hand
fights take place ; the players strike with all their force,
and many a shank gets the blow intended for the
kurah, and fractures are not rare. The ball is generally
made of wood, more frequently of rags, or of wool or
cow-hair. The game is generally without stakes, but
cases are known when there is a stake, a goat, a lamb,
and sometimes even an ox, which is eaten together at
the expense of the camp losing the game. In certain
regions, at Miliana, for example, the rule of the game is
more difl&cult; the ball has to be returned not only to
one of the camps but also to a hole dug out in the ground.
The game of the kurah is at Algeria also played without
a bat; in this case it bears in the Little Kabylia of
CoUo and Jijeli the name of dukha; the ball is thrown
by hand.
"It is most remarkable that in Morocco the kurah is
generally played by the tolba, the priests, which is not so
common in Algeria. In Morocco itself, the game of
the kurah, without bat and with the feet, is the mon-
opoly of the tolba, and only they play it, at least in this
fashion. Thus, for example, in the Haha, where we
have made a special study of it, only the tolba play at
the kurah, and in the following manner: they are, to
begin with, divided into two camps, and they kick
alternatingly the ball with their feet; by degrees the
camps approach each other, and a veritable fight takes
place. Everybody tries to throw him who comes near
to the ball, in order to send it ahead with his foot, but
they are not allowed to use their hands they may push
;
only with their chests, or shoulders, or legs, or feet;

they may even kick in the shin, but they must keep
their arms down. At Mogador, everybody plays at the
kurah with the feet, but without kicking it: only the
tolha, who play

separately, follow the method used in the
Haha, so that this game is at the same time almost a
battle. This way of playing seems for the rest to be the
most popular among the tolba of Morocco.
"This curious specialization of the ball game as being
in some way an attribute of the tolba, the clerics, occurs
also in other places without going any further than our
;
own country, we know that in the Middle Ages they play-

ed ball in France in the churches up to the revolution the
;
Bishop of Avranches and his canons played on Shrove

Tuesday a game of lacrosse in the cemetery, and the
signal for it was given by ringing all the cathedral bells.
It is precisely at the same time that in northern Africa
they play at the kurah. Indeed, in a very large number
of countries, it is played only in springtime, and in
every country in which it is played spring remains the
classic period; especially for the tolba at Aur^s, and
perhaps elsewhere, it is an integral part of the cere-
monies which take place in springtime. But often they
organize in a drouth at any time a game of kurah in
order 'to bring about rain.' Hence the kurah cannot
be identified with a simple sport, because they do not
generally play it during the native feasts it is not, like :
the fantasia, the usual accompaniment of any cele-

'
'
bration. If one stops to consider all these different

circumstances, one cannot fail to see that the kurah,
which takes place during certain solar dates, or to bring
about a change in weather, and which often is the priv-
ilege of a class of people of a religious character, has
entirely the appearance of a survival of agrarian cere-
monies celebrated by a special caste. I do not doubt
that a more extended investigation and one more
precise than mine will some day establish this in a
definite manner."^
1 Doutte, Merrdkech, Paris 1905, p. 318 ff.; see also his Magie et religion
dans I'Afrique du Nord, p. 554.
We shall now investigate the ball game discussed by
Doutt^ from its beginning andin its ramifications which
concern us here. In Firdusi's Shah-Nameh there are
frequent references to the game of chaugan-goy.
" Chaugan means a bat as well as the ground on which
the game is played, goy means a ball. "^ These two
words are not of Persian origin. Although there is a
word chaul, chul "crooked, " gan means nothing, and no
native etymology for goy is possible. have un- We
questionably here three Chinese words, ts'uh "to kick
(a ball)," kan "a bat," k'iu"a ball," hence the game is
most likely of Chinese origin. It cannot now be de-
cided whether the Arabs first brought it out of China or
borrowed it from the Persians, but the fact that it was
already popular as a royal game in Byzantium in the X.
century would indicate its importation by the Arabs.
The Greeks called the game rlôxdvtov, while rlôxaviâ was
"to play golf or polo" and THûxavcarijpcov was "the place
where the game was played." We have a good de-
scription of the game by Cinnamus: "At the end of
winter, when the weather had cleared, Nicephorus de-
voted himself to the game which since ancient times had
been a special privilege of kings and their children.
The youths divide themselves into two equal parts and
drive a leather ball of the size of an apple to a previously
designated spot. Then they ride to it at full speed as
to a prize placed in the middle, having in the right hand
a stick of fair size, which suddenly ends in a circular
breadth, the middle of which is made of dried guts woven
in the form of a net. Each party strives to drive the
ball beyond the other to a previously designated spot,
for whoever with his racket first reaches the goal with
the ball is considered to be a victor. Such is the game,
which is dangerous and leads to falls. It is necessary
' J. J. Modi, The Game of Ball-Bat among the Ancient Persians,
as Described
in the Epic of Firdousi, in The Journal of the Bombay Branch of the Royal
Asiatic Society, vol. XVIII, p. 39.
161 162
163 164
HOCKEY STICKS, from Artin Pacha's Contribution a Vkude du hlason en Orient.

for those who

take part in it constantly to lean to one
side, and
to turn to the other side, so as to guide the
horse to the ball, and they have to undergo all kinds of
motions, in order to reach the ball. "^
In Persia the game became very popular at court
under the name of chaugdn, which led to Arabic o^j^
§aukan and c^j^ saulân, while the ball itself was
known among the Arabs as 'J^ kurah or °^' 'ukrah.
According to Mas'udi, Harun al-Rasid was the first
khalif to play the game in a polo ground.^ Neither
Persian goy nor Arabic kurah has become the usual
word for "ball" in those African countries outside of
the Berber regions, but another Persian word. We
have from a universal Indo-European root for "hot,"
Persian tabah "tile, brick." It is assumed that
Persian tab "curling" is from another root, but it is
more likely that it evolved from the first. In any case
the two meanings have become united in all the
languages in which these Persian words have been
borrowed. We have Arabic v^ tab game at
"ball,
balls," vj^ tub "brick," -*> tauf "go around." The

Tatar languages have the latter root top "around, " but
it is hard to ascertain whether this is borrowed from the
Persian; in any case the Arabic root is not found in the
other Semitic languages and, although already used in
the Koran, is unquestionably a borrowing from the
Persian.
In the Negro countries this root is very popular.
We have Swahili tufali "a, brick dried in the sun" and
tuffe "ball (to play at tennis or cricket); the natives
put a stone or sand into rags and sew them up or tie
them as a ball." Here the relation of "brick" to
1 Corpus scriptorum Mstoriae Byzantinae, vol. XXV, p. 263 f.
2 Quatremfere, op. cit., vol. I, part I, p. 121 ff.

'

"ball" is clear, since pieces of brick or round pieces of
sunburnt clay were used in many of tbe games where
a ball is generally employed. Similarly we have Hausa
tubali "a ball or brick made of mud." Hassania tob
"brick" is found in Soninke and Bambara as tufa,
but in Bambara we also have kura "round, anything
round," and we are thus still under Arabic influence.
Although we have no information as regards the Negro
ball games, the vocabulary shows that the same rough
games were known there as among the Berbers.
Polo, as recorded for Byzantium, was a seasonal
game, coinciding with the beginning of spring and an
exclusive prerogative of the emperors. Among the
Arabs it was equally a royal pastime, though we have
no reference to a seasonal celebration. Among the
Berbers, where the spring festival, beginning in March,
was generally confused with the Christian New Year,
on the first of January, many of the celebrations refer
equally to one season as to another. Among the
Berbers the prerogative was transferred to the clergy, a
prerogative which unquestionably also existed in Spain,
purely from philological considerations. In Arabic we
have, from the root kiir, f y^ mukawwir "theologian,
magistrate, man of law," because they alone in Spain
and elsewhere were allowed to wear a mikwar turban,'
'
'
but as this is from the root kur, it may also mean "one
who plays ball." As the beginning of the year was
transferred from the first of March to the first of
January, the playing of the ball was identified with
the haûza days, that is, with the time when vetula
"the old woman" was part of the celebration.

The ball games were brought by the Arabs into
Europe, where they became very popular and led to
their introduction among the clergy and, in a rougher
and simpler form, among all classes of people. In 1165
John Beleth, referring to the Church at Poitiers, told of

"a certain December hcense which at time is ob- this
served in certain places. There are certain churches in
which it is customary that even bishops and archbishops
in their monasteries play with their inferiors and give
themselves up to the ball game, and this license is
called the December license, because it was formerly a
custom among the pagans for servants and shepherds
to enjoy in that month the privilege of being equal to
their masters and of celebrating common feasts after
the harvest time. Although some large churches, such
as that at Rheims, observe this custom of playing, it
seems more worthy not to play. "^ The game is re-
corded in France as choula as early as 1152,^ and
later we find for it the names soule, sole, choule, chaule,
solce, cheole, soulette, etc., and ceoler, choler, soler means
"to play, kick the ball." At Berry it was generally
played by the clergy and in the diocese of Bourges it
took place on the holidays of Saint Ursula and St.
John the Evangelist, that is, on December 27 and 29.'
The Arabic c^j^ §aul^dn is also found in the abbrevi-
ated form ^j^ sauli^, ûli^, and from this is formed the
root ^ salaâ "to beat, cudgel." This has led to the
French forms soule, etc. But the Arabic o" >r gokan
produced Portuguese choca, Spanish chueca "football,
golf," which in its turn has led to French choquer,
originally "to kick," then "to shock." But we have
in French the fuller form for the game, namely chicane,
which has similarly led to chicane "chicanery," from
the rough and brutal way in which the game was played
by the populace. We have also derivatives from the
Arabic v*^ tab, namely French taper "to tap," etc.
1 W. Mannhardt, Wald- und FeldkuUe, Berlin 1904, vol. I, p. 477.
* Ducange, sub houla.
' H. F. Jaubert, Glossaire du centre de la France, Paris 1864, p. 623.
That the Romance words of this class are really de-
rived from the polo game follows from Arabic v^M»

tdbtab, vW* tahtab "bat for tennis or polo," obviously
from v^ tabah "ball," hence Arabic yU^ tahtab also

means "to tap, knock lightly." In the Germanic
languages we have the earliest and most striking de-
rivative from the "polo" words,
produces Gothic and
for here Arabic
OHGerman
^
salaga "to beat"
slahan "to beat." OHGerman slaga "hammer" shows
how "to beat" developed from "the polo bat." In
Anglo-Saxon, the word is not recorded in the Leyden
Glossary and but sparingly in the Corpus Glossary
with one meaning "coUisio, bellicum, " exactly as in
the French souler.
In Africa and in Europe the ball game became asso-
ciated with the intercalary days and gained a semi-
religious significance. Precisely the same has taken
place in Mexico, where the names connected with the
game, derived from the Arabic, bear conclusive proof
of the African origin of Mexican culture. Torquemada
described the Mexican ball game as follows: "These
Indian tribes know the game of ball as we play it,
though they play it differently from us. The place
where it was played is called tlachco, which is like our
cricket ground. They used to make the ball from the
gum of a tree which grows in the hot lands and which,
when tapped, distills thick, white drops that very soon
congeal and being pressed and kneaded become blacker
than pitch. From this ulli they made their balls,
which, although heavy and hard to the touch, were
well adapted for the manner of their playing. They
were as resilient as air-filled balls, and better, too, since
they did not have to be blown, nor did they play to
stop it, but to win it, just as at the chueca, which is to
MEXICAN BALLGROUlXD, from Seler's Codex Borgia.
kick it to the wall held by the adversary or to pass it

overhead. They kicked the ball with the haunches
or the buttocks, and not with any other part of the
body, for every other stroke was foul. It was agreed
by them that the game was lost by him who touched
the ball with any other part of the body than the haunch,
buttock, or shoulder, and this was done very gracefully
by them, and, that the ball might more easily rebound,
they bared themselves and wore only the maxtlatl,
which was their loincloth, and put on a stiff piece of
leather over their buttocks. They could make it re-
bound as they pleased, and it did rebound in rapid
succession so as to appear like a living thing. They
used to play two against two or three against three,
and sometimes two against three; and in the chief
tlachcos there played lords, princes, and great players,
and, to celebrate their market days, especially the days
of the fair, they went there to play, and they played for
stakes, such as a bundle of mantles, more or less, accord-
ing to the players' wealth, whether they were kings,
cities, or municipalities. They also played for gold
and feathers, and sometimes they put themselves up for
stakes. This tlachco was in the very market square,
but there were also others in other parts or wards. The
playgrounds were built in such a way as to form a street
of two heavy walls, narrower at the bottom than at the
top, so that although the players were in the narrower
part, the game widened out overhead. The greatest
width of the space was twenty fathoms, and even less,
and in some places these tlachcos were topped by very
curious battlements. The walls were higher at the
sides than at the ends. To play the more easily, they
had them whitewashed and smooth, even as they
fixed the floor. At the sides of the walls they placed
certain stones, resembling millstones with their holes,
in the middle of the grounds, whither the ball rarely
reached; and lie who put the ball through it "won the
game, and, as this was a rare victory, which only few
attained, the mantles of all the onlookers belonged to
him by an ancient custom, and it was considered a bit
of fun, when the ball went through the hole in the
stone, for the people to laugh and run off with their
capes, while others deprived them of their mantles for
the benefit of the victor; but he was obliged to make
certain sacrifices to the idol of the playgrounds and the
stone, through the hole of which he put the ball.
"Upon seeing this trick of putting the ball through,
which to the spectators appeared a miracle, which it
was perchance, they said and asserted that the winner
must be a thief and adulterer, or that he would die soon,
since he had had such a piece of luck; and the memory
of this victory lasted for many days, until the next
victory made them forget the first. Each playground
was a temple, because they placed in it two images, one
of the god of the game, the other of the god of the ball,
above the two walls which were lower toward midnight
(north?), on a day of good omen, with certain cere-
monies and sorceries, and in the middle of the ground
they made other similar ones, singing songs; then there
came a priest from the chief temple, with certain min-
isters, to bless it (if this detestable superstition may be
called blessing), and he spoke certain words; he threw
the ball four times to begin the game, and with this
they said that it was consecrated and they could play
with it, and not before. This was done with much
authority and attention, because they said that with it
came the ease and peace of the heart. The master of
the playground, who was always a lord, did not begin to
play ball until he had carried out certain ceremonies and
offered sacrifices to the idol of the game, from which it
may be seen how superstitious they were, since even in
matters of pastime they held such communion with
their idols. Montezuma sometimes took the Span-

iards to this game, because he considered it good.
Lords and chiefs went from one town to another, and
brought with them good players, to play against others,
and they gave this game more attention than we do,
and those who played better or won made fun of the
others and said: 'Tell yom* women to hurry up the
spinning, because you will need other mantles.'
Others said: 'Go to such and such a fair to buy clothes,'
and with this the spectators laughed. They served the
ball, and if it was not good they would not accept it.
Later, when it began to pass, those who threw it over
the front wall or struck the wall, got a score against
them; or if they struck with the ball against the ad-
versary's body or played badly, that is, not with the
haunches, they got a mark also, and the whole game
consisted in these scores; and they did not drive the
ball back. People made bets for one side or the other,
and thus there were more stakes than in the principal
of the game. Those who played either loudly or ment-
ally called upon a demon who, they said, was of
prominence in this game to help them. Of the good
player who was successful in the game they said that
his good fate or luck or sign under which he was born
had helped him, while the loser's misfortune they as-
cribed to his bad sign. "^
Seler has shown that the Mexican ballgrounds were
generally placed strictly from north to south, which
pointed to their relation with the vernal equinox,
especially since the god Tecatlipoca is shown playing
the game about the beginning of the New Year, which
coincided with the vernal equinox.^ On the other hand
the twin god Xolotl is the regent of the seventeenth
day sign, olin, which means "rubber, from which the
> J. de Torquemada, De la monarquia Indiana, Madrid 1723, vol. II, lib.
XIV, cap. 12.
"

ball is made. " Xolotl is distinctly mentioned as the god
of the ballgame.^ It has already been shown that the
god Xolotl is derived from the Arabic -A* 'aiara in its
various connotations. But the Arabic JJj^ 'aiura,

that the tenth day of the month of Muharam, which
is,
approximately coincides with the vernal equinox, is

also used for the New Year, on which games are played,
as already discussed at great length. We thus see that
as 'aSura in northern Africa became identified with the
carnival time during which the ballgame took place, so
in America this 'aSura of necessity led to the identifica-
tion of the god Xolotl with the ballgame, since Xolotl
is philologically derived from the same root.^
Thus it is made clear that the ballgame in vogue

throughout the regions visited by the early voyagers is
of African origin. This can be further proven by the
very vocabulary connected with this game. In Caraib,
according to Breton, the gummy substance from which
made and the ball itself are called tibueli.
the ball was
The same name is found in the Aztecan tapayolli, tapa-
yulli "ball;" both are ultimately from Arabic '^^ *^
tabah 'alek "rubber ball." This alek is found in the
'
Mande languages as Mende hole "gum," Mandingo

folio, Malinke fole "rubber, landolphia owayensis.
In the West Indian islands the ball and the place where
it is played is by the early writers given as batey, but
this is not a native word, for Durand informs us that
batel was the Spanish term for "rubber."^ That the
first part of the Nahuatl word is from Arabic tdbah
1 Ibid., p. 193.
="
See p. 279.
' "U&mase la materia de esta pelota olin lo cual en nuestro castellano he
oido nombrar por este nombre hatello cual es una resina de un 4rbol particu-
lar que cocida se hace como unos nierbos," D. Duran, Historia de las Indias
de Nueva EspaHa, Mexico 1880, vol. II, p. 244.
tel
o
<o
Sf2
follows from the fact that the same root in the com-
pound means "brick, tile, potsherd," exactly
tapalcatl
as in Arabic. The name of the ball-ground and of the
game itself is in Nahuatl tlachtli, where tlach represents
the Germanic slag, for Arabic ^
salaâ "to beat,
kick the ball." Coptic 6loi "ball" indicates that the
Arabic had not only a form ^>« ûl^, but also ^j^ slu^
or ^ sla^ for the game. From this is derived Nahuatl
"the ballgame." The ball grounds are repre-
tlachtli for
sented in the picture writing by a long H-shaped space,
in which the two stones containing the holes through
which the ball has to pass are represented in the middle
of the structure, but more frequently a spindle crossing
the picture represents the millstone. This is due to
the fact that the hole in the millstone is, like the
spindle, called in Nahuatl malacatl, generally temalacatl,
in which te means "stone." The strange relation be-
tween "spindle" and "hole" is at once made clear from
the Arabic words from which the two are derived.
Arabic 'i^^-* maslak means "a breach, a place of pas-
sage," from "^^ salaka "he inserted (a thread in the
eye of a needle)," while "^^^ mislakah is "a reel upon
which thread wound," so that the two words from
is
the same root lead both to "hole" and "spindle."
The technical term for sending the ball through the
aperture seems to have been in Nahuatl petla, for this
verb means "to throw oneself through the ranks of the
enemy, make a breach, " while petlaua is "to take away
the clothes," which is the forfeit the spectators had to
pay to the winning party. But the same root has some
other, totally different connotations. Thus petlani
means "to shine," petlania "to make an object shine,"
and petlatl is "the mat" on which the Indians sat and
slept. would seem impossible to connect these
It
meanings, which, however, are found side by side in the
Mande languages. There is a universal Semitic root
fitil "wick, cord." We have Assyrian pitiltu "cord,
loop," Hebrew /O? pdtil "cord," which leads to
Arabic *^ fatllah "wick." But we have also Egyp-
tian petr "cord, thread, wick of a lamp. " In north-east
Africa, in Soho, we have fatal "to plait, weave." In
the Sudanese languages the word is universally in use.
Hausa has fatilla, fitilla "lamp" and Songay has
fitila "lamp" and fitina, which has nothing to do with
this word and is from Arabic if^ fatn, with the meaning
"war, rebellion." In Bambara the two have become
confused,and fitne means "lamp, war, rebellion."
The meaning "to weave," found in Soho, is very old,
for we have, outside of the Semitic languages, Persian
patil"mat." In Nahuatl we have both "mat," and
the same confusion as in Bambara, namely petla "to
make a warlike onrush" and petlaua "to shine," from
the meaning "lamp." The Nahuatl petla produced
Kiche petel "spindle," which is found in most of the
Maya languages. We have Kekchi petet, Pokonchi
pitejt, pechech "spindle," but in Maya we also
Maya
have pechatah "to crush with the feet, " pech' "crushed,
oppressed, flattened," pet "circle," petelpet "flat,
round things," hence peten "island." This idea of a
flat round thing in Maya is apparently derived from the
flat round stone, through the hole of which the ball had
to pass. In any case, already Arabic J^ fitila means

"to turn in a circle."
The resemblance of the American ball game to the

African, its semi-religious signiflcance in both continents,
its celebration during the vernal equinox or at the be-
ginning of the New Year, whether it fell in March or

January, the identity of the nomenclature connected
with the game, in Mexico and in Africa, preclude all
chance development: the African civilization was
not transferred to America piecemeal but as an organic
whole.
CHAPTER XIV.
CoNCLtrSIONS.^
1. It is yet too early to write the history of the ad-

vance of civilization from its incipiency to the present
time. We do not possess sufficient data to assert
whether the trend of civiUzation was from east to west
or from west to east, but within historic times it appears
that most cultural influences proceeded from east to
west, from a central region in Asia towards Egypt and
Europe. While some such influences began very early
in the dimness of time, other benefits of civilization
proceeded with extraordinary slowness, not merely
centuries, but whole millenniums being consumed in
the establishment of practices essential to the life of
the eastern nations. It is useless to expatiate on such
differences, —
we can only ascertain the facts. Thus
cotton, known to Assyria and India long before the
VII. century, imported into Egypt and, possibly, into
Greece before the beginning of the Christian era, never
roused the peoples that received the valuable products
by commerce to introduce the plant at home. Iron,
which the European nations knew since earliest times,
remained a rare article along the Nile, and the best
steel, long in use in India and China, and imported into
Rome together with the eastern products, was almost
totally unknown to the metallurgists along the Med-
iterranean until the arrival of the Arabs. (II. 2-32,
III. 1-18.)
2. The Arabs had unquestionably been acquainted
with the use of cotton before their westward move-
' The references are to the volumes and pages of this work.
CONCLUSIONS 353
ment, but the fact that one of the names for "cotton"
is derived from an Egyptian, or, rather, Coptic word
for "rehgious purification" indicates that the Egyp-
tian treatment of the dead, which must have descended
in substance to the Copts, determined an intensive use
of this material in the religious observances of the
Arabs, and the extensive distribution of this "purifi-
cation" word for "cotton" through the northern part
of Africa shows that it was chiefly its ritualistic use
which led to the cultivation of cotton throughout the
Sudan region. Wherever the "purification" word was
not adopted, as in the Mande languages, it shows once
more that the Sudan owes the cultivation of cotton
for industrial purposes to the Arabs. The cotton
plant was chiefly not native to Africa, for not a single
reference to it or representation of it on Egyptian
monuments has been discovered. (II. 18-22.)
3. It is possible that the Bantu peoples knew the use
of iron even in antiqioity, but of this no proof exists.
The new impetus to the iron industry in Africa was
given by the Arabs, who popularized the Hindu and
Chinese methods of hard steel manufacture in the west.
The ancient alchemy, which was concerned with the
production of mineral compounds and metallic objects
for industrial purposes, did not advance much, if any,
beyond the traditions established for ages by the
metal-workers of Egypt and Greece. The Arabs, in
inheriting the science, enriched it at once by new pro-
cesses observed in their wide commercial ventures in
the Indian seas and in China. The result of this
semi-scientific interest of the Arabs may be observed
in the extremely large number of alchemistic ex-
pressions and concepts which have entered into the
languages of Europe. The effect of this activity, to a
lesser extent, is seen in Africa in the greater use of iron
among the Sudanic tribes in the Middle Ages. (III. 1-53.)
4. The most remarkable westward movement of
civilization may be observed in the enormous ex-
tension of shell-money. Known to China and India
in the dimmest antiquity, but almost entirely unknown
to Europe and western continental Asia, the cowrie
industry and trade, bearing in the vocabulary the direct
evidence of its Arabic initiative, in the Middle Ages
reaches the shores of Guinea, is recorded in the Malli
kingdom, and takes a firm footing off the shores of
Angola. The direct relation of the African cowrie
currency to its Asiatic prototype is unmistakable.
The ancient Chinese and Hindu double standard of
white and blue cowries dominates the African trade at
Angola and in Guinea, and leads to the predominance
of the blue currency, hence to the adulteration of shell-
money by Chinese and Venetian glass-blowers, who
thus establish the unprecedented popularity of blue
beads and aggries in the Sudan. (II. 203-248.)
5. We possess no satisfactory account of the trade
routes that converged at the bend of the Niger River,
though we have much indirect evidence, especially in
the vocabularies of the Sudanic peoples, to show the
probable caravan routes over which the gold mines
lying between the Mande and the Asante nations and
the pepper-bearing regions were reached since Arabic
occupation of the northern part of Africa. The same
routes which the Hausas still use in their trading ex-
peditions from Egypt and Tripoli were traveled over
by the Arab merchants and their Negro servants in
ancient times. The Berber element in the Mande
languages is of the Zenaga type in the extreme west of
Africa, whence it appears that either land routes or
sea lanes hugged the shore about Cape Bajador. But
the most intimate intercourse lay from the Magreb,
that is, from Morocco and Algiers, over the Sahara
and the Arabic oases and settlements in the neighbor-
CONCLUSIONS 355
hood of Lake Chad. The very considerable linguistic

element of the Hassania Arabic in most of the lan-
guages of the Western Sudan is proof of the close relation-
ship between the Magreb Arabs and the peoples of the
Niger plateau. (II. & III., Philology, in Word Index.)
6. Since the foundation of the Gana state in the VII.
or VIII. century until modern times several more or
less powerful states arose in the Western Sudan, the
Malli, Songay, Soso, Peul, Benin, Asante, Hausa
kingdoms, some of them giving evidence of considerable
cultural ability, though seldom rising above a bloody
autocracy. The information we have about their his-
tories is scanty in the extreme, but certain facts stand
out prominently enough to admit of historical de-
ductions. In the first half of the VII. century the
Hindu Zotts, who had for some time hved in Persia
as musicians, received certain privileges from the Arabs
for their acceptance of Islam, and with their buffalo
herds began a westward movement through Africa,
ultimately reaching the bend of the Niger, where they
settled, and, in all probability, formed the substratum
of the lighter-colored Negroized Peuls of later times.
They brought with them the combined Hindu and
Syrian traditions of civilization, and introduced the
agricultural habits, for which the Peuls are known even
to-day. (III. 111.)
7. In the midst of these Zotts were the blacksmithing
Hindu Gypsies, who moved westward with them, and
from Egypt, in the naiddle of the VII. centiiry, began
their penetration of Europe and of Africa. Being
nomads, blacksmiths, and cheats, they typified the
Biblical descendants of Tubal Cain and of the Egyp-
tians, upon whom was a curse that made of them
"slaves of slaves," neither servants nor again free men.
It is not unlikely that the Hindu low caste which they
represented had something to do with the formation
of the low caste pariahs, as which they appear through-
out northern Africa. Thus we have the anonaaly that,
although the blacksmiths were respected for their
craft, they were in the contemptible class politically,
and Sudanic society developed, side by side with the
native free man and the slave, also the caste of the
griots, the Gypsies without a political and social
status, without privileges and without duties, hence
more nearly akin to the free men than to the slaves,
and frequently as wealthy and as powerful as the
upper class. (III. 78-100.)
8. The pariah condition of the griots is, if not
originally, yet preeminently the result of Arabic leg-
islation. According to Islamic law not only were
criminals dealt with severely, but individuals and
classes that were likely to develop criminal tendencies
were visited with premonitory punishments. The
name of the griots in the Sudan, more especially among
the Mandingos, not only means "cheat, musician,"
but also "tarred," unquestionably from the punish-
ment meted out to the Gypsies. But "tarring and
feathering," mentioned by Richard Coeur de Lion,
clearly as an eastern method of punishment, explains
the method of pasting feathers about the head and
appearing with a bird mask, adopted by the griots as
their distinctive mark, as which the bard and dancer
before the king of Malli is by Ibn-Batutah described
in the middle of the XIY. century. The griots are the
preservers of the ceremonial and religious dance, the
bards and parasites at court, the fortune-tellers and
medical quacks, the executors of Islamic rites, the
blacksmiths and metal-workers, in short everything
that characterizes them as Gypsies. It is not im-
possible that some of the fetishistic practices, not
definitely derived from Arabic sources, are, through
the griots, of Hindu origin. In any case, the griots
CONCLUSIONS 357
are the mediators, if not the originators and priests,

of what is known as Sudanic fetishism. (III. 100-
107, 213-217.)
9. Many cultural influences were carried by the
Arabic merchant, the tarûman, the interpreter who
penetrated the interior, not only along the northern
trade routes, but also from Zanzibar and Mogambique.
Unfortunately we have no records of the early activities
of these merchants, but from the XVI. century refer-
ences to the tangomaos, tangomanes, as which they were
known to the Portuguese and Spaniards, it appears
that they were composed of all kinds of western Europe-
ans, renegades and outlaws, who infested the mouths
of Guinea rivers, became thoroughly Negroized in their
manner of Ufe, surrounded themselves with black
wives, often ascended a throne, and carried on exten-
sive commercial enterprises in the interior, with the
aid of Negro trusties and couriers. Their cultural in-
fluence upon their retinues and customers was always
considerable. Being unable to trade with Europe,
except as smugglers and traitors, they extended their
operations to the western islands, whatever these may
have been. They were chiefly instrumental in ob-
taining slaves, gold, ivory, in exchange for a few
European articles, such as beads and iron staves, and
transferring native articles from the interior to the
islands of the west. (III. 99-113.)
10. More important, on account of their constant

and intimate relation to the Sudanese people, were the
religious teachers fakirs, the Marabuts, who found
and
their way from the Islamic countries and developed
their fanatical activities among a wonder-loving and
ignorant people. Timbuktu and other cities on the
Niger and in Guinea attracted Moslem teachers and
saints, who developed by
their activities, undisturbed
factional disputes and persecutions. They brought
with them the undercurrent of the Islamic religion, as
tainted by gnosticism in Egypt and in the Berber
countries. With this they frequently combined that
art of the physician, where the physician's realm en-
croached upon that of the religious teacher, through
the health-giving amulet. This activity of the Mara-
but was still more direct in those cases where the Ne-
groes formally became Mohammedans, but even the
heathen Bambaras were strongly affected by the
Moslem saints and teachers, although they did not
subscribe to the precepts of the Islamic faith as such.
(III. 130, 144-150, 163-179.)
11. The chief phylactery of the Mar abut was the
piece of paper or leather with some Koranic verses and
apocryphal injunctions written upon it, hence the
case in which the amulet was placed, the paper or
leather itself, the ink with which the verses were
written became "medicine" and "rehgion." This
led to an enormous multiplication of fetishes in the
sense of "amulets" and "phylacteries." The names
of most of these in the Sudan are of Arabic origin.
The most important amulets were those which had
mystic words, the names of the constellations, the
guardian angels, astrological and astronomical specu-
lations written in squares variously constructed and
known as âdwals. (I. 107-110, III. 268-280.)
12. The medical practices introduced by the Mar-
abut were confined to the simple remedies that had
become the stock in trade of the itinerant quack, such
as cupping and cure of colds, toothache, distempers,
etc. by fumigation, but in all such cases the medical
practice became inseparable from the magical amulets,
and the whole assumed a religious, fetishistic aspect.
The cupping is simplified to suction with the lips, the
magician claiming supernatural powers and surrepti-
tiously introducing an object into his mouth, which he
CONCLUSIONS 359
claims to be the cause of the trouble. The fumigation

may be external or, as in Graeco- Arabic medicine, may-
be taken through a pipe into the mouth, in which case
the tabbaq, the styptic, glutinous substance used in
such fumigations, reduces itself to the particular
glutinous plant of Africa adapted for styptic purposes,
the Nicotiana tabacum and rustica, the sovereign rem-
edy par excellence. In those cases where the Arabic
quack meets the European quack, the European name
bitumen, equally used for such a substance, may pre-
vail, but such is not the case in the Sudan. (II. 85-
134.)
13. The vast amount of fetishistic beliefs which the
Marabut has helped to spread through the Sudan may
be roughly classified into two divisions, one being the
direct result of Islamic teachings and practices, the
second being an undigested mass of gnostic ideas,
some of them of pre-Arabic, Berber or Egyptian pro-
venience, some of them possibly of Negro origin, but
all of them incorporated into the gnostic observances
of African Moslems, and frequently resting on nothing
more than philological calculations. The limits be-
tween the two are ephemeral, and the whole mass of
fetishism is eclectic and syncretic, varying not only
from nation to nation, but frequently also within the
Unguistie group from village to village. Through all
the maze of conflicting and disjointed cosmogonic or
folkloristic stories which refer to the Sudanic fetishism
the fundamental concepts, all of them of Arabic origin,
remain undimmed through the ages and aid us in
bringing order out of chaos and rearranging the relig-
ious motifs into orderly Islamic sequences. (III.
116 passim.)
ff., et
14. The chief Sudanic religious concept, which under-

liesa vast number of specific fetish worships as well as
the larger worship of the invisible god or gods, is based
on the evolution of the Islamic Allah in the Sudan.
From the start he is confused with the heavens, the
sky, where he is naturally placed in Moslem thought.
Hence the Arabic words for "sky," that are more
tangible to the untutored mind, have everywhere taken
the place of the invisible Allah. Arabic gannah "par-
adise, the abode of recompense" on the one hand leads
to the identification of God with heaven, and on the
other leads to the identification with gan " the serpent."
Throughout the Sudan, in Timne, Mande, Yoruba,
Nupe, Bornu, Songay, Hausa, gannah or al-ganndh
forms the word for "heaven, sky," and in Yoruba a
derivative from this means "God." (III. 172-174.)
15. In the Mande languages Arabic sama "heaven,"
suman "rain" leads to the "heaven, rain, year, season"
words, which in Malinke and Bambara are apocopated
to san, sa, sa, and the latter also meaning "serpent,"
sa "the serpent" also became the equivalent of Allah.
But in the Mande languages s and k are interchange-
able, wherefore we also find the forms ka, kan, kali for
"serpent, above," which through Wolof dyan are
clearly derived from Arabic gannah "paradise," gan
"serpent." But here a confusion has arisen with
another Arabic word for "hoe," hence in Malinke and
Bambara we have also ka, kan, kana "he cut with a
sickle." Thus we have here also a confusion of "heav-
en" with "serpent" and "hoe." The relation of the
whole group of these words to the same "paradise"
words in the previous paragraph is made clear from the
Asante language, where the words derived from the
same Arabic sama, suman lead to the concept of the
"abode of the blessed, abode of the departed spirits,
spirit." (III. 174-177.)
16. The untutored Negroes stood in greater awe of
the evil spirits of the Islamic mythology than of the
beneficent §inn, hence the Arabic habbal "a malicious
CONCLUSIONS 361
ginn, Satan" has led to the legion of the Sudanic boli

or bori, the chief elements of the northern fetishism.
Just as the Arabic habbal is connected with the idea of
obsession, so the African boli is closely associated with
obsession and epilepsy, and leads to practices generally
connected with these, such as mad dances and delirious
prophetic utterances. The boli becomes the essence
of fetishism and is identified with "amulet" and
"medicine," hence the bolitigi "the master of the
boli" assumes the role of priest and doctor, and oc-
cupies an important place in the religious conceptions
of the Sudanese peoples. (III. 142-162.)
17. To the Negroes the Moslem preacher was just
such a superior person who lived on a familiar footing
with the world of spirits and possessed all the know-
ledge of amulets and "medicine," hence the Arabic
qurrd "holy man, devotee," qard "to recite the Koran"
have led in Africa to an enormous mass of words signi-
fying "reading, sooth-saying, call upon a fetish," and
"the master of the boli" also becomes "the possessor of
the kara," which may be wisdom or "medicine," or a
spirit, a ghost. With many of the nations such a person
is distinctly connected with the Moslem religion, as
in Asante Kramo "Mohammedan," Peul karamoko
"reader of the Koran." (III. 163-166, 190-193.)
18. A "naive Moslem theology," influenced,no doubt,
by Christian ideas, is responsible in the Sudan for the
invisible triune divinity, consisting of the male, fecun-
dating principle, representing the Sun or Heaven, the
female, generating principle, identified with the earth,
and their variously conceived.
eldest child, There
seems to be in an Arabic astrological speculation,
this
as preserved in a Mossi tradition, that the invisible
god had to restrain the sun from burning up the world
by shutting it up at night and placing it in charge of
nine of his children. (III. 293-296.)
19. Similar astrological considerations, as applied to
the calendar, have preserved in northern Africa the
divisions of the solar year, according to which the 360
days are followed by five "useless" days, while the
Arabic lunar mansions have led among the Berbers
to the 28 zodiacal periods of 13 days each, whereas in
western Africa we find the matter reversed, the zodiac
being divided into 13 parts, obviously of 28 days each.
(III. 270 1, 326-331.)
The New Year generally began in March, and
20.
its celebration was confused with that of the vernal
equinox, which comes two weeks later. Thus the
tenth day of the month of Muharam, the Arabic 'aSura,
was identified among the Berbers and elsewhere in
northern Africa with the New Year, and the ball game
or polo, formerly played by Persian kings in the begin-
ning of spring, gained religious significance in Africa,
as it did in Europe. The Berber name of the ball,
taSurt, from this same 'asura, shows that it was intimate-
ly connected with the Arabic word, which, from its
connotations "companion, pot," must have led to a
number of speculations upon these terms. (III. 279-
280, 286-293, 334-344.)
21. Similar speculations upon similar connotations
of the same root
referring to a mythological concept,
and a confusion of similar roots referring to religious
ideas have led in the Sudan to fetishistie worships.
Thus, from the Arabic terms duwd "a species of owl,
the handsomest bird of all" and du'a "blessing, crying
for aid or succor" arises the Mandingo dug a, which
combines both connotations, and leads to their "re-
doubtable bird of omen." (III. 315 f.)
22. Far more interesting is the double confusion of
Arabic nabl "prophet" with 'aman "faith," and the
latter with 'amir "hyena," from which arises the famous
Sudanic Naba-Nama worship, where the fetish is
CONCLUSIONS 363
identified with an old hyena, and the Namatigi, the

master of the Nama, among the Mandingos is considered
to be the chief sorcerer, the holder of wisdom. (III.
160-162, 249-252.)
23. It is not always possible to get at the Arabic
prototype of a particular worship among the Mandingos,
as we possess but fragments of the original beliefs, but
in some cases there seems to be a substratum of Arabic
medicine that has led to fetishistic worships. Such
is, no doubt, the case with Mandingo nana "the itch,
or syphilis," possibly from Arabic nagila "the itch,"

and the related Bambara manyan "the itch," "a
fetish whose pyramidal altar, in Hausa called dakali,
may be seen near the villages." (III. 285 f.)
24. It is similarly difficult to trace the precise origin
of the dasiri worship, which consists in sacrifices under
a upon which the dasiri stays, and beneath which
tree,
is placed a bowl to receive the libations; nor can we
locate the origin of the connection of the long-beaked
Nama with bee-culture, though we may surmise that
in the latter case we have, as in the case of the syphilis
god, a faint medical recollection, honey having been
considered in the Koran as a sovereign remedy. (III.
146-150, 252, 264.)
25. A very powerful Arabic influence in the Sudan
was exerted by the astrological âdwal, a phylactery,
in which the encysted graphic signs written in columns
do not represent phonetic elements, but are intended
for astrological sentences of augury, arising from the
casting of sticks in geomancy. Such signs may be
seen in the rock inscriptions photographed by Des-
plagnes. The signs themselves have distinct names and
in the Sudan the leading ones among them have come
to mean "tale, story." From the connotations "to
chant the Koran" and "spider," for the root ratlah,
has similarly arisen Bambara ntale "story, spider,"
and this has led throughout the Sudan to a mythical
valuation of the spider in Negro stories and beliefs.
(III. 271-278.)
26. Many fetishistic practices are derived from. Sufi
associations, since Sufl sectarians were entrenched at
Timbuktu and elsewhere in the Sudan. To this in-
fluence may be ascribed the African ecstatic dances
and self-castigation, which form an important part of
Sudanic secret celebrations and initiations. At the
same time the secret organizations bear evidence of
being fashioned in the manner of the Moslem brother-
hoods. (III. 163-179.)
27. Outside the religious influences, the Arabs ex-
erted also powerful influences in the political and social
orders of the Sudanic states. In the political order we
find the name of the kingly power, mansa, derived from
the Arabic man§a, while the royal dignity found its
outward expression in the Arabic stool, the Arabic
taht, tuhut, best preserved in the Asante dufua, Soninke
takhade. The Arabic conical hat, qub^ and gifarah,
associated with the Magi, is at the base of all the "hat"
names in the Sudan, and, apparently, was originally
restricted to the kingly power. The priestly power,
with which the conical hat was originally associated,
received for its insignia the Arabic mitraqah, the rattle,
which, through a homonym which in Arabic means
"gourd" and "religious wisdom," was made from a
gourd. 192 f., 218 f., 318-321.)
(III.
28. Among the Bambaras the Arabic niSan, the
distinction of nobility or the king, became as n'tene
a mere totemic sign for a family or tribe. The Man-
dingos preserved the Arabic social distinction of free
men and slaves, to which were added "the slaves of
slaves," the griots, whose chief occupation was music.
Though the Mandingos did not adopt the Arabic
habiliments, those that they wore bore names derived
CONCLUSIONS 365
from the Arabic Hmlah, mai'zar, gifarah, la^dm, Idbis,

bandarlah, Msd. (III. 219-221, 230-239, 318-321.)
29. The presence of Negroes with their trading

masters in America before Columbus is proved by the
representation of Negroes in American sculpture and
design, by the occurrence of a black nation at Darien
early in the XVI. century, but more specifically by
Columbus' emphatic reference to Negro traders from
Guinea, who trafficked in a gold alloy, guanin, of pre-
cisely the same composition and bearing the same name,
as frequently referred to by early writers in Africa.
(I. 33 f., 174 1, 159-161, II. 116-119, 262 f., 265-270.)
30. There were several foci from which the Negro

traders spread in the two Americas. The eastern part
of South America, where the Caraibs are mentioned,
seems to have been reached by them from the West
Indies. Another stream, possibly from the same focus,
radiated to the north along roads marked by the pres-
ence of mounds, and reached as far as Canada. The
chief cultural influence was exerted by a Negro colony
in Mexico, most likely from Teotihuacan and Tuxtla,
who may have been instrumental in establishing the
city of Mexico. From here their influence pervaded
the neighboring tribes, and ultimately, directly or
indirectly, reached Peru.
31. That the Negro civilization was carried chiefly
by the trader is proved not only by Columbus' specific
reference, but also by the presence of the African mer-
chant, the tangoman, as tiangizman in Mexico, hence
Aztec tiangiz "market," and by the universality of the
blue and white shell-money from Canada to La Plata, and
the use of shells as a coin in the Peru-Guatemala trade.
The exceptional position of the merchants in Mexico,
with the chief worships directly attributed to them,
similarly testifies to the importance of the trader in the
pre-Columbian, Africa-America relations. (II. 249-
270, III. 230 1, 239-245, 259 f.)
32. The African penetration in religion and civic
life and customs was thorough and, to judge from the
survival of the Arabic words in a MaUnke or Soninke
form in America, especially among the Caraibs and
Aztecs, proceeded almost exclusively from the Man-
dingos, either the ancestors of the present Malinkes,
or a tribe in which the Soninke language had not yet
completely separated from its MaUnke affinities.
^ 33. In Mexico we have the same confusion of "god,"
"rain," and "serpent" as among the Mandes, and the

same root coa in Aztec, can in Maya-Kiche, as in the
Mande, leads to a confusion of the three concepts in
Aztec, where coa has also the meaning "sickle," as
in Mande. (III. 307-314.)^
34. Just as in Mande, so throughout America, the
Arabic habal, in forms derived from Mande boli, rep-
resents the idea of spirit or anything related to religion
or medicine. In America, too, the bolitigi, the master
of the boli, appears as boratio and a large number of
forms Unguistically derived from this, with the identi-
cal powers as in Africa and wielding the gourd rattle,
the mitraqah, in Tupi and other South-American lan-
guages denominated maraca. The Mande tigi in the
sense of "master" is also separately represented in
many American languages and in the appellation
cacique, formed by the early voyagers. (III. 158, 222-
227, 228-230.)
35. The Asante kara "rehgious wisdom," Asante
Kramo "Mohammedan" were similarly applied in
South America and the West Indies to the fetishist,
leading to the Tupi carai, caraiba "the foreign sor-
cerer," by Columbus, through an identification with
Cambalu, in China changed to canibal, and applied to
the race, which apparently practised religious canni-
CONCLUSIONS 367
balism, as in some regions of the Sudan, and where the

African rites seem to have taken firm root. (III. 180-
198, 218-221.)
36. As in the Sudan, we have in Mexico the concept
of the male and female divinity, forming with their
descendant a kind of trinity, and designated, as in the
Sudan, the " old " divinities. The detailed resemblance
of the two trinities is shown in the case of the identifi-
cation of the male principle with the sun, which, in
the Mexican belief, as among the Mossi in the Sudan,
is restrained by the nine guardians of the night. (III.
296-302.)
\ 37. Similar astrological considerations, derived from
the Arabic source, have led in Mexico to a calendar
year, as in Africa, of thirteen months of twenty-eight
days each or twenty-eight months of thirteen days each.
But just as in Africa the old Coptic calendar of 360
days has survived, so in Mexico the year of 360 days
is followed by the same "useless" days as in the Coptic
and Berber calendar. (III. 270 f., 333 f., 344-351. )\
38. Just as the Arabic 'aSura was in Africa identified
with the New Year, which began in March, and was
celebrated by ball games, so the Mexican Xolotl, phil-
ologically derived from 'aiura, was identified with the
New Year and the ball game, which assumed a reUgious
significance. The same philological speculations as in
Arabic led to the identification of Xolotl with twins
and as boiling in a pot. The ball game itself in Mexico
uses the same terms for the grounds in which it was
played, and the ball with which it was played, as in
Africa. (III. 279-286, 344-351.)
39. The African bird of omen, from the Arabic duwd
"a species of owl, the handsomest bird of all," led to
Aztec uactli from touactli, for which we find the pop-
ularized toluactli "the Mexican bird of omen," but
we also have for the owl tecolotl, Mayan tunculuchu,
from Malinke duga. The feathers of the handsomest
bird are in Mexican called quetzalli, related to a Su-
danic "owl" word, and the relation of the quetzalli to
the Sudanic owl is seen in the fact that the region
where the feathers are found is called "Owl-land."
(III. 314-316.)
40. The Sudanic confusion of "prophet," "faith,"
and "hyena" is found over a large territory in America
where the Arabic nabl lies at the foundation of "Nag-
ual" words, while 'aman "faith" produces, parallel
to the Sudanic Namaiigi, the Aztec amanteca, Kechua
amauta, Caraib omeoto, the wise man par excellence.
The confusion with the hyena, which in Mande leads
to namakoro "the old hyena," leads in Mexico to
Ueuecoyotl "the old coyote" and Coyotlinaualli "the
coyote wizard," where we once more have the Arabic
naM. (III. 241-255, 296.)
41. The Mandingo "itch" or "syphilis," nana, and
Bambara manyan "a fetish whose pyramidal altar
may be seen near the villages" leads in Mexico to the
god of syphihs, Nanauatzin, to whom a pyramidal
altar, called tzaqualli, was erected, and this tzaqualli
is identical with Hausa dakali "a pagan altar." (III.
,280-286.)
42. The Mexican picture-writing resembles the rock
inscriptions in the Sudan, which are unquestionably
of âdwal astrological origin. The importance of the
spider in connection with such ga^wals in Africa, which
is of universal significance in the Sudan, appears in a
large number of Mound-builder gorgets in America
with tjie gadwal cross in the middle. (III. 263-267,
273 f.)
43. The Sufi element of the ecstatic dance and self-

castigation are important elements of many American
reUgious ceremonials, and the name for the dance and
CONCLUSIONS 369
the refrains of the songs accompanying it are identical

with, those in Africa. (III. 199-212.)
44. The external representation of the royal dignity
by means of a stool, generally of one piece of wood,
called in the Sudan by its Arabic name taht, tuhut, was
found among the Caraibs with the identical name,
duho. The conical hat, representing royalty or high
dignity in Mexico, among the Huastecans, and else-
where, and in Aztec known as copilli, is linguistically
identical with the "hat" words in Africa derived from
Arabic gifdrah, gufarah. (III. 218 f., 321 f.)
45. The Mande totemic n'tene, originally a sign of
distinction, is preserved in Caraib nitaino "man of
distinction," and the Sudanic identification of the griot
with the musician has led in Mexico to an identity of
name for "slave" and "musician." The Arabic des-
ignations of wearing-apparel are preserved in Mexico
for the identical garments as in the Sudan. The
Simlah appears here as chimalli and tilmatli, the mai'zar
as maxtli, the gifarah as copilli, the Idbis as tlauiztli,
the banddrtah as pantli, the kisd as quachtli, and the
loâm is found among the Caraibs as nagua. (III.
220 f., 230-239, 318-322.)
\^46. The identity of the spiritual civilizations, down
to minutest details in the Sudan and in Mexico and
elsewhere in America, leads to the assumption that
other cultural elements, identical in both continents
and frequently bearing the same names, are of African
origin. This is preeminently the case with cotton,
which in Africa has a religious pviriflcation significance,
and the presence of which in America before Columbus
outside of its reUgious use in connection with burials
cannot be proved from documentary evidence. (II.
1-82.)
^
47. great resemblance of agricultural methods
The
in America to those in Africa leads to the conclusion
that tobacco and the bread roots of America, some of
which are conceded by the early writers to be of African
origin, owe their origin to the advanced Arabic agricul-
ture, which may be traced in the Sudan, to judge from
philological considerations. (I. 102-268, II. 83-200.)
48. A thorough investigation of the archaeological
remains in the Western Sudan, coupled with a further
painstaking philological study of the Arabic influences
in Africa, may reveal other African elements that are
the prototypes of similar conditions in the civilization
of America.
WORD INDEX
dbSin, Gypsy, 10. ahuiimu, Asante, 174.
abonsam, Asante, 179. ahunum, Asante, 174.
abosom, Asante, 175, 179. afj,uviwat, Arabic, 263.
abosonsom, Asante, 179. aidênnye, Mandingo, 172.
abrabo, Asante, 172. 'all, Arabic, 304.
abuhumam, Soa, 230. akka, Kannada, 40.
accearium, LLatin, 15. akom, Asante, 166.
acceris, LLatin, 15. akrade, Asante, 164.
accersa, LLatin, 15. akrafo, Asante, 165.
acer, LLatin, 15. akra-kwd, Asante, 165.
acerlis, LLatin, 15. al, Manenguba, 134.
acero, Spanish, 15. alaghom naom, Tzental, 247.
acerra, LLatin, 15. alale, Wolof, 230.
acersu, LLatin, 15. 'aldmu, Arabic, 305.
aeier, French, 15. alanaua, Arabic, 278.
acieris, LLatin, 15. alapra, Timne, 320.
oftn, Chamir, 10. alaênna, Vei, 173.
adae, Asante, 169. al-'ayah, Arabic, 128.
'adana, Arabic, 47. albios oculos, LLatin, 20.
iSapa, Greek, 331. alcin, Turkic, 28.
adden, Berber, 208. al-ddr, Arabic, 318.
iSvpa, Greek, 331. 'alek, Arabic, 348.
aiwirah, Arabic, 317. aldênna, Kuka, 173.
'ddzan, Arabic, 207. aldênne, Timbuktu, 173.
adz6, Yoruba, 175. ale, Cameroon-Cross River, 134.
adSwa, Asante, 219. aledânna, Pika, 173.
adzwo, Asante, 208. Skevpov, Greek, 53.
aerugo, Latin, 46. iXiday, Greek, 53.
aes, Latin, 4. alezonda, Boko, 173.
aeUtaelg, ASaxon, 39. alfinde, Spanish, 16.
afrodifi, Syriac, 21. al-gdnnah, Arabic, 172.
afsan, Ossetic, 10. alguexi, OPortuguese, 35.
aga, Yoruba, 219. alhiTide, Spanish, 16.
aênna, Tuareg, 223. ali, Bakairi, 212.
aghau, Egyptian, 10. alidena, LLatin, 11, 16.
agifu, Egyptian, 10. al-'iksir, Arabic, 39.
agua, Asante, 219. alinde, Spanish, 16.
aguade, Asante, 187. aljama, Hausa, 278.
aguadi, Asante, 187. aljima, Hausa, 278.
aguadini, Asante, 187. alkaria, Hausa, 318.
'aûz, Arabic, 325. alia, Bambara, 305.
dhen, Persian, 10. 'Allah, Arabic, 305.
dhingar, Persian, 78. allama, Bambara, 305.
ahoma, Fante, 178. 'allamu, Arabic, 305.
ahonhom, Asante, 174. almaizar, Spanish, 232.
ahum, Asante, 174. Hausa, 113.
almajiri,
ahuii, Asante, 174. almalâmah, Arabic, 49.
WORD INDEX 373
alman, Songay, 230. dr, Kannada, 4.

almara, Hausa, 278. ara, LLatin, 15.
almardkam, Arabic, 50. arddênna, Mandingo, 172.
almit4, Zenaga, 112. ar'ba, Zenaga, 234.
al^qamarani, Arabic, 280. arcati, Sanskrit, 40.
alqamaru, Arabic, 280. ardsenne, Bambara, 173.
alquicel, Spanish, 239. ardyine, Bambara, 173.
alquicer, Spanish, 289. arefi, Timne, 174.
alti, Huasteca, 317. areyto, Taino, 199.
aliepetl, Nahuatl, 317. ari,Bakairi, 212.
altun, Turkic, 28. arianna, Timne, 172.
aiutacion, Spanish, 33, 35. arid, OHGerman, 38.
alutatio, Latin, 34. aridsanna, Soso, 173.
aluiatium, Latin, 34. arld§enne, Mandingo, 172.
alutiae, Latin, 35. aritin, Arawak, 211.
aluz, OPortuguese, 35. ari utcha, Egyptian, 132.
alveice, OPortuguese, 35. ariyanna, Soso, 173.
alvexi, OPortuguese, 35. arize, OHGerman, 38.
dma, ONorse, 26. arizin, OHGerman, 38.
amalgama, LLatin, 49. arizzae, OHGerman, 38.
'aman, Arabic, 160. arizze, OHGerman, 37.
aman san, Asante, 218. arka, Kannada, 40.
amana, Hausa, 160. arkas, Sanskrit, 40.
amanteca, Nahuatl, 160. arrogium, LLatin, 36.
amantecatl, Nahuatl, 241. arroser, French, 36.
amaoti, Caraib, 241. arroyar, Spanish, 36.
amaotta, Aymara, 241. arroyo, Spanish, 36.
amauta, Kechua, 241. arrozar, Provencal, 36.
amdebber, Berber, 240. arrugia, LLatin, 36.
amehbul, Tamazirt, 154. dru, Kannada, 4.
amina, Hausa, 161. arutos, OHGerman, 38.
'aminah, Arabic, 162. arutz, OHGerman, 38.
'dmir, Arabic, 162. aruz, OHGerman, 27, 38.
Ammo, African, 297. aruzi, OHGerman, 38.
dmr, ONorse, 26. aruzz, OHGerman, 38.
an, Manenguba, 134. arvt, OHGerman, 38.
avr-bar, Sumerian, 6. 'arwat, Arabic, 38.
andan, Votyak, 16. a'idb, Arabic, 133.
andanicum, LLatin, 16. asamaii, Asante, 176.
ande, Wolof, 188. a§aman, Guanche, 223.
andena, LLatin, 11, 16. aSano, Guanche, 223.
andun, Ossetic, 16. a'Sdr, Arabic, 280.
anga, Subiya, 132. 'asara, Arabic, 12.
anicula, LLatin, 325. 'aiara, Arabic, 348.
'dnft, Arabic, 12, 16. asem, Egyptian, 40.
anmii4, Zenaga, 112. dsin, Pehlevi, 10.
annabi, Soso, 250. dslngdr, Pehlevi, 78.
annicula, LLatin, 325. 'ailr, Arabic, 279.
anula, LLatin, 325. dsir, Hebrew, 14.
aoualetago, Caraib, 211. 'asr, Arabic, 12.
apakyi, Asante, 193. 'aSHra, Arabic, 348.
dippodlTT], Greek, 21. aiuwQfi, Yoruba, 177.
d^poirAijcoi, Greek, 50. atejmud, Zenaga, 112.
apsin, Gypsy, 10. iBlyyami, Greek, 67.
'air, Arabic, 11 f. bel, Coptic, 65.
aveugle, French, 20. bele, Fernandian, 134.
awsin, Kurd, 10. bSllm,Hebrew, 68.
iilpri, Greek, 10.
beorht,ASaxon, 23.
ay a, Kannada, 4.
beornan, ASaxon, 26.
ayas, Sanskrit, 4.
here, Vei, 233.
ayi, Asante, 209.
beredi, Persian, 21.
ayo, Avestan, 4.
beremo, Vei, 234.
ayya, Kannada, 3. OHGerman, 22.
berhta,
'aza'r, Arabic, 78.
bert, Ethiopic, 21.
'azma' , Arabic, 170.
besi, Malay, 7.
besnat, Coptic, 10.
bhrae, Sanskrit, 23.
haaêrirpet, Egyptian, 7. bibrantia, LLatln, 24.
babara, Sumerian, 17. bieri. Fan, 158.
bairhtjan, Gothic, 23. bieti, Fan, 158.
bairhts, Gothic, 23. bila, Malinke, 172, 233.
bakal, Tagalog, 8. bildnd, Hindustani, 65.
balamo, Gypsy, 68. biland, Gypsy, 65.
balamu, Gypsy, 68. bila'Aov, Gypsy, 65.
balant, Arabic, 19. bilaiii, Syriac, 21.
balaqu, Assyrian, 34. biri, Bambara, 235.
balaur, Arabic, 19. birindi, Persian, 21.
balim, Syriac, 68. birr, Arabic, 172.
ballvbca, Latin, 34. biii, Malinke, 235.
baluce, Latin, 34. bitinkS,, Malinke, 235.
baluqah, Arabic, 34. bitia, Mandingo, 235.
balur, Arabic, 19. bjartr, ONorse, 23.
band, Persian, 238. bla, Bambara, 172, 233.
banddriyah, Arabic, 238. blam, Syriac, 68.
pdvdoy, Greek, 238. blind, German, 21.
bankarey, Songay, 238. blm, Hebrew, 68.
bante, Hausa, 238. bins, Arabic, 18.
bar, Sumerian, 6. blm, Arabic, 18.
bara', Arabic, 172. boa, Mandingo, 159.
bara, Songay, 230. boatio, Mandingo, 159.
b&ra, Asante, 171. bo (&en, Akra, 174.
baraka, Berber, 332. bohique, Taino, 226.
barand, Persian, 25. bohuti, Taino, 223.
barand, Arabic, 25. bol, Coptic, 65.
baraq, Hebrew, 23. boli, Bambara, 142.
baraqa, Arabic, 23. boratio, Venezuela, 223.
barika, Bambara, 332. bori, Hausa, 154.
barka, Bambara, 332. borom alale, Wolof, 230.
barot, Coptic, 21. borome, Wolof, 230.
bdrqd, Syriac, 24. boviti, Taino, 223.
barqtd, Talmudic, 24. boye, Caraib, 226.
barr, Arabic, 172. bra, Asante, 171.
barzSl, Hebrew, 7. bractea, Latin, 23.
basSi, Cam, 7. brand, ASaxon, 25.
basnet, Coptic, 10. brandeum, LLatln, 24.
batel,Spanish, 348. brandir, OFrench, 25.
batey, Taino, 348. brandisium, LLatln, 22.
behique, Cuba, 226. brando, LLatln, 24.
WORD INDEX 375
brantia, LLatin, 24. chag, Kiche, 308.
bramia, LLatin, 24. chak, Pokonchi, 308.
braq, Syriac, 23.
chakij, Pokonchi, 308.
brattanea, LLatin, 24.
chaklaj, Pokonchi, 308.
brattea, Latin, 23.
brattealis, LLatin, 24. xoXk^s, Greek, 4.
bratteatus, LLatin, 24. Xap&aaav, Greek, 4.
bratthea, LLatin, 24. xap/ctdj, MGreek, 73.
brenna, ONorse, 26. chartolaratum, LLatin, 71.
bri, Bambara, 235. xopri^s, MGreek, 73.
brindisium, LLatin, 22. chaudronnier, French, 76.
brinnan, Gothic, 26. chaugan, Persian, 340.
brinno, Gothic, 26. chaul, Persian, 340.
brand om, ASaxon, 26. chaule, OFrench, 343.
PpovTâ-tov, Greek, 22. chedas, Sanskrit, 8.
xeX'Soy/o, Greek, 41.
bronze, EngUsh, 22.
bu-el-mal, Berber, 200. chelidonium, LLatin, 42.
buhuitihu, Taino, 223. chemijn, Caraib, 222.
buri, Tuareg, 154. cheole, OFrench, 343.
chi, Siyin, 7.
buti, South Nyasaland, 134.
bwStS, Bateke, 159. chicane, French, 343.
bmti, Piot, 158. chicha, Taino, 331.
byrnan, ASaxon, 26. chidram, Sanskrit, 8.
chim, Maya, 231.
chimal, Maya, 231.
chimalli, Nahuatl, 231.
ca, Tibetan, 7. chimil, Maya, 231.
caan, Maya, 308. chindtti, Sanskrit, 8.
caanal, Maya, 808. xma, Egyptian, 53.
cabalone, Venetian, 59. choca, Portuguese, 343.
caldaia,^ Italian, 76. choquer, French, 343.
calderaio, Italian, 76. choula, OFrench, 343.
calibum, LLatin, 13. choule, OFrench, 343.
canal, Maya, 308. Xpva-is, Greek, 27.
Candida, LLatin, 36. chueca, OSpanish, 343.
caniba, Taino, 185. chUl, Persian, 340.
canima, Taino, 185. d, Kongo, 134.
eapilla, Spanish, 321. iil, Turkic, 28.
cappa, LLatin, 319. cimi, Taino, 222.
caragius, LLatin, 191. cimini, Taino, 222.
carai, BraziUan, 185. iini, Kannada, 8.
caraiba, Brazilian, 185. cinna, Kannada, 8.
cards, Norman French, 191. ciii, Kannada, 8.
caraut, OPrench, 191. cloi, Coptic, 349.
Cariba, Taino, 185. coa, Caraib, 313.
cary'ba, Brazilian, 185. coaca, Nahuatl, 313.
eastrdm, Avestan, 9. cobali, LLatin, 57.
cazzuola, Italian, 76. cohofde, Spanish, 59.
cemi, Taino, 222. codardo, Italian, 59.
eemla, Berber, 231. con, Nahuatl, 313.
cervolo,LLatin, 325. copilli, Nahuatl, 321.
chaac, Maya, 308. corrugium, LLatin, 36.
choc, Maya, 308. couard, French, 59.
chacah, Maya, 308. couatl, Nahuatl, 308.
chacal, Maya, 308. covali, LLatin, 57.
Coyotlinaualli, Nahuatl, 250. derbe, Songay, 232.
cuachim, Huasteca, 231. deuk, Wolof, 219.
cuachimal, Huasteca, 231. di, Kilimanjaro, 133.
cuachimzal, Huasteca, 231. dia, Bambara, 301.
cueitl, Nahuatl, 234.
diar, Hassania Arabic, 317.
cuprum, Latin, 7.
djam, Songay, 113.
cuyule, Maya, 234.
djiri, Malinke, 135.
dka, Aramaic, 17.
dkar, Berber, 12.
da, Bambara, 241. dla, Bambara, 241.
daba, Bambara, 241. dloki, Bambara, 232.
ddba, Baga, 313. do, Malinke, 208.
dabak, Bambara, 241. doci, Tanganyika-Bangweulu, 131.
dabara, Hausa, 240. doki, Kongo, 132.
dabari, Songay, 240. dokili, Malinke, 208.
dabbar, Arabic, 240. doloki, Bambara, 232.
dabiri, Hausa, 236. dolokie, Peul, 232.
dabr, Arabic, 240. domma, Soso, 232.
dabu, Hausa, 236. don, Bambara, 208.
dadaga, Sumerian, 17. don, Songay, 273.
dag, Sumerian, 17. dondiko, Mandingo, 232.
da§ala, Arabic, 107. doroke, Soninke, 232.
dagara, Arabic, 150. dowe, Rumpe, 132.
dajgal, Arabic, 107, 214. dozi, North-west Nyasa, 131.
daggala, Arabic, 214. dâlo, Adirar, 313.
dd'irah, Arabic, 317. dsarai, Wadai, 313.
dakali, Hausa, 286. dsawa, Gura, 317.
dakara, Hausa, 150. dsenna, Bornu, 173.
dakl, Arabic, 286. dhi, Dahome, 175.
dala, Bambara, 241. d§imo, Goali, 136.
dalia, Hausa, 209. dînnawes, Kanika, 173.
ddmlna, Hausa, 171. dsinnukuzu, Hwida, 173.
dana, Bambara, 192. dSlS, Arabic, 331.
daquir, Arabic, 12. dso, Aku, 208.
dar, Arabic, 317. du, Mandingo, 219.
da'r, Arabic, 78. du'a', Arabic, 315.
dare, Brafeda, 317. duaria, Timbuktu, 317.
d-areii, Timne, 173. dufud, Asante, 219.
daria, Hausa, 209. duga, Bambara, 315.
dauri, Hausa, 318. dugare, Bambara, 302.
dasara, Arabic, 150. dugu, Bambara, 317.
dasiri, Bambara, 150. dugutigi, Bambara, 146.
dawa', Arabic, 132. duho, Taino, 219.
dawa'ir, Arabic, 317. duma, Baga, 232.
dawdr, Arabic, 317. duna do arjana, Malinke, 279.
dawwdr, Arabic, 317. dungare, Bambara, 302.
debber, Berber, 240. dungari, Malinke, 302.
debe, Soninke, 318. dux), Mandingo, 219.
defS,, Wolof, 241. duo, Adampe, 208.
defard, Wolof, 240. duruki, Mandingo, 232.
deger, Berber, 150. duwa', Arabic, 315.
dejjal, Berber, 107. dya, Bambara, 171, 301.
delegoa, Gura, 232. dyalli, Peul, 209.
demir, Turkish, 5. dyama, Bambara, 170.
WORD INDEX 377
dyamandyan, Bambara, 170. el, Sumerian, 41.
dyamani, Bambara, 170. AotAs, Greek, 42.
dyamaso, Bambara, 171. ele, Ruwanzori, 134.
dyamburu, Wolof, 113. electre, ASaxon, 42.
dyame, Wolof, 113. elehtre, ASaxon, 42.
dyamome, Wolof, 171. ijXeKTpos,Greek, 40.
dyamu, Bambara, 171. ^'KiKTwp,Greek, 40.
dyamu, Wolof, 171. eleiarum, LLatin, 42.
dyamuru, Bambara, 113. ellu, Assyrian, 41.
dyan, Bambara, 170. elidrium, LLatin, 41.
dyan, Wolof, 312. elothr, ASaxon, 42.
dyeli, Malinke, 107. elotr, ASaxon, 42.
Dyeli-ba, Mandingo, 108. Greek, 41.
iXiSptov,
dyelli, Peul, 209. emba. North-west Zambezia, 134.
dyelo, Bambara, 209. enmud, Zenaga, 112.
dyere, Songay, 279. enti, Bedauye, 16.
dyo, Songay, 113. epinatzi, Yabitero, 226.
dzakar, Arabic, 12. er, Sanskrit, 3.
die, Akra, 174. er, Kannada, 2.
dieii, Akra, 174. Sraka, Kannada, 2.
dzefiba, Akra, 174. ere. Central Ogowe, 134.
dzi, Ewe, 173. iri, Kannada, 2.
dn4udu, Ewe, 173. eremeri, Caralb, 209.
dziifose, Ewe, 173. eremi, Galibi, 209.
dzimeanyisi, Ewe, 173. eruvve, Tamil, 3.
dzihgo, Ewe, 173. eru, Assyrian, 1.
dzinko, Ewe, 173. erugo, Latin, 46.
dzifikusi, Ewe, 173. IpvBpbs, Greek, 5.
dziw'adola, Ewe, 174. esejret, Redames, 135.
dziw'o, Ewe, 173. esi, Mabi, 175.
dSo, Akra, 208. esinole, Adampe, 175.
dzu'dr, Arabic, 78. esmali, Provencal, 48.
dzwa, Asante, 219. esmart, Provengal, 48.
diwento, Asante, 208. esmaudus, LLatin, 48.
diwonto, Asante, 208. esmaut, Provengal, 48.
dzwumM, Asante, 113. estandart, OFrench, 137.
esum, Asante, 175.
ete, North Central Congoland, 134.
else,Spanish Guinea, 134.
S, Upper
Ituri, 134.
ewyi, Asante, 174.
earn, ONorse, 14.
ewyiase, Asante, 174.
ebara, Filham, 313.
ewyim, Asante, 174.
ebon, Bayon, 230.
exmaltitus, LLatin, 42.
eci, Spanish Guinea, 134.
ey, Kannada, 3.
ed', Manenguba, 134.
edejmun, Zenaga, 112.
edmnikiLsu, Mabi, 173.
edfinuali, Anfue, 175. facdo, LLatin, 129.
edyi, Mabi, 175. faciicius, Latin, 129.
edzi, Yoruba, 175. factio, LLatin, 129.
edzwom, Asante, 208. faitis, OFrench, 129.
efila, Aku, 233, 321. falala, Hausa, 230.
egua, Asante, 187, 219. fale, Bambara, 230.
eisarn, Gothic, 13. falin, MaUnke, 230.
el', Manenguba, 134. faling, Malinke, 230.
fana, Malinke, 230. gaharde, Albanian, 59.
fanu, Malinke, 238. gabb, ONorse, 58.
farzala, Arabic, 7. gabban, ASaxon, 58.
fatal, Sotho, 350.
gabbo, Italian, 58.
fa\ilah, Arabic, 350.
gabe, Albanian, 59.
fatn, Arabic, 350.
gabel, Albanian, 59.
fatsah, Arabic, 129.
gabil, Fulup, 233.
fau(, Arabic, 109.
gabim, Albanian, 59.
feitieo, Portuguese, 128.
gabn, Arabic, 58.
felis,OHGerman, 19. gaboj, Albanian, 59.
fendeli, Soninke, 238.
gadi, Bornu, 312.
ferrum, Latin, 7.
ga4wal, Arabic, 268.
fi-al-mdl, Arabic, 230.
gafir, Arabic, 233.
fila, Hwida, 321.
gafr, Arabic, 233.
fila, Yoruba, 233.
3030, Soninke, 316.
filazz, Arabic, 18.
gagar, Arabic, 78.
file, Gbe, 321.
gal, Sumerian, 3, 7.
filizz, Arabic, 18.
galdbd, Syriac, 8.
filizzafu, Arabic, 18.
galbu, Assyrian, 8.
fini, Malinke, 238.
gale, French, 285.
firi, Malinke, 230.
âli, Arabic, 107.
firila, Malinke, 230.
gallabu, Assyrian, 8.
firind, Arabic, 25.
galpoli, Guresa, 233.
fitila, Songay, 350.
gam', Arabic, 170.
fitina, Songay, 350.
gamd'ah, Arabic, 170,278.
fitne, Bambara, 350.
gamalteinais, Gothic, 51.
fjall, ONorse, 19.
gamara, Arabic, 167.
flins, OHGerman, 20. garni, Berber, 168.
flinta, OHGerman, 20.
âmi', Arabic, 170.
flnt, Arabic, 20.
gamm, Arabic, 167.
flnz, Arabic, 18.
gammah, Arabic, 167.
foalerah, Gurma, 233, 321.
gammi, Berber, 168.
fogome, Soninke, 230.
gamrah, Arabic, 167.
fola, Dsarawa, 321.
^dn, Arabic, 312.
fole, Malinke, 348.
gan, Arabic, 172.
folio, Mandingo, 348.
gana, Zapotec, 247.
fonta, OSpanish, 58..
gandadia, LLatin, 36.
foro, Malinke, 99.
gandeda, LLatin, 36.
fomla, Mossi, 233, 321.
ganga, Spanish, 36.
frodltl, Syriac, 21.
ganga, Nyika, 132.
fuale, Bambara, 230.
gangadia, LLatin, 36.
Fudh, Peul, 109.
gangue, French, 36.
fula, Uupe, 233.
Hannah, Arabic, 172.
fvla, Mandingo, 233.
gar, Sumerian, 3.
fule, Dewoi, 321.
gardSu, Assyrian, 8.
fuluzz, Arabic, 18.
gari, Hausa, 318.
fura, Bambara, 150.
gariatun, Hassania Arabic, 318.
Fut, Peul, 109.
garku, Kannada, 4.
garpal, Bagbalan, 233.
âs, Sumerian, 10.
gab, OFrench, 58. âSiS, Arabic, 331.
gabn, Malinke, 320. gaiana, Hausa, 278.
Sabdn, Arabic, 58. gau, Olrish, 59.
gabar, Portuguese, 58. §aukdn, Arabic, 341.
WORD INDEX 379
gawadjon, Gothic, 187. guSkin, Sumerian, 27.

gaz, Sumerian, 10. 'gUz, Berber, 327.
gbara, Gbese, 320.
gbaralei, Gbande, 320.
gbawere, Mende, 320. hob, Arabic, 58.
gbilen, Gio, 321. ifabal, Hebrew, 57.
gbira, Gio, 320. habal, Arabic, 60.
gbola, Mano, 320. âbdld, Syriac, 57.
gboro, Legba, 320. habalu, Assyrian, 57.
ge}^, Egyptian, 10. Jiabbdld, Hebrew, 57.
gehgeJj,,Egyptian, 10. habbilu, Assyrian, 57.
gSlabd, Talmudic, 8. habel, Berber, 154.
gekiis, Lithuanian, 5. habl, Arabic, 57.
gesmipon, Gothic, 52. âbul, Arabic, 57.
ghnaoghel, Tzental, 247. hagagu, Assyrian, 9.
gibur, Boritsu, 320. haganu, Assyrian, 9.
gi4du, Kannada, 8. ^gind, Syriac, 10.
gifarah, Arabic, 233, 320. kagind, Talmudic, 10.
gimdr, Arabic, 168. haginnu, Assyrian, 9.
gin, Sumerian, 16. hdd, Arabic, 10.
gingi, Sumerian, 316. hiadid, Arabic, 10.
giri, Gio, 135. hddânna, Wolof, 173.
girigiri, Mandingo, 130. haegtes, ASaxon, 330.
gismelze, OHGerman, 42, 51. haegtis, ASaxon, 330.
glaber, Latin, 8. haehtis, ASaxon, 330.
y\a</>vp6s, Greek, 8. haftâuS, Persian, 18.
y\iwTeiv, Greek, 8. haguza, Berber, 326.
goba, Olrish, 59. haidUza, Berber, 326.
gob'd, Talmudic, 319. hd'in, Arabic, 58.
gobelin, OFrench, 59. hair, Arabic, 196.
gof, Welsh, 59. halba?, Arabic, 84.
gofdfer, Adirar, 233. halib, Arabic, 84.
gofdra, Beran, 233. hdli?, Arabic, 35.
gona, Zapotec, 247. halldbah, Arabic, 84.
gora, Hausa, 193. halub, Arabic, 84.
gow, Cornish, 59. halU?, Arabic, 35.
goy, Persian, 341. Kama, Mende, 135.
gregory, English, 130. 'hampi, Kechua, 226.
grLgri, French, 130. hanga, Mochi, 132.
griot, French, 106. J),aragu, Assyrian, 27.
gser, Tibetan, 28. hdradênna, Biafada, 172.
gua, Asante, 219. JlMraku, Assyrian, 27.
guatiao, Taino, 186. iârigu, Assyrian, 27.
guatigi, Bambara, 151. harlgah, Arabic, 36.
guelu, Bambara, 316. haris, Sanskrit, 28.
guflrah, Arabic, 283. harq, Arabic, 36.
gugute, Soninke, 316. harru, Assyrian, 27.
guiriot, French, 106. karU, Assyrian, 27.
gum, Berber, 168. i}driig, Hebrew, 27.
gumma', Arabic, 167. ha§db, Beran, 133.
gummnh, Arabic, 167. ]}a$in, Arabic, 10.
gumr, Arabic, 167. }pa?in, Ethiopic, 10.
gurumbd, Bornu, 320. haklS, Arabic, 331.
gurumpa, Karekare, 320. Jpa§n, Arabic, 9.
hdiim, Arabic, 128. honnir, French, 58.
haiin, Ethiopic, 10. honte, French, 58.
hatumi, Hausa, 128. hor, Turkish, 59.
haulube, Peul, 107, 114. Ho-re, Mandingo, 155.

horiat, Turkish, 59.
haun, Arabic, 58.
horo, Malinke, 99.
hauni, Berber, 263.
hrabi, Malinke, 114.
hauniyah, Arabic, 263.
huan, Berber, 263.
hauns, Gothic, 58.
hiMToge, Cumanagoto, 212.
haur, Arabic, 36.
hub, Arabic, 58.
fiaure, Caraib, 193.
hubdld, Syriac, 57.
hawwdr, Arabic, 59.
hubbal, Arabic, 57.
hawwur, Arabic, 59.
hufune, Peul, 233, 320.
hazes, OHGerman, 330. humar, Arabic, 26.
hazind, Talmudic, 10.
humrah, Arabic, 26, 278.
hdzis, OHGerman, 330.
huroQU, Assyrian, 27.
ftazMS, OHGerman, 330.
hurru, Assyrian, 27.
Man, Gothic, 58.
hurruq, Arabic, 36.
heb, Egyptian, 58.
huwwal, Arabic, 57.
êftoM, Egyptian, 58.
hebbata, Berber, 154.
huwwarai, Arabic, 59.
huwwdrun, Arabic, 59.
bebbel, Berber, 154.
jiu'wUr, Arabic, 36.
hebbuel, Berber, 154.
hwang, Chinese, 28.
Jêben, Egyptian, 58.
hwar, Arabic, 59.
êbul, Ethiopic, 57.
hyira, Asante, 193.
hechizo, Spanish, 129.
hegg, OEngUsh, 330.
hegitisse, ASaxon, 330.
heka, Egyptian, 132. i.Upper Ituri, 134.
hielebi, Arabic, 79. iarn, Olrish, 14.
hemba, Angola, 134. 'ibldm, Arabic, 68.
her, Berber, 193. icheiri, Caraib, 222.
J!^ra, Malinke, 193. 'Id, Arabic, 170.
hera, Bambara, 193. igenni, Berber, 172.
hesin, Kurd, 10. iîimah, Arabic, 278.
}fetem, Egyptian, 20. ijenni, Berber, 172.
Hexensabbat, German, 330. il, Kiche, 305.
hi, Kilimanjaro, 133. ilah, Maya, 306.
hiarn, Olrish, 14. ilama, Nahuatl, 305.
Mb, Arabic, 58. Ilamatecutli, Nahuatl, 305.
hik, Coptic, 132. ilemi, Galibi, 209.
hilliba, Arabic, 84. ilhuia, Nahuatl, 306.
hind, Arabic, 12. ilhuicatl, Nahuatl, 306.
hindl, Arabic, 12. ilhuice, Nahuatl, 306.
Mr, Berber, 193. ilhuiz, Nahuatl, 306.
iiirarpyam, Sanskrit, 28. ilil, Maya, 306.
hiiritu,Assyrian, 27. ilnamigui, Nahuatl, 306.
hira, Assyrian, 27. Ilia, Kechua, 303.
hldhla, Zulu-Kafir, 134. illa-cabra, Kechua, 303.
hlohla, Zulu-Kafir, 134. 'Illah, Arabic, 305.
hobe, Coptic, 58. illaj, Kechua, 306.
iJwZe, Mende, 348. illapa, Kechua, 306.
homan, ASaxon, 26. illay, Kechua, 306.
honhom, Asante, 174. ilvuic, Pokonchi, 306.
Miien, OHGerman, 58. inumu, Telugu, 1.
WORD INDEX 381
tos, Greek, 46. kabe, Soso, 166.

ir, Tamil, 2. kac, Sanskrit, 17.
Ir, Sanskrit, 3. kdcas, Sanskrit, 17.
ira, Kannada, 2.
kadaku, Kannada, 4.
irame, Soninke, 232.
kadi, Bornu, 312.
ir&, Yoruba, 194.
iri, Yoruba, 194.
kadi, Kannada, 3.
KaSfila, Greek, 20.
Iren, ASaxon, 14.
kaû, Kannada, 3.
iri, Muin, 135.
ka4upu, Kannada, 3.
iri, Kannada, 2.
kUgarU, Gimini, 135.
irimba, Malayalam, 1.
kakku, Kannada, 4.
Iro, Kabungo, 135.
kal, Kannada, 4.
iron, English, 14.
kal, Sumerian, 3.
'irsa', Arabic, 208.
kalan, Bambara, 163.
irumbu, Tamil, 1.
kalanfa, Bambara, 191.
irya, Sanskrit, 2.
kalfa, Bambara, 166.
isarn, OHGerman, 14.
kali, Mende, 312.
isar-un, Arabic, 14.
kalpak, Tatar, 319.
Isen, ASaxon, 14.
kalte, German, 76.
Isern, ASaxon, 14.
kalU, Assyrian, 8.
iso, Aruwimi-Lomami, 134.
kalugu, Bornu, 232.
issimd, Galibi, 222.
kaluru, Bornu, 232.
'isiiharah, Arabic, 194.
kamme, Soninke, 171.
ite, North Central Congoland, 134.
kamu, Soninke, 171.
'itr, Arabic, 12.
kan, Chinese, 340.
'izdr, Arabic, 232.
kan, Kannada, 16.
kan, Bambara, 312.
kan, Asante, 163.
jal, Turkic, 28. kana, Asante, 163.
jar, Turkic, 28. kana, Bambara, 150, 312.
jdrn, ONorse, 14. k&na, Asante, 163.
je, Cameroons-Cross River, 134. kdiic, Sanskrit, 17.
jelli, Malinke, 209. kanfa, Bambara, 191.
ji, Angola, 134. kani, Kannada, 16.
ji, Dahome, 173. kanja, Tamil, 16.
jijoho, Dahome, 173. kaiikye, Asante, 163.
jijoo, Dahome, 173. kanna4i, Kannada, 16.
jijowd, Dahome, 173. kanta, Malinke, 166.
jik, Northern Cross River Basin, 134. kantiba, Mandingo, 191.
jikpa, Dahome, 173. kajnra, Gurma, 320.
jikpome, Dahome, 173. kar, Sumerian, 3.
jine, Bambara, 142. kara, Bulom, 313.
jingdnih, Syriac, 78. kara, Kannada, 3.
jinkusu, Dahome, 173. k&ra, Asante, 164.
jongo, Mandingo, 113. kara, Malinke, 163.
juvenu, Dahome, 175. kara, Baga, 313.
kara, Sanskrit, 4.
kara, Vei, 163.
kU, Akra, 164. karabe, Bambara, 234.
ka, Bambara, 312. kara4u, Kannada, 4.
kaba, Bambara,166. karafe, Malinke, 234.
kabako, Bambara, 166. karakacas, Sanskrit, 4.
kabarde, Albanian, 59. karaku, Kannada, 4.
kabburi, Toda, 3. karamo, Mandingo, 191.
karamoko, Peul, 163. kid, Sumerian, 8.
karan, Vei, 163. klkin, Malinke, 316.
karang, Mandingo, 163. kil, Turkic, 28.
karas, Sanskrit, 3. kin, Chinese, 27.
karatu, Hausa, 163. king, Chinese, 16.
karba, Tulu, 3. klr, Arabic, 114.
karballatu, Assyrian, 318. kira, Hausa, 114.
karbdltd, Syrian, 318. kirbel, Hebrew, 318.
karbalfd, Talmudic, 319. kiri, Kannada, 8.
karbe, Bambara, 234. kirtne, Cheremissian, 4.
karbun, Kannada, 3. kisa', Arabic, 239.
Kapxd\eos, Greek, 4. k'iu, Chinese, 340.
Kcipxapos, Greek, 4. kla, Akra, 163 f.
kard, Persian, 4. klamo, Akra, 163.
kax^, Kannada, 3. koba, Talmudic, 319.
karfa, Malinke, 166. )c6|3aXos, Greek, 56.
karfe, Bambara, 234. kobe, Albanian, 59.
karkaras, Sanskrit, 4. kobi, Rumanian, 59.
karku, Kannada, 4. kobi, OBulgarian, 59.
karte, Ostyak, 4. kobim, Albanian, 59.
karukku, Tamil, 4. kobolt, German, 59.
karumbon, Tamil, 3. kobon, Albanian, 59.
kassi, Hassania Arabic, 239. kddali, Kannada, 10.
kathirpas, Sanskrit, 10. kofla, Swahili, 321.
kalhora, Sanskrit, 10. ko-gane, Japanese, 27.
kali, Kannada, 4. kolibante, Peul, 107.
kalidra, Sanskrit, 10. kom, Asante, 166.
katumi, Hausa, 128. koma, Malinke, 168.
kaiura, Kannada, 10. komfo, Asante, 167.
)taTf£/3eXos, Greek, 76. 'komo, Akra, 167.
kauye, Hausa, 318. komo, Bambara, 168.
kava, Malinke, 166. KwixoSpbiuK, Greek, 69.
kay, Kannada, 3. 'komoyeli, Akra, 167.
kdy, Kannada, 4. kora, Asante, 193.
kebs, Tibetan, 319. kore, Mandingo, 261.
Kemeno-rom, Coptic, 69. koro, Bambara, 193.
kend, Asante, 164. korofo, Bambara, 193.
kefikan, Asante, 108. koroni, Bambara, 193.
keri, Beran, 313. kort, Votyak, 4.
keri, Soso, 312. kort, Zyrenian, 4.
ketsala, Timne, 313. kote, Mandingo, 261.
key, Kannada, 3. kovali, OBulgarian, 59.
khabari, Soso, 233. kovalinya, OBulgarian, 59.
khabi, Soso, 114. kovarinU, OBulgarian, 59.
kharamokho, Soso, 191. kovU, OBulgarian, 59.
kharan, Soso, 163. koyra, Timbuktu, 318.
kharas, Sanskrit, 4. kra, Asante, 164.
kheb, Egyptian, 58. krd, Asante, 163.
khebent, Egyptian, 58. Krdmo, Asante, 163.
khebenti, Egyptian, 58. Krdmoni, Asante, 163.
khebt, Egyptian, 58. kran, Bambara, 163.
kkergej, Wolof, 316. kuare, Mandingo, 261.
khi, Siyin, 7. kvh'dh, Talmudic, 319.
khui, Soso, 164. kubbu, Assyrian, 319.
WORD INDEX 383
kubSu, Assyrian, 319. lemterget, Berber, 192.
kuera-mikbi, Soso, 157. letanum, LLatin, 42.
kufune, Soninke, 233, 320. letarum, LLatin, 42.
kufura, Soso, 233. leu, Chinese, 5.
kujil, Hassania Arabic, 316. li, Becuana, 134.
kvr-kin, Sumerian, 27. li, Pangwe, 134.
kukulcan, Maya, 316. libide, Malinke, 320.
kukum, Maya, 316. libiri, Bambara, 320.
kukumel, Maya, 316. libri, Malinke, 320.
kulbak, Persian, 319. 16, Kwa-Kasai, 132.
kulbe, Persian, 319. 16, North Central Congoland, 134.
kum, Asante, 167. loga, Koro, 232.
kuma, Malinke, 168. logo, Kwa-Kasai, 132.
kuntigi, Bambara, 158. l6go, Nyanza, 181.
kur, Arabic, 342. I6ha, Ruflji-Ruvuma, 131.
kura, Bambara, 342. lohas, Sanskrit, 1, 5.
kura, Hausa, 162. loi. Southern Rodesia, 131.
kurah, Arabic, 336. lok', Ogowe-Gaboon,132.
Kvppapla, Greek, 319. loko, Nyanza, 131.
kurbe, Persian, 319. Ion, South Congoland, 134.
kuria, Hausa, 193. longo, Ogowe-Gaboon, 132.
kurpaSa, Sanskrit, 319. longu, Central Ogowe, 134.
kurufe, Soninke, 320. lono, Kwa-Kasai, 132.
kufhdra, Sanskrit, 10. I6si, Tanganyika-Bangweulu, 131.
kusi, Coptic, 325. I6ti, Zambezia, 132.
kworia, Hausa, 193. lowi, Zulu-Kafir, 132.
loya, Zulu-Kafir, 132.
loyi, Nyanza, 131.
Idbis, Arabic, 235. M, Coptic, 132.
labisu, Arabic, 235.
I6zi, Wunyamwezi, 131.
labiti, Malinke, 235.
lu, Chinese, 5.
lubus, Arabic, 235.
labo, Malinke, 160.
lufudi, Hausa, 235.
labus, Arabic, 235.
lugod, Dsarawa, 232.
ladsdru, Adirar, 133.
lukukuk, Soninke, 316.
lagam, Arabic, 234.
lumo, Bulom, 232.
lagha, Malinke, 234.
liinta, Malinlce, 160.
laibe, Wolof, 278.
lakhdbe, Wolof, 232.
luioi, Luba-Lunda, 132.
lamaha, Malinke, 234.
laman, Berber, 160.
lamo, Mandingo, 297. mabiti, Malinke, 235.
lamuni, Hausa, 160. mabori, Mende, 160.
Idna, Kandin, 173. maboya, Caraib, 222.
landi, Ruflji-Ruvuma, 133. ma'dana, Arabic, 47.
lapora, Bulom, 320. ma'din, Arabic, 47.
latege, Malinke, 234. mUdii, Akra, 218.
lato,LLatin, 41. maestling, ASaxon, 50.
latonum, LLatin, 42. mage, Soso, 218.
lavdro, Tene, 320. magma, LLatin, 49.
laya, Hausa, 128. mahhul, Arabic, 153.
lebsa, Berber, 235. mahlibdS, Arabic, 84.
leggal, Peul, 135. maiboli, Hausa, 153.
l^dm, Hassania Arabic, 234. maibori, Hausa, 153, 160.
lekki, Peul, 135. maikira, Hausa, 114.
mai'zar, Arabic, 233. maS'altah, Arabic, 79.
makanari, Bakairi, 212. masent, Egyptian, 10.
makeri, Hausa, 114. masiri, Bambara, 233.
maldS, Ethiopic, 44. masirili, Bambara, 233.
malafa, Hausa, 320. madti, Malinke, 233.
malafdre, Peul, 320. maslak, Arabic, 349.
niXay/m, Greek, 49. masubori, Hausa, 160.
HaKay/iaTlliiP, Greek, 49. matala, Arabic, 47.
ma'lam, Arabic, 112. maialaka, Hausa, 192.
niXafM, MGreek, 49. matsokachi, Hausa, 302.
malasaho, Bambara, 209. m,a'tur, Arabic, 12.
malasa, Bambara, 209. muuta, Sicilian, 48.
malata, Arabic, 44. maxtli, Nahuatl, 233.
maldwa, Bornu, 320. mbaraca, Guarani, 193.
maliia, Arabic, 44. m'bieri, Pahuin, 159!
malma, Gothic, 50. m'biri, Abambu, 159.
maltare, Italian, 45. m'biti, Abambu, 159.
maltha, Latin, 44. mbor, Wolof, 230.
pdXfla, Greek, 44. mborahei, Guarani, 211.
malthare, Latin, 45. mborasei, Guarani, 211.
malufa, Pika, 320. mhuan, Duala, 230.
mafi, Asante, 218. m'bweri, A-duma, 159.
mandeopolos, LLatin, 75. m'bwiri, Ba-Kale, 159.
mandindes, LLatin, 75. m'bwiti, Ba-Kale, 159.
mandinga, Spanish, 226. mealmstan, ASaxon, 50.
mandopoli, LLatin, 74. med, Slovenian, 50.
mandopolijn, Dutch, 75. med', OBulgarian, 50.
mandopolini, LLatin, 75. medina, Slovenian, 50.
mandropoli, LLatin, 75. medinar, Slovenian, 50.
mandropolos, LLatin, 75. medor, Slovenian, 50.
mandia, Vei, 218. melango, Mandingo, 285.
/iipms, Greek, 37. melaton, Syriac, 47.
mansa, Mandingo, 218. fi^KSeiv, Greek, 45.
mania', Arabic, 218. melet, Hebrew, 44.
mantaraka, Bambara, 192. m^lm, OHGerman, 50.
manterge, Malinke, 192. melma, Italian, 50.
fj.avTnr6\os, Greek, 73. meltan, ASaxon, 52.
/iivTis, Greek, 73. men, ONorse, 37.
mniitse, Akra, 218. mend, Egyptian, 37.
manyU, Bambara, 285. menango, Mandingo, 285.
manyan, Bambara, 285. menat, Egyptian, 37.
mapoya, Caraib, 222. mene, ASaxon, 37.
maraca, Tupi, 193. menit,_ Egyptian, 37.
maracainbara, Tupi, 193. msnni, OHGerman, 37.
mardfa, Esitako, 320. merango, Mandingo, 285.
marafua, Masu, 320. mesew, Egyptian, 9.
marata, Arabic, 44. mesniu, Egyptian, 9.
marbori, Mende, 160. messing, German, 50.
marda, Sanskrit, 44. mestallium, LLatin, 50.
marham, Arabic, 50. m^stallum, LLatin, 50.
marraka, Arawak, 193. mMail, OFrench, 50.
marsdh, Arabic, 208. wMdl, Talmudic, 47.
masa, Bambara, 218. iih-aXkov, Greek, 47.
masa, Malinke, 218. metalon, Syriac, 47.
WORD INDEX 385
metalon, Talmudic, 46. muma'4an, Arabic, 47.
metdlyah, Talmudic, 46. muntalaka, Hausa, 192.
mStil, Hebrew, 47. murasi, Tupi, 211.
mfoar, Afudu, 320. muii, South Nyasaland, 134.
mfon, Bayon, 230.
mfumu, Mimboma, 230.
migbd'ah, Hebrew, 319. na'am, Arabic, 332.
migfar, Arabic, 320. naat, Maya, 246.
rnildi, Arabic, 44. naatah, Maya, 246.
mildeaw, ASaxon, 46. naatil, Maya, 246.
/utXriu, Greek, 45. naba', Arabic, 249.
itCKros, Greek, 45. ndba, Mossi, 250.
miltou, OHGerman, 46. nabah, Arabic, 249.
mina, LLatin, 36. nabi, Arabic, 249.
mind, Persian, 37. nabina, Wolof, 250.
minci, OWelsh, 37. nablu, Peul, 250.
minerale, LLatin, 37. nabo, Malinke, 160.
minuS, Avestan, 37. naboria, Taino, 221.
mislakah, Arabic, 349. nabure, Soninke, 230.
mitraqah, Arabic, 192. naburugume, Soninke, 230.
mixtum, Latin, 50. nacina, Zapotec, 247.
mi'zar, Arabic, 233. na^wah, Arabic, 38.
mjedenica, Serbian, 50. nafolo, Bambara, 230.
mldt, Syriac, 44. nafuli, Soso, 230.
mldtd, Syriac, 44. nafulu, Malinke, 230.
ndit, Syriac, 44. nagal, Arabic, 285.
mlt, Semitic, 44. nagila, Arabic, 285.
mlton, Syriac, 48. naguas, Taino, 234.
md, Bambara, 297. naguii, Zapotec, 247.
moQaray, Tupi, 212. nahaa, Zapotec, 247.
mdkhi, Bambara, 297. naib, Arabic, 278.
mola, LLatin, 49. na'ima, Arabic, 332.
molde, ASaxon, 48. naima, Hausa, 332.
molt, OHGerman, 38. rmpes, Spanish, 278.
molta, OHGerman, 48. nak'a, Pedi, 132.
molto, OHGerman, 38. naka, Tlapi, 132.
morale, Latin, 37. nam, Berber, 332.
monisto, OBulgarian, 37. ndma, Berber, 332.
fiAvvos, Greek, 37. nama, Malinke, 160, 250.
morose, Tupi, 211. namaha, Malinke, 234.
mortarium, Latin, 44. namakoro, Malinke, 250, 255.
moteca, Nahuatl, 229. namatigi, Malinke, 160.
mpakyiwafo, Asante, 193. namfa, Timne, 333.
mpendel, Wolof, 238. nammadiri, Peul, 332.
mpfuama, Musentandu, 230. nammxiduru, Peul, 332.
mrat, Syriac, 44. nammude, Peul, 332.
mrt, Hebrew, 44. namo, Mandingo, 297.
m'ion, Syriac, 48. namra, Timne, 333.
muince, Olrish, 37. namsarne, Timne, 333.
muinde, Olrisb, 37. namu, Arabic, 297.
mukawwir, Arabic, 342. namucheni, Tarascan, 334.
mvXda, Gothic, 48. nana, Zapotec, 247.
muldeins, Gothic, 48. nana, Malinke, 285.
mulmus, LLatin, 49. nanaol, Maya, 246.
naflUto, Malinke, 285. nima, Hausa, 332.
nanauatl, Nahiiatl, 285. nimah, Kiche, 334.
Nanauatzin, Nahuatl, 280. nimaj, Pokonchi, 334.
nandi, Rufiji-Ruvuma, 133. niman, Kekchi, 334.
nanga, Rongo, 132. nimani, Tarascan, 334.
nao, Kiche, 246. nimaqua, Tarascan, 334.
naoghi, Tzental, 247. nimar, Kiche, 334.
naol, Kiche, 246. nimatehpei, Tarascan, 334.
naom, Tzental, 247. ninini, Tarascan, 334.
naotizan, Kiche, 246. nink, Kekchi, 334.
naSa', Arabic, 218. niSan, Persian, 137.
natal, Kiche, 246. niSanddr, Persian, 137.
natege, Malinke, 234. nitayno, Taino, 221.
natube, Kiche, 246. ilka, Asante, 163.
nana, Nahuatl, 247. nkom, Asante, 166.
naual, Maya, 246. nkommo, Asante, 167.
naualin, Kiche, 246. nkomo, Akra, 167.
naualli, Nahuatl, 246. nkongya, Asante, 167.
nauallotl, Nahuatl, 247. nkonya, Asante, 167.
nauati, Nahuatl, 247. iikra, Asante, 164.
nauatl, Nahuatl, 247. hkra^hea, Asante, 164.
nauatlato, Nahuatl, 247. v^kuzi, Coptic, 325.
nautal, Maya, 246. no. North Central Congoland, 134.
Nawar, Arabic, 79. noeit, Coptic, 53.
nayaa, Zapotec, 247. noh, Maya,
246.
nayanii, Zapotec, 247. noit, Coptic, 53.
n'dana, Bambara, 192. nok'. Central Kongo, 132.
ndarka, Songay, 192. noke, Central Kongo, 132.
ndoma, Mende, 232. nona, Zapotec, 247.
ndomai, Gbandi, 232. nout, Coptic, 53.
nduwe, Dahome, 208. nsunsuan, Asante, 175.
nSIpoSei, Hebrew, 5. n'tale, Bambara, 272.
nei, Cameroons-Cross River, 134. n'tana, Bambara, 192.
nem, Asante, 333. n'iene, Bambara, 139.
nema, Bambara, 333. nterge, Bambara, 279.
neme, Hausa, 332. n'thi, Sho, 7.
nemi, Nahuatl, 334. nti, Kongo, 134.
nemilia, Nahuatl, 334. nuhas, Arabic, 5.
nen, Nahuatl, 334. nu'm, Arabic, 332.
nemitia, Nahuatl, 334. num, Maya, 334.
nemoTiiemi, Nahuatl, 333. nume, Kekchi, 334.
ngaka, Sotho, 132. numu, Bambara, 112.
ngala, Bambara, 305. nunta, Malinke, 160.
nguru, Gbandi, 135. Nuri, Arabic, 79.
nhdSd, Syriac, 5. nwoma, Asante, 178.
hkoma, Asante, 178. nya, Bambara, 142.
ni, Cameroons-Cross River, 134. nyama, Bambara, 157.
niam, Hausa, 333. nyame, Asante, 170.
ni'am, Arabic, 332. nyamena, Soso, 171.
nican, Berber, 139. nyamo, Asante, 170.
nicteca, Nahuatl, 229. nyanga, Zulu, 132.
nidwah, Gothic, 38. nyena, Bambara, 142.
nim, Pokonchi, 334. nyuma, Malinke, 333.
WORD INDEX 387
oalitago, Caraib, 211. . pamitl, Nahuatl, 238.
obil,Fulup, 233. pampaniya, Taino, 238.
obra, Asante, 171. panitl, Nahuatl, 238.
obryzum, Latin, 51. pantli, Nahuatl, 238.
od?t, Dahome, 175. para, Sanskrit, 6.
odzwumfo, Asante, 113. parand, Persian, 25.
ofdra, Adirar, 233. parasei, Tupi, 211.
ogo, Wele-Aruwimi, 132.
parasu, Assyrian, 7.
oguasoni, Asante, 187. pari, Eannada, 6.
oguasonipa, Asante, 188. pari, Sanskrit, 6.
ok&ra, Asante, 164. pari magas, Persian, 25.
okarawa, Asante, 164. parniydn, Persian, 25.
okla, Akra, 165. parSigu, Assyrian, 7.
okomfo, Asante, 167. paru, Tamil, 6.
okra, Asante, 164. parzillu, Assyrian, 6.
olin, Nahuatl, 347.
old,North Central Congoland, 134.
pasSi, Cam, 7.
patachim, Igorot, 8.
Oman, ASaxon, 26.
patatjim, Bontok, 8.
omUfi, Asante, 218.
paii, Nahuatl, 226.
omanha, Asante, 2.
patil, Persian, 350.
omaiisofwe, Asante, 218.
pafil, Hebrew, 350.
om' bin, Galoa, 159.
patli, Nahuatl, 226.
Omeciuatl, Nahuatl, 297.
patolli coachtli, Nahuatl, 239.
omeoto, Bakairi, 241.
pech', Maya, 350.
Ometecutli, Nahuatl, 297.
pechatah, Maya, 350.
omig, ASaxon, 26.
pechech, Maya, 350.
omiht, ASaxon, 26.
peraht, OHGerman, 23.
ondanique, OFrench, 16.
peru, Tamil, 6.
onde, Yoruba, 178.
pet, Maya, 350.
ono, South Congoland, 134.
petel, Kiche, 350.
onta, Catalan, 58.
peielpet, Maya, 350.
onyankopon, Asante, 170.
peten, Maya, 350.
oomI, Coptic, 331.
petes, Egyptian, 129.
ori, Bakairi, 212.
petet, Kekchi, 350.
'opiyfiara, Greek, 27.
petla, Nahuatl, 350.
orufi, Yoruba, 174.
petlani, Nahuatl, 349.
'opi(T(TCii, Greek, 27.
petlania, Nahuatl, 349.
osaman, Asante, 176. petlatl, Nahuatl, 349.
ospanah, Afghan, 10.
petlaua, Nahuatl, 349.
ospina, Afghan, 10.
petr, Egyptian, 350.
osumanni, Asante, 175.
piacha, Chayma, 226.
oSuwofi, Yoruba, 177.
piaye, Galibi, 226.
oiomfo, Asante, 113.
piaze, Bakairi, 226.
owo, Yoruba, 188.
pighldnd, Hindustani, 65.
ozi, Dahome, 175.
pilerU, OHGerman, 19.
pinata, Baniba, 226.
pindai, Caraib, 226.
page, Caraib, 226. pin-lien, Chinese, 25.
pai, Nahuatl, 241. piri, Kannada, 6.
paitia, Nahuatl, 241. piring, Persian, 21.
paje, Tupi, 226. pitejt, Pokonchi, 350.
palaca, Spanish, 33. pitiltu, Assyrian, 350.
palacurna, Latin, 34. jrXdyijs, Greek, 18.
palaga, Latin, 34. plindi, Armenian, 21.

pochteca, Nahuatl, 229. raudi, ONorse, 6.
polman, Tzotzil, 230. ravi, Arabic, 38.
p'olom, Maya, 229. rauta, Finnish, 6.
p'olomkay, Maya, 229. rautah, Arabic, 38.
por, Tamil, 6. rawata, Arabic, 38.
pothao, Bisaya, 8. rd, Yoruba, 194.
pracht, German, 23. ree, Wolof, 209.
pragalaniyam, Sanskrit, 65. rege, Soninke, 207.
irpivBiov, Greek, 24. rem, Coptic, 65.
irpavSunrpiTai, Greek, 24. rere. Central Ogowe, 134.
pichiachi, Tamanaco, 226. rê, Yoruba, 194.
puiacie, Aparais, 226. reri, Aku, 209.
putau, Moro, 8. rerifi, Yoruba, 209.
puto, Bagobo, 8. retla, Hassania Arabic, 272.
ri, Kilimanjaro, 133.
riga, Hausa, 232.
qabllar, Arabic, 321. rimo, Tene, 232.
qaddai, Arabic, 36. rin, Yoruba, 209.
qamardni, Arabic, 280. ris,OHGerman, 38.
qar', Arabic, 198. rnir-n-keme, Coptic, 65.
qara', Arabic, 163, 190. roda, Livonian, 6.
qdrd, Hebrew, 190. r'ofdra, Beran, 233.
qar'ah, Arabic, 193. rogo, Nyanza, 131.
qarditd, Syriac, 190. rohiias, Sanskrit, 5.
qaryah, Arabic, 318. roi,Sengwe-Ronga, 132.
qimar, Arabic, 278. rdji,Nyanza, 131.
qobue, Ethiopic, 319. rom, Gypsy, 65.
qrda, Berber, 198. rome, Coptic, 65.
quachichpalli, Nahuatl, 239. rost, OHGerman, 38.
quachpanyo, Nahuatl, 239. rotz, OHGerman, 38.
quachpepechtli, Nahuatl, 239. rowa, Southern Rodesia, 131
quacfdli, Nahuatl, 239. rowi. Southern Rodesia, 131.
quauhchimalli, Nahuatl, 231. royi, Southern Rodesia, 181.
quh', Arabic, 320. roz, OHGerman, 38.
quh'd, Syriac, 319. roza, Zulu-Kafir, 132.
queras, Norman French, 191. rozi, Nyanza, 131.
Quetzalcouail, Nahuatl, 316. rro, Zangian, 133.
quetzalli, Nahuatl, 316. man allah, Hausa, 311.
quimar, Arabic, 278. man sama, Hausa, 311.
qur'a, Arabic, 192. ruber, Latin, 5.
quarai'ah, Arabic, 193. rubrica, Latin, 5.
qurra', Arabic, 163. rudhiras, Sanskrit, 5.
nLdU, OBulgarian, 5.
r'ufara, Zenaga, 233.
rufus, Latin, 5.
ruQei, Guarani, 211.
ruma, Temne, 232.
rafalla, Soso, 241.
rumadiah, Arabic, 78.
Zenaga, 233.
r'afar,
rumeli, Syriac, 78.
ramadidh, Arabic, 78.
movdde, Lappish, 6.
rang, Persian, 137.
rust, ASaxon, 38.
rank, Arabic, 137.
mfyiilah, Arabic, 272.
raqaz, Arabic, 207.
r-aren, Timne, 173.
rasa, Arabic, 208. s, Upper Gambia, 134.
ra\lah, Arabic, 272. so, Malinke, 311.
WORD INDEX 389
sd,, Bambara, 311. sangirigiri, Bambara, 175.
sU, Malinke, 175. sard, Malinke, 175.
sd'ah, Arabic, 311. sS,ni, Malinke, 313.
sdba, Baga, 313. sdnila, Malinke, 313.
sabado, Spanish, 330. sanikela, Bambara, 313.
sdb'at, Arabic, 330. safima, Yoruba, 175.
sabbath, English, 330. santo, Mandingo, 312.
sdbih, Arabic, 158. ?aqala, Arabic, 17.
âdaqah, Arabic, 145. sar, Lalung, 7.
sadar, Arabic, 21. sar, Turkic, 28.
sad?ar, Beran, 133. Sar, Zenaga, 135.
sadsdrun, Adirar, 133. sara, Bambara, 175.
saed, Avestan, 8. Sara, Arabic, 175.
sSfe, Malinke, 312. saraka, Bambara, 143.
Sag, Sumerian, 17. sat, Annamese, 7.
sagala, Eufiji-Ruvuma, 134. saul^dn, Arabic, 341.
saâla, Arabic, 34. saulig, Arabic, 343.
saganâl, Arabic, 17. 'fx'f'"', Greek, 8.
Sagar, Arabic, 133. scindo, Latin, 8.
Sa^^dr, Arabic, 133. scuchar, OHGerman, 17.
Safârlah, Arabic, 133. sdar, Beran, 133.
saih, Arabic, 158. se. Central Nigerian, 134.
sakeld, Hindu, 18. êddr, Soa, 133.

sake. Western Zambezia, 134. seduma, Landome, 232.
sala, Nyanza, 133. segullo, Spanish, 33.
salaâ, Arabic, 343. segullum, Latin, 34.
salaka, Arabic, 349. segutilum, Latin, 34.
salasd'u, Arabic, 127. sSi, Cam, 7.
salimah, Hebrew, 232. sejar, Hassania Arabic, 135.
sail, Nyanza, 133. sel, Sengmai, 7.
sama, Malinke, 174. Selaib, Arabic, 87.
samd', Arabic, 312. semetti, Arawak, 222.
sama', Arabic, 168. ffc^WoXis, Greek, 53.
sama, Gbese, 175. sen, Andro, 7.
samxii, Gbandi, 174. ser, Garo, 7.
samak, Arabic, 311. sera, Yao-Ngindo, 134.
samxika, Arabic, 311. Set, Cameroons-Cross River, 134.
samanya, Malinke, 174. Se{, Egyptian, 9.
samaghe, Soninke, 311. Sethlans, Etruscan, 9.
samba, Bambara, 313. sher, Dhima, 7.
samba. Western Zambezia, 134. shi, Taungtha, 7.
samia, Bambara, 174. shingwo, Gbari, 136.
samian, Bambara, 174. shoora, Mech, 7.
saminya, Bambara, 174. shurr, Bara, 7.
Sdmiri, Syriac, 61. Si, Angola, 134.
samit, Coptic, 53. sibri, Malinke, 320.
samiy, Arabic, 174. sidara, Avestan, 8.
Samlah, Arabic, 231. siddri, Soa, 133.
samo. West Central Africa, 134. (TlSripos, Greek, 9.
samyen, Bambara, 174. si4i, Kannada, 8.
san, Malinke, 175. sig, Sumerian, 17.
sandu, Aruwimi-Lomami, 134. Sigbon, Nupe, 136.
sandyi, Bambara, 175. sigillum, Latin, 34.
sanfe, Bambara, 175, 312. ^kara, Bulom, 313.
sil, Sumerian, 7. smipa, Gothic, 52.
silumo, Bulom, 232. smoedum, ASaxon, 52.
UrnaX, Arabic, 231. smolt, ASaxon, 52.
simSldh, Hebrew, 231.
smultar, OHGerman, 51.
smylte, ASaxon, 52.
simila, Latin, 53.
smylting, ASaxon, 42.
similago, Latin, 53.
soke, OFrench, 343.
Simlah, Arabic, 231.
sole, OFrench, 343.
sind, Avestan, 8.
somaro, Vei, 174.
siparru, Assyrian, 7.
somyen, Bambara, 174.
slqll, Syriac, 17.
sor, Tipura, 7.
siri, Malinke, 233.
so-tigi, Bambara, 146.
sisa, Aruwimi-Lomami, 134.
soule, OFrench, 343.
sitti, Mandingo, 233.
souler, French, 344.
skuggwa, Gothic, 17.
soulette,OFrench, 343.
slaga, OHGerman, 344. sauqar, Arabic, 313.
slahan, OHGerman, 344. staerblind, OHGerman, 21.
Sleb, Arabic, 85.
stahal, OHGerman, 16.
?lina, Bode, 173.
stahel, OHGerman, 16.
Sma, Coptic, 53.
stahl, OHGerman, 16.
smaete gold, ASaxon, 52.
stakan, Russian, 18.
fffiaySiivw, Greek, 45.
stakla, OPrussian, 18.
ir/iiySos, Greek, 45.
stall, Coptic, 16.
smaidar, OHGerman, 40. star, OHGerman, 21.
smalt, OHGerman, 51.
smaltare, Italian, 45.
staraplint, OHGerman, 20.
steli, ASaxon,16.
fffid'Xe'ri, Greek, 45.
stelli, Coptic, 18.
smaltire, Italian, 45.
stikls, Gothic, 18.
smaltito, Italian, 45.
stiklo, OBulgarian, 18.
smalto, Italian, 45.
subaha, Malinke, 157.
ff/juiKToi, Greek, 45.
subahi, Hausa, 158.
smaltum, LLatin, 42.
subarha, Mandingo, 157.
smalz, OHGerman, 51.
subh, Arabic, 158.
smantire, Neapolitan, 48.
sudhr, Spanish, 33.
a/idpaySos, Greek, 45.
?ulg, Arabic, 349.
smardare, Neapolitan, 48.
siili^, Arabic, 343.
smardire, Neapolitan, 48.
suma, Baga, 232.
(T/iApSoi, Greek, 45.
suma, Bambara, 175.
smautire, Neapolitan, 48.
suma, Asante, 175.
smauto, Sicilian, 48.
suman, Arabic, 174.
smede, ASaxon, 52.
surnari, Asante, 175.
smedie, ASaxon, 52.
sungu. Central Congoland, 134.
smeduma, ASaxon, 53. sunsumma, Asante, 175.
smeidar, OHGerman, 52.
susUma, Asante, 175.
smelting, ASaxon, 42, 52.
Supparuru, Assyrian, 7.
smelzan, OHGerman, 51.
susu, Akra, 177.
sm.elzi, OHGerman, 51.
susu, Fante, 177.
smelzigold, OHGerman, 52.
susuma, Akra, 177.
smetume, ASaxon, 53.
SMSJtmo, Akra, 177.
smid, OHGerman, 52.
susuw, Asante, 177.
smida, OHGerman, 52.
Suva, Goali, 136.
smlda, Syriac, 52.
smidian, ASaxon, 52.
smilcit, OHGerman, 51. tab, Persian, 341.
smilting, ASaxon, 52. tdbah, Persian, 341.
WORD INDEX 391
tdbtab, Arabic, 344. tecuacan, Nahuatl, 228.

taelg, OHGerman, 39. tecunenenqtie, Nahuatl, 228.
Arabic, 35.
taf^lis, teeuti, Nahuatl, 228.
Arabic, 218.
jofej, iecutilia, Nahuatl, 228.
takhade, Soninke, 219. tecutli, Nahuatl, 228.
taia,Malinke, 192. tecutocaitl, Nahuatl, 228.
talammudz, Arabic, 112. tecuuia, Nahuatl, 228.
talango, Mandingo, 192. tehaset, Egyptian, 9.
talaqa, Arabic, 39. tejra, Kabyl, 135.
talata, Mandingo, 127. tela, Rufiji-Ruvuma, 133.
talbis, Arabic, 235. temir, Tatar, 5.
talSo, Coptic, 39. temur, Mongolian, 5.
tali, Malinke, 272. ten, Kaure, 136.
taltn, Malinke, 272. tende, Rufiji-Ruvuma, 133.
talgo, Coptic, 39. tene, Bambara, 139.
tali, Yao-Ngindo, 134. teflgo,Yao-Ngindo, 134.
ialmidz, Arabic, 112. tenu, Kasm, 136.
talq, Arabic, 39. teo, Kramba, 136.
talutium, Latin, 34. teotl, Nahuatl, 298.
tam, Songay, 113. tepal, Maya, 317.
tamha, North Central Congoland, tepani, Tarascan, 317.
134. tepaquaro, Tarascan, 317.
tamoanchan, Nahuatl, 279. tepamani, Tarascan, 317.
tamra-loham, Sanskrit, 5. tepeu, Kiche, 317.
tamu, Songay, 113. tepetl, Nahuatl, 317.
tan, Legba, 136. tequeba, Kiche, 228.
tana, Bambara, 192. iequi, Nahuatl, 229.
tana, Malinke, 139. tequina, Castilla del Oro, 228.
tangoman, African, 240. tequitl, Nahuatl, 228.
tapalcatl, Nahuatl, 349. tequiutl, Nahuatl, 228.
tapayolli, Nahuatl, 348. tequio, Nahuatl, 228.
tapayulli, Nahuatl, 348. tequiotl, Nahuatl, 228.
taper, French, 343. tequitlato, Nahuatl, 228.
tdro, Legba,_ 136. ter, Bulom, 317.
tariar, Arabic, 60. tera, Rufiji-Ruvuma, 133.
taseonio, Spanish, 33. tera, Mossi, 136.
tasconium, LLatin, 39. teri, Malinke, 279.
tasinia, Hausa, 278. terike, Bambara, 279.
ta?mir, Adirar, 232. terke, Bambara, 279.
tasunia, Hausa, 278. tete. Northern Cross River Basin,
taSurt, Berber, 345. 134.
taterman, OHGerman, 60. teisu, Japanese, 7.
tatsuniyu, Hausa, 278. tewuri, Tene, 135.
tauf, Arabic, 341. tezeatl, Nahuatl, 302.
te, Nahuatl, 229. Tezcatlipoca, Nahuatl, 302.
te. Western Zambezia, 134. thenda, Ukinga, 133.
teca, Nahuatl, 228. thende. North Ruvuma, 133.
teeari, Tarascan, 229. thi, Khougzai, 7.
tecoani, Nahuatl, 313. thiet, Annamese, 7.
tecolotl, Nahuatl, 316. thih, Sairang, 7.
tecpan, Nahuatl, 228. thili, Beni, 301.
tecpana, Nahuatl, 229. thlr, Chairel, 7.
tecpanir, Kiche, 228. thiri, Bot'iua, 301.
tectzecpa, Zoque, 229. ti, Nyanza, 133.
tla, Bagbalan, 136. tsi, Zulu-Kafir, 134.
tiangue, Am. Spanish, 240. ts'ieh, Chinese, 8.
tianguiz, Am. Spanish, 240. tSigbe,Nupe, 136.
iSigbon, Nupe, 136.
tianquizco, Nahuatl, 239.
tsigwo, Basa, 136.
tianquizmanako, Nahuatl, 240.
tsiSa, Berber, 331.
tianquizmanaloyan, Nahuatl, 240.
ts'it, Chinese, 8.
tianquiztli, Nahuatl, 239.
tsokachi, Hausa, 302.
tiantiztli, Nahuatl, 239.
tSu, Akra, 113.
Ware, Becuana, 134.
tsugba, Esitako, 136.
tiamictli, Nahuatl, 239.
is'uh, Chinese, 340.
tihara, Filham, 313.
tSulo, Akra, 113.
tibueli, Caraib, 348.
tsumo, Akra, 113.
t'ieh, Chinese, 7.
i^wam, Asante, 170.
tige, Bambara,135.
tu, Spanish Guinea, 134.
tigi, Bambara,158.
tub, Arabic, 341.
tiji, Zenaga, 301.
tubali, Hausa, 342.
tilbi, Songay, 232.
tilmatli, Nahuatl, 231.
tufa, Bambara, 342.
tufali, Swahili, 341.
tio, Mandingo, 158.
tuffe, Swahili, 341.
tir, Shonshe, 7.
tugma, Ebe, 136.
iira, Yoruba, 136.
tuhui, Arabic, 218.
tlrah, Arabic, 136.
tuigjan, ASaxon, 40.
tirh, Lai, 7.
tuluq, Arabic, 40.
tiri, Foro, 135.
tumbu. North-west Zambezia, 134.
fit, Chinese, 7.
titere, Spanish, 60.
tunculuchu, Maya, 316.
ixo, Cameroons-Cross River, 134.
iitra, ONorse, 60.
tya, Cameroons-Cross River, 134.
tlv, Dselana, 136.
tylg, ASaxon, 39.
tjal, Berber, 107.
tzak, Pokonchi, 229.
tjari, Chauya, 135.
tzakal, Pokonchi, 229.
tjere, Songay, 279.
tzakol, Pokonchi, 229.
tjero, Songay, 279.
tzaqualli, Nahuatl, 286.
tjert, Chauya, 135.
Greek, 340.
T^vKdi>iov,
tjur, Chauya,135.
Greek, 340.
TfKKawo-TiJpioi',
tlachco, Nahuatl, 344.
T^vraWfw, Greek, 340.
tlachtli, Nahuatl, 349.
Ware, Becuana, 134.
tlata-ghe, Dahome, 127.
tlauiztli, Nahuatl, 235. uacili,Nahuatl, 314.
tne, Malinke, 139. uad, Berber, 187.
tob, Hassania Arabic, 342. udda, Berber, 187.
togore, Peul, 232. udkor-bar, Sumerian, 6.
tolonyi, Soso, 192. 'udzah, Arabic, 128, 178.
toltecatl, Nahuatl, 240. uetula, LLatin, 324.
toUecauia, Nahuatl, 240. Ueuecoyotl, Nahuatl, 255.
toluactli, Nahuatl, 315. ufuma, Kiriman, 230.
tondd, South Congoland, 134. ugnu, Assyrian, 16.
tondo. Western Zambezia, 134. 'ugruf, Arabic, 320.
tongo, Yao-Ngindo, 134. ukku, Kannada, 1.
truchement, French, 180. uknu, Assyrian, 16.
tsanrm, Bornu, 173. 'ukrah, Arabic, 341.
tsdra, Baga, 313. iil, Tamil, 40.
tse, Cameroons-Cross River, 134. ulku, Kannada, 40.

tsetsala, Timne, 313. ulli, Nahuatl, 348.
WORD INDEX 393
'umma 'amir, Arabic, 162. womi, Adanme, 178.

'umrah, Arabic, 162. woii, Akra, 177.
'unia, Arabic, 16. wot, Yoruba, 177.
ur, Sumerian, 1. wondi, Malinke, 188.
ur, Kannada, 2. woMMmo, Akra, 178.
ura, Tamil, 2. wonsuomo, Akra, 178.
urhu, Kannada, 1. wontSemei, Akra, 178.
urium, LLatin, 36. wontSumo, Akra, 178.
urku, Kannada, 1. wchyeli, Akra, 178.
urraida, Basque, 5. woyeli, Akra, 178.
uru, Gbese, 135. wudi, Soso, 135.
uru, Kannada, 2. wuh, Akra, 178.
urudu, Sumerian, 1. wuri, Soso, 135.
urukku, Tamil, 1. wy, Malagassy, 7.
uruvu, Kannada, 2. wyiasefo, Asante, 174.
'uSara', Arabic, 279.
'utr, Arabic, 12.
xeque, Portuguese, 158.

^^piov, Greek, 38.
vadimonium, Latin, 187. Xohtl, Nahuatl, 279.
vadium, LLatin, 187. x-qna, Tzental, 247.
valuka, Sanskrit, 34. x-qnaulai, Tzental, 247.
vibrantia, Latin, 24.
vipilli, Nahuatl, 233.
vodu, Dahome, 128, 178. ya, Akra, 209.
vodUhwe, Dahome, 178. yano, Akra, 209.
vodHno, Dahome, 178. yar, Padsade, 317.
voodoo, English, 129. yara, Akra, 209.
vwama, Kongo, 230. yele, Malinke, 209.
yelema, Bambara, 209.
yere, Akra, 209.
wada, Hausa, 187. yipura, Yula, 320.
wada'a, Arabic, 187. yire, Takponin, 135.
wo'd, Arabic, 187. yiri, Gbin, 135.
wadda, Hausa, 187. yite, Soninke, 135.
wadi, Gothic, 187. yiyi, Toronka, 135.
waî', Arabic, 187. ymaeti gold, ASaxon, 52.
wadl'ah, Arabic, 187. yra. Ewe, 194.
wakhale, Wolof, 230. yre. Ewe, 194.
wall, Bambara, 188.
wall, Arabic, 188.
wasei, Carolinian, 7. za-bar, Sumerian, 6.
wail, Arabic, 35. zag, Hebrew, 17.
waim, Arabic, 137. zagd^, Arabic, 17.
wele, Fernandian, 134. zagangal, Arabic, 17, 302.
wesi, Javanese, 7. za-gin, Sumerian, 16.
m, Maori, 7. zakku, Hebrew, 17.
wo, Adanme, 178. zalll, Arabic, 300.
wodda, Hausa, 187. zam^ Songay, 113.
wodzi, Akra, 177. zam a, Arabic, 170.
woli, Malinke, 188. zaman, Arabic, 170.
wolo, Akra, 178. zami', Arabic, 170.
wolomo, AlJra, 178. za'r, Arabic, 78.
woma, Asante, 178. zar, Turkic, 28.
zaranya, Avestan, 28. zi§a§, Arabic, 17.
zaray, Avestan, 28. zU, Arabic, 300.
zarona, Avestan, 28. zil, Turkic, 28.
feipai, Greek, 232. zim, Arabic, 170.
ielizo, OBulgarian, 4. zirir, Persian, 28.
zer, Persian, 28. zitaron, OHGerman, 60.
zerd, Persian, 28. zka, Aramaic, 17.
zervn, Persian, 28. zkdra, Berber, 114.
zerlr, Persian, 28. zkh, Hebrew, 17.
zâ^, Arabic, 302. zullah, Arabic, 300.
;
SUBJECT INDEX
Abyssinia, gypsies in, 95 f. 196 ff.; and interpretation of
Africa, its fetishism chiefly of Arabic dreams, 197; their sword play and
origin, 131. African gun firing, 206; their laife
AFRICAN FETISHISM AND in America, 228 f their Simlah, in
. ;
TOTEMISM, 116-141. America, 231 f.; in the Sudan, 232;

'A^Hz of the Arabs, and the "old their mai'zar, in America, 232 f.;
woman," 325 flf.; among the Ber- in Africa, 233; their gafr, in Africa,
bers, 326 ff. 233; in America, 234; their laâm
Akra "spirit" words, and Arabic in Africa and America, 234; their
qara', 164; and Arabic "amulet," labisu in Africa, 235 f.; their kisa'
177 f. in Africa and America, 239; their
Alchemy, and the seven metals, 18; tarâmdn in America, 239 f.; their
and the Arabs, 40. dabbar in Africa, 240; their naba
Altar, three-pointed, of the Habbes and American nagualism, 249;
triune god, 295 f. their ^blj, in the Nama worship,
Amanteca, confusion of the mythical, 252; their use of honey as medicine
with the artisans, 245 f. in the Sudan, 252; history of their
Amantecatl, history of the word, 241 ga4wal, 268 ff.; their geomancy in
f. Africa, 277 f.; their theology
Amauta, Peruvian, the amantecatl of among the Mossis, 295; their
the Mexicans, 241. baraka and the eating of porridge
Amulet words, African, from the at New Year, 332; brought ball
Arabic, 129 £f. game to Europe, 342.
Ancestor worship in Africa and the AREYTO, THE, 199-212.
bori, 159. Areyto, history of the, 199 ff.; and
Arabs, their metallurgy in Europe, eulogy of the dead, 201.
40; their words for "gypsy," 78 ff. Arttw Poc/jo on Arabic blazonry, 141,
gypsies among the, 85 fi.; their 236 ff.
week among the Mandingos, 127; Asante fetishism and sufism, 164 ff.;
their "amulet" words in Africa, "spirit" words and Arabic qara',
129 ff.; their "medicine" words in 164; festivals and Arabic greater
Bantu, 131 ff.; Arabic "tree" as and lesser festivals, 169 f.; "dyna-
foundation for African "medicine" mic force" and Arabic bar a', 171;
words, 133 ff.; their medicine and "spirit" words and Arabic
African fetishism, 136; their bla- "heaven," 175 ff.
zonry in Africa, 141; their âdaqah Athinganoi as gypsies, 66 f.; see
and Bambara sacrifice, 145; their Touch-me-nots.
"religion" and Malinke nama, Avicenna on steel, 11 f.
160; their "reading" and Sudanic

fetishism, 163 f.; their festivals
among the Asantes, 168 f.; their Bacdbs, Plate of the, and Brinton,
"heaven" and Sudanic "spirit" 266; and Cyrus Thomas, 267 f.;
words, 175 ff.; their "amulet" and and the Arabic §a4wal, 269 f.; and
Sudanic "spirit" words, 178 f.; the Mound-builder gorgets, 273 f.
Arabic qara' in America as Caraib, Ball game and baraka among the
.^ 190 ff.; Arabic 'isiihdrah, and im- Berbers, 334 ff.; in Persia, 340; in
precation in Africa and America, Byzantium, 340 fi.; brought to
. . . .

Europe by the Arabs, 342; among history of the word in Africa, 158
the European clergy, 343; in Mex- flf.; among the Hausas, 160; and
ico, 344 ff.; and the intercalary by Ramon Pane, 223; in America,
days, 344; its religious significance
223 fi.
in Mexico, 347 f.; and the spindle

Bosman on the funeral ceremony in
the Sudan, 203 flf.
in Mexico, 349; and the vernal
equinox, 350 f
Bowl of the Bacabs and the dasiri
tree, 266 f
Bambaras and the bori, 143 S.; their
Brinton on nagualism, 246 fl.; and
fetishism a retrograded Islamism,
the Plate of the Bacabs, 266.
143; borrowing dervish practices,
Bronze represented red in Egyptian,
153 f.; their fetishism compared
3; see Brundisium.
with that of the Asantes, 168 f.;
Brotherhood, Islamic, and the Bam-
and the spider stories, 272; see
bara kuare, 263 f
Mande.
Banner, Mexican, and Persian band,
Brun on African totemism, 139 f.
Brundisium, and mirrors, 22.
238.
Bantu, Egyptian "medicine" words
in, 131 fi.; Arabic "medicine" Cacique, its history, 158.
words in, 132 ff. Calends of January and the "old wo-
Baraka in the Sudan and Arabic man," 323 f.
na'am, 332 f.; the chief, among the Cannibalism in Brazil of a religious
Berbers a game of ball, 334 ff. character, 190.
Bee culture andNama worship, 252; CARAIB SOCIAL ORDER, THE,
and the Ekchuah worship, 258. 218-221.
Beryl and blindness, 20 f. CARAIBS, THE, 180-198.
Bezoar stone of the Arabs in Peru, Caraibs, as described by L6ry, 180
305. ff.; as a ghost word, 185; in Brazil,
Bird element in African fetishism, "strangers," 185; soothsayers, ac-
214 f. head on Mound-builder gor-
; cording to Stade, 188 ff. ; from Ara-
gets, 273; of omen in Africa and bic qara', 190 ff.; refrain of their
Arabic du'a', 314 f.; and Quetzal- songs in Africa, 193 ff.; and spit-
couatl, 314 f. ting, in America and Africa, of
Black god, see Medicine god. Arabic origin, 197; refrain of their
Blacksmiths, in the Western Sudan, songs in America, 197; their
98 fif.; words for, in the Western "dance" words, 209 f.; and nobil-
Sudan, 112 flf.; among the Berbers, ity, 219 ff.; and household serv-
113 f.; see Griots. ants, 221.
Blazonry, history of, 137 ff.;in Amer- Cords and dice related, 278 f
ica, 235; in Europe, 236; its Per- Carnival and the ball game, 348.
sian origin, 236 f. Castes in the Western Sudan, 98 ff.
Blindness and the beryl, 20 f.; and Cave of Treasures and the gypsies, 63
the hyena, 21. ff.; originated in VII. c, 68.
Boli, see Bori. Ceiba tree, ceremonies connected with
BORATIO, THE, 222-227. the, in Africa and America, 286.
Boratio, see Bori. Chalybes and prpsies, 55 ff.
BORI, THE, 143-162. Chastisement in the Bambara and
Bori, among the Bambaras, 143 flf.; Quetzalcouatl worships, 264.
confusion of, with the Arabic înn, Cheron on the griots, 103 ff.
143; major and minor, 150; cult of Chicha in America and porridge of
the, 150 ff.; three groups of, 152; the Berbers, 331.
talking, 152 f.; derived from Ara- Cimi, American fetish, 227.
bic fpoidl, 153; bloody sacrifices Conical hat of the Magi and in Africa,
to, related to pre-Islamic practices, 3 1 8 ff in Mexico of Persian origin,
. ;
153; talking, in Jobson, 154 flf.; 321.

. . .
SUBJECT INDEX 397
Contaminations, philological, in the Egypt, gypsies in, 79 ff.

African lanâges, 171. Egyptian culture and Sumerian, 9;
Copper, its history, 1 flf.; and iron applied to gypsy, 65 f.; element
words confused in India, 1; pro- in African fetishism, 131 ff.
ceeded from a non- Aryan race, 5; Ekchuah of the Mayas corresponds
alloys and philology, 21; alloy and to Mandingo Nama, 255 fl.; his
brundisium, 22. long nose the beak of the Nama
Coptic word in gypsy language, 65; god, 259.
a gypsy word in, 68 f Enamel, history of, 43 ff.
Coyote, see Hyena, Ueuecoyotl, Coyot- Eulogy of the dead at an Arabic
linauatl. funeral, 201 ff.; in the Sudan, 203
Coyotlinauatl, the Mexican coyote ff.; words in the Sudan of Arabic
god, 245 f.; precise rendering of origin, 208 f.; among the Mandin-
Mandingo Namakoro, 250. gos, as related by Ibn-Batutah,
Crescent in Arabic blazonry and on 209.
Mexican shields, 237 f.
Cross on Mound-builder gorgets_ iiot
Christian, but African in origin, FEATHERS AND MASKS, 213-
271. 217.
Crown, see Conical hat. Feathers and fetishism, 213 ff.
Festivals among the Asante and the
Arabic festivals, 168 f.
Festus, interpolation in, 15 f.
Fetish day among the Mandingos,
127 f.; as understood by de
Dance and song in America, 199 flf.; Marees, 128; and grigri, 129 f.
funeral, in Arabic, 207 f.; words in FETISHISM AND SUFISM, 163-
America, 211 f.; god, Ixtlilton, 179.
and medicine god, 252 ff. Fetishism in Africa, as described by
Dasiri, Bambara, and the minor fes- de Marees, 116 ff.; and popular
tival of the Arabs, 146 f.; and Que- medicine, 131; in Africa chiefly of
tzalcouatl, 260 f.; his keepers old Arabic origin, 131; Egyptian ele-
men, 264; and tree worship, 264; ment in, 131 ff.; according to De-
and the secret society of the kuare, lafosse, 157; and feathers and
264.
masks, 213 fl.; American, and the
among the cimi, 227.
Defensive armor, Arabic,
Fetishist, boratio in Venezuela, 223
Mandingos, 235.
ff.; see Bori.
Delafosse on Mandingo fetishism,
Firing of guns in the Sudan and eu-
157.
logy of the dead, 204; and Arabic
De Marees on African fetishism,
116 ff.
sword play, 206.
Dervishes, their practices in the Bam-
Freemen of the Western Sudan, 98 f.
bara ritual, 153. Funeral ceremony in the Sudan and
Dice and cards related, 278 f eulogy of the dead, 207; in Amer-
Dravidian, its "strong" words related ica, 209 f
to Sanskrit, 2 f.; compared with

Sumerian, 6.
Dreams, interpretation of, and the Gadwal, history of the, 268 f.; and
Arabic isfihdrah, 196. the Plate of the Bacabs, 269 f.;
Duga, bird of omen in Africa and meaning table of the zodiac, 271;
tecolotl in Mexico, 316. ornamentation among the Hau-
Duho, see Stool. sas, 273; on Moundbuilder gor-
Dya of the Bambaras the "shade of ' gets, 273 and tonalamatl, 274;
f.;
the Arabs, 301 f. and pacMsi, 278.

. . ; .

Garments, Mexican, of African origin, Habbes and their triune god, 295 f.
230 ff. Hausas and their bori worship, 160;
Gemini and Xolotl, 279 f and their spider stories, 272; their
Geomancy, Arabic, in Madagascar, "story" words and Arabic geo-
274 fl.; among the Yorubas, 277; mancy, 277 f.
and "story" words in Hausa, 277 Heaven words in the Sudan, 172 ff.
f. ffelebis, gypsies of Egypt, 79 ff.
Ghost word in the glosses, 15 f. Henry on the Bambara bori, 143 ff.,
150 ff.; criticism of his concept of
(jinn, Arabic, and the Bambara spir-
African fetishism, 145.
its, 143 ff.
Heraldry, see Blazonry.
Glass words in Sumerian and Arabic,
Hieroglyphic ivriting of Mandin-
16 f.
gos like that of the Tuxtla statu-
Goblins and gypsies, 56 f ; history of
.
ette, 271.
the word, 59 f
God and "heaven" confused in Af-
Hindu metallurgy in Egypt, 65.
i?<me^ as medicine in Africa based on
rica, 223; among the Caraibs, 227;
among the Mossis, 295. a Koranic prescription, 252.
Gold digging in Pliny's account, 28
Hyena and blindness, 21; and the
ff.
Nama worship, 161 f. of the Man-
;
dingos becoihes the coyote of the

Golf, see Ball game.
Mexican religion, 250; see Coyot-
Gothic Bible and "iron," 13 f.
linauatl.
Gourd and rattle related through a
philological contamination, 192 f.;
rattles, see Rattle.
Grigri, see Fetish.
Griots in the Western Sudan, 100 ff.; Ibn-Batutah on the griots, 100 f. and ;
in Ibn-Batutah, 100 f.; in Sene- the Mande totem, 140.

gambia, 106; no longer pure gyp- Ibn-Errumi on male and female steel,
sies, 107 f.; Niger the river of the, 12.
108. Ibn-Haukal on steel, 12.
Guatiao, peaceful Indian, from Ara- 'Id-al-sagir, the minor festival, 146 f
bic wadla'h, 186 ff. Ilia in Peru and the triune divinity,
GYPSIES IN AFRICA, 78-115. 303; something that is old, 305;
GYPSIES IN EUROPE, 54-77. refers to celestial divinity, 306.
Gypsies in Charlemagne's law, 54; in Intercalary days and the "old wom-
Europe, 54 ff and Chalybes, 55 ff.
. ; an," 330 ff.; and good cheer, 331;
as cheats, 55 ff.; and goblins, 56 f.; and the ball game, 334 ff.
in Jannes and Jambres, 63 ff.; and Interpolations in Pliny's account of
the Cave of Treasures, 63 ff.; gold digging, 34 ff.
Coptic word in Gypsy, 65; as Iron, its history, 1 ff.; and copper
Egyptians, 65 f.; and the Athing- words confused in India, 1 repre-
;
anoi, 67 f.; Gypsy word in Coptic, sented blue in Egyptian, 3; in the

68 f.; in Egypt in the VII. c, 68; Tatar and Mongolian languages,
Gypsy word for "Greek," 68; 5; proceeded from a non- Aryan
Gypsy in Greek, 69 f., 73; in the race, 5; in Chinese and other lan-
Balkans, 70 ff.; and Mandopoli, 73 guages, 7 f.; in Semitic, 7; in the
ff.; in Arabic, 78 ff.; called Belebi, Malay languages, 7; in Greek and
79 ff ; called Ndwars, 79 in Egypt,
.
; Etruscan, 9; in Ethiopic, 10; in
79 ff.; Sleb of Arabia possibly, 85 Persian, 10; male and female, 12;
fl.; in Abyssinia, 95 f.; among the in the Germanic languages, 14;
Somalis, 96 ff.; called âli in the and the gypsies in the VII. c, 55 ff.
Western Sudan, 107; their migra- Islam and Bambara fetishism, 143.
tions in Africa, 108 ff.; and tarring Ixtlilton related to Ueuecoyotl, the
and feathering, 214. "old coyote;" see Medicine god.
.
SUBJECT INDEX 399
Jannes and Jambres and the gypsies, Merchant words in America, 230; in
60 ff.; in the Koran, 61; and the Mexico, 239; traveling, god of the,
gypsies of the VII. c, 76 f. and Ekchuah, 259.
Jobson on the griots, 101 ft.; and the Metals first worked in Central Asia,
talking bori, 154 S.; on the Man- 9; the seven, 18.
dingo funeral ceremony, 206 S. MEXICAN NEW YEAR, THE,
. 323-351.
Mexican civilization and the Man-
dingos, 228 ff.; garments of Afri-
Komo worship among the Bambaras
can origin, 230 ff . ; triune divinity,
and Asante kom, 168.
296 political
f.; structure of
Kuare, secret society of the Bamba-
African origin, 317 ff.; calendar
ras and the Islamic brotherhood,
and the "superfluous" days, 334 f .;
264.
and the ball game, 344 ff.
Minor festival of the Arabs, 146 f.
Mirrors, Brundisian, 22.
Lapidary of Aristotle on steel, 14. Mossis and their one god, 295.
Lery and the Caraibs, 180 ff. Moundrbuilder gorgets, spider on,
like that on Mandingo inscrip-
tions, 271; and the Plate of the
Bacabs, 273 f.
Madagascar, Arabic geomancy in,
274 ff.
Mummery prohibited by the Coun-
cils, 323 f.; of the Arabs, 325; see
Mande totem and Ibn-Batutah, 140 Veiula.
f "read" words and Arabic qara',
. ;
163 f.; "god" words from the

Arabic, 170; see Mandingos, Bam- Nagualism, see Naualli.
baras. Nama, its philological history, 160;
Mandinga, South American word for of the Malinkes corresponds to
fetishist, 226. Bambara Dasiri, 160 ff.; and the
MANDINGO ELEMENTS IN hyena, 161 f.; celebration of the,
THE MEXICAN CIVILIZA- among the Bambaras, 250 f.; a
TION, THE, 228-322. confusion of Arabic naba and
Mandingos, griots among the, 100 'amdn, 251 f.; and bee culture,
ff.; their social order of non- Afri- 252; and the Arabic §ubh, 252; see
can origin, 113 ft.; their seven-day Hyena.
week of Arabic origin, 127; and Namakoro of the Mandingos the an-
fetish day, 127 f.; and the Amer- tecedent of the Mexican Coyot-
ican Caraib, 191 ff.; their defen- linauatl, 250; and Ueuecoyotl, 255.
sive armor of Arabic origin, 235 Nanauatzin, god of syphilis, 280 f.;
f.; their gods, 293 f. and his pyramidal altar, in Amer-
Mandopoli and the gypsies, 73 ff. ica and in Africa, 286.
Mantle, Mexican, and the device Naualli, Brinton's history of the
upon it, 235. word, 246 ff.; and its derivatives
Mask and fetishism, 213 ff.; bird, in from Arabic naba, 249 f.
Africa, 214 f.; and dance in Amer- Nine Lords of the Night in Mexico and
ica, 215 ff. in Africa, 298.
Medicine, popular, and fetishism, Nobility, Caraib, and African totem-
131; words in Africa of Egyptian ism, 220.
origin, 131 ff.; and "tree" in Afri-
can languages, 133 ff.; honey as
chief, 252. Old god, the, in Mexico, 305.
Medicine god in Mexico, 252 f Old woman, the, a philological blun-
Melchizedekites and Touch-me-nots, der, 326; and intercalation, 330 ff.;
66 ff. see Vettda.
Ometecutli of the Mexicans and of voodoo, 129; of fetish, 129; of
Ammo of the Habbes, 297. grigri, 129 f; of African "medi-
Owl, bird of omen, in Africa and in cine" words, 131 ff.; of standard,
America, 316.
137; of rank, 137; of Bambara da-
siri, 150; of Mandingo subarha,
157 f.; of cacique, 168; of Malinke
Pachisi and the ga4wal, 278. nama, 160; of Arabic gara in the
Paradise of the Mexicans and the Sudan, 163 f.; of Asante kom, 166
Malinkes, 279. ff.; of Arabic âm' in the Sudan,
Patecatl, history of the word, 241. 170 f.; of Arabic zam'a in the Su-
Pehlevi origin of Athinganoi, 67. dan, 170; of Arabic bara' in the Su-
Peru, its amauta, the amantecatl of dan, 171 f.; of Arabic al-§annah in
Mexico, 241; and the triune divin- the Sudan, 172 ff. ; of Taino guatiao,
ity, 302 f.; and the bezoar stone, 185 ff.; of LLatin caragius, 191; of
305. "dance" words in the Sudan, 208;
Peuls, history of the, 108 ff.; and of "eulogy" words in the Sudan,
Touch-me-nots, 109; and Zotts, 208 f.; of Taino areyto, 211; of
111; their migration to the Niger, American "dance" words, 211 f.;
111 f. of Arabic rasa in America, 211 f.;
Philology of Sumerian urudu, 1 ff.; of Mande raansa, 218; of Taino
of "strong" words, 2 ff.; corre- duho, 218; of hori words in America,
spondence of y and r in Sanskrit- 226; of Mexican tec, 228 f.; of
Dravidian, 3f.; of Dravidian "cut" Maya p'olom, 229 f.; of "merch-
words, 4; of Greek ;taX/c(is, 4 f.; of ant" words in Africa, 230; of Na-
"red" words in Europe, 5 f.; of huatl chimalli, 231; of Nahuatl
Sanskrit lohas, 5; of Tatar temir maxtli, 232 f.; of Nahuatl vipilli,
"iron," 5; of Assyrian parzillu, 6 234; of Spanish naguas, 234; of
ff.; of Sumerian bar, 6 f.; of Semi- Nahuatl panitl, 238; of Nahuatl
tic and Malay "iron" words, 7; quachtli, 239; of Nahuatl tiamictli,
of Chinese "iron" words, 7 f.; of 239 f.; of Arabic dahbar in the Su-
Latin cuprum, 7; of "cut" words dan and in America, 240 f of Na-
. ;
in Indo-European, Sumerian, and huatl toltecatl, 240; of Nahuatl

Semitic languages, 8 f.; of "axe" amantecatl, 241; of Nahuatl pate-
words, 9 f of Assyrian l}asinnu, 9
. ; catl, 241 of Arabic naba in Africa,
;
ff.; of Arabic 'atr, 11 ff.; of "steel" 250; of "syphilis" words in Africa

words, 12 ff.; of Gothic eisarn, 13 and in America, 285 f.; of Nahuatl
f.; of French acier, 14 f.; of LLatin tzaqualli, 286; of Bambara dya, 300
andena, 16; of "steel," 16 f.; of f. of "mirror" words in Africa and
;
"glass" words, 16 ff.; of Arabic in America, 302; of Kechua ilia,

filizz, 18 f.; of irXdvris, 18 ff.; of 304 f.; of "serpent," "heaven,"
beryl, 19; oi flint, 19 f.; and blind- and "rain" words in Africa and in
ness, 20 f.; of German starblind, America, 312 f.; of "buy" words
21; of bright, 23 f.; of brand, 24 ff.; in Africa and in America, 313; of
of OHGerman aruz, 26 ff.; of Nahuatl Qu^tzalcouatl, 316; of
"gold" words, 27 ff.; of mineral, Nahuatl tepetl, 317 f.; of "hat"
36 f.; of Arabic talq, 39 f.; of elect- words in Africa, 318 ff.; of hag,
rum, 40 ff.; of smelt, 42 ff.; of m^lt, 330; of "witches' sabbath," 330;
44 f.; of amalgam, 49; of German of Arabic na'am in the Sudan, 332
massing, 50; of smith, 52 f.; of gob- f.;of Arabic wo' am in America, 334;
lin, 57 ff.; of Gothic hauns, 58; of of "ball" words in Africa, 341 f.; of
coward, 59; of Romany, 65 f.; of Arabic saulân, "beat," in Europe,
Greek KufioSpbiwi, 69; of OHGerman 343 f.; of "ball" words in America,
chaltsmid, 76; of German Zigeu- 348 f.; of "spindle" words in Amer-
ner, 78; of Mandingo §ali, 107; ica, 349 f.
. . . .
SUBJECT INDEX 401

Planets, the seven, and the seven Shfde in Arabic conception, 300;
metals, 18, in Berber and Sudanic languages,
Pliny's account of gold digging, 28 ff 300 f.; and Tezcatlipoca of the
Polo, see Ball game. Mexicans, 302.
Porridge and the "old woman," 331. Shield, Mexican, and the device upon
Pulque as medicine corresponds to it, 234 f.; round, and blazonry
honey in the Sudan, 254. among the Arabs, 237.
Pyramidal altar of god of syphilis in Singing, see Dance.
Africa and in America, 286. Sky and serpent in Africa and in Am-
erica, 308 ff.
^leb of Arabia, possibly of gypsy
a tutelar deity that
Quetealcouatt, origin, 85 ff.
gives rain=Bambara Dasiri, 260 Smiths in Egypt and Assyrian cul-

f.; and tree worship, 264; as time- ture, 9 f.
measurer related to Bambara Somalis, gypsies among the, 96 ff.
Dasiri, 264; as an old man, 264; Sorcery among the Mandingos, 157.
and the Plate of the Bacabs iden- Spider on Mandingo inscriptions
tical with the tree of the Dasiri, like that on the Mound-builder
266 and the uactli, 314; the
f.; gorgets, 271; in Africa source of
Mexican god, 315; and Owl-land, many stories, 271 fl.; and proverb
316. the same in Bambara, 272; stories
of the Tchis, 272; stories of the
Hausas told in its name, 272 ff.;
Raids of the Arabs and the Mandin- and story due to Arabic origin, 272
f.; and the zodiac, 273.
go were-wolf, 158.
Spindle in Mexico and the ball game,
Rain words in the Sudan and Arabic
349.
"heaven," 172 ff.; god of, the
Spirit words in Asante, 164.
Dasiri of the Bambaras and
Stade and the Caraib soothsayer,
Quetzalcouatl of the Mexicans,
188 ff.
261.
Steel, hard and soft, 11 f.; see Iron.
Rattle, African words for, from Ara-
Stool, history of the royal, in Amer-
bic mitraqah, 192; gourd, of Afri-
ica, 218; among the Asantes, 219.
can fetish man, 192; in North
America, 1 92 f.; in Brazil, warafca^
Subarha in Mandingo belief, 157.
Arabic mitraqah, 193. Sufism and African fetishism, 163 ff.;
Richard I, and tarring and feather-
and Asante fetishism, 164 ff.
ing, 213.
Sumerian related to Dravidian, 1 ff.;
its "strong" words identical with
Royal dignity, see Stool.
Dravidian, 2 f.; origin of Egyptian
Roger Bacon on steel, 11.
culture, 9; and Chinese words for
"gold."
"Superfluous" words in Mexico and
Sacrifices, Bambara, of Arabic ori-
the intercalary days, 334 f.
gin, 43 S.; no, to the old god of
Syphilis and the itch confused, 280
the Mexicans, 297.
f.; the same word for, in Malinke
Sahagun on the amanteca, 241 ff.
Samaritans and Touch-me-nots, 63.
and Nahuatl, 285 f
Sanskrit, its "strong" words iden-
tical with Dra vidian, 2 f Talisman, see Fetish.
Semitic, its "strong" words identi- Tarring and Feathering and Arabic
cal with Dravidian, 2 f ta'slr, 213 f.; as in a law of
Serpent and the Mexican Tlaloc, Richard I, 213 f.; and the gypsies,
307; and sky confused among the 214; among the Bororo, 216; at
Bambaras and in Americaj 308 ff.; Cumana, 216; among the Bakai-
rain, and heaven in America, 312. ris, 216; of African origin, 217.
. .

Tattooing in Africa, its connection Tutelary god, see Dasiri and Qnetzal-
with blazonry, 138 f. couall.
Ta'zlr of the Arabs and tarring and Tuxtla statuette, its hieroglyphics
feathering, 213 f. like those of the Mandingos, 271.
Tchis and their spider stories, 272. Twenty of the American calendar of
TecoMl of the Mexicans and African African origin, 270.
duga, 316. Twins, their treatment in Africa, 286
Teotl of the Mexicans and the sun, flf.; in a cooking pot due to philo-
298 ff. logical speculation in Arabic, 293.

Tepetl of the Mexicans and Mandingo
debe, 317 f.
Tezeatlvpoca and Arabic "shade," Uaetli and Quetzalcouatl, 314.
302. Ueuecoyotl as a dance god, 255; re-
Thirteen in America, 270; parts of lated to Coyotlinauatl, 255; pre-
zodiac in Africa, 270 f.; of the cise translation of Bambara Nama-
American calendar of African ori- koro, 255; the old god of the Mexi-
gin, 270 f.; days' time unit among cans, 296.
the Arabs and the Hindus, 271;
among the Berbers, 271; in cards
related to thirteen divisions of the Vernal equinox and ball game, 350 f
zodiac, 279. Vetula and the Graeco-Latin glosses,
Thomas, Cyrus, and the Plate of 323 ff.
the Bacabs, 267 f. Vincent of Beauvais on steel, 11.
Timotheus Presbyter and the Melchi- Voodoo words and African "amulet,"
zedekites, 66. 178 f.
Tlaloc, the Mexican rain god, and
the serpents, 307; and Xolotl con-
fused, 308.
Week of African tribes, 270.
Were-wolf and Mandingo subarha,
Tobacco and the horatio, 227.
157 f.
Toltecs, history of the word, 240 f.
Western Sudan, castes in the, 98 ff.
Tonalamatl and the Sa^'wal, 274.
Totemism, African, its history, 137
flf.; among the Mandes, 139; in the Xolotl and the twins, 279 f.; of Ara-
Western Sudan, 139 f.; and nobil- bic origin, 280; and the cooking
ity among the Caraibs, 220. pot, 280; and Nanauatzin, 280 f.;
Tovch-me-nots and Samaritans, 63; and the ball game, 347 f.; and Ar-
and Melchizedekites, 66 flf.; and abic 'w^Hra, 348.
Peuls, 109.
Tree and "medicine" in African lan-
Yaeatecutli, the Mexican god
of the
guages, 133 ff.; the Bambara Da-
siri and Quetzalcouatl connected
merchants = Ixtlilton,
259.
Yorubas, their geomancy of Arabic
with the, 264 f
origin, 277.
Triune gods of the Mandingos, 293;
among the Habbes, 295 f.; of the
Mexicans, 296 f.; of the Mayas of Zodiac of thirteen parts on West-
Christian ori^n, 298; of Africa of African calabashes, 270; and the
Christian origin, 298; in Peru, spider, 273.
302 f. Zotts ia Africa, 111.

Africa and The Discovery of America III

Uploaded by

Copyright:

Available Formats

Africa and The Discovery of America III

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Africa and The Discovery of America III

Uploaded by

Copyright:

Available Formats

What is the book about based on the table of contents?

What is the book about based on the table of contents?

What institutions are mentioned on page 2?

What institutions are mentioned on page 2?

CORNELL

BOUGHT WITH THE INCOME

The original of this book is in

the Cornell University Library.

There are no known copyright restrictions in

INNES & SONS

VII. THE CARAIBS ... 180-198

IX. FEATHERS AND MASKS 213-217

XIV. CONCLUSIONS 352-370

The powerful Moslem interpenetration in spiritual

xii AFRICA AND THE DISCOVERY OF AMERICA

Bacon, R. See Brewer.

Cardinall, A.W. The Natives of the Northern Territories of the

Defrfimery, C. and Voyages d'Ibn Batoutah, vol. IV, Paris 1879.

Ellis, A. B. The Tshi-Speaking Peoples of the Gold Coast of

Perrand, G. Les Musulmans a Madagascar, vol. I, Paris 1891.

Garcia Icazbalceta, J. Memoriales de Fray Toribio de Motolinia, Mljico

Hall, H. See King.

Ibn-al-Baitar. See Leclerc.

Jaubert, H. F. Glossaire du centre de la France, Paris 1864.

Kersaint-Gilly, F. de Le Nama, in Bulletin du Comit6 d'Etudes his-

Lane, E. W. An Account of the Manners and Customs of the

D. Joannis Praefationes, Venetiis 1740.

Nasmith, J. Itinerarium Symonis Simeonis, et Hugonis Illum-

Oppenheim, M. Vom Mittelmeer zum Persischen GoU, vol. I,

Palmer, H. R. Notes on the Kororofawa and Jukon, in Journal

Quatremfere, M. Histoire dea sultans mamlouks, vol. I, parts I,

Raccolta di docimienti dalla R. Commissione Colombiana, part I, vol. I,

Kiley,H. T. See Bostock.

Sachau, E. Muhammedanisches Recht, in Lehrbucher des

V&mb&y, H. Etymologisches Worterbuch der turko-tatarischen

Wallis, C. B. The Advance of our West African Empire, London

Wherry, E. M. A Comprehensive Commentary on the Quran,

Zayas y Alfonso, A. Lexicograffa antillana, Habana 1914.

The Histoey op Copper and Iron.

(a) Sumerian urudu.

The Assyrian name for "copper" or "bronze" is

THE HISTORY OF COPPER AND IRON 3

enemies, lord," ir "to move, rise, elevate," er "to

"to whet" are all related to the Dravidian roots,

Lithuanian geleiis "iron" we have borrowings from a

So far the etymological history of the words discussed

(b) Assyrian parzillu.

In Sumerian bar means 1) "to divide, open, a half;"

parasu "to divide," parsigu "band," §upparuru "to

The Sumerian sil, which, as we see, led to the mean-

ing "iron" represented in a large number of languages.

form Qastram means "a cutting metal, iron," hence one

combination and originally refers to iron.

(c) Assyrian haginnu.

it," ilrr^ hasln "axe." The relation of the two words

hazina, Syriac hagind, points to a common borrowing.

(d) Arabic ^atr.

Vincent of Beauvais, quoting Avicenna, says: "There

glitter," hence J/ ^ ma'tur "a sword having in its