Reality, Symbolism, Time, and Space in Medieval World Maps Author(s): David Woodward Source: Annals of the Association

of American Geographers, Vol. 75, No. 4 (Dec., 1985), pp. 510-521 Published by: Taylor & Francis, Ltd. on behalf of the Association of American Geographers Stable URL: http://www.jstor.org/stable/2563109 . Accessed: 04/03/2011 13:17
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=taylorfrancis. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Taylor & Francis, Ltd. and Association of American Geographers are collaborating with JSTOR to digitize, preserve and extend access to Annals of the Association of American Geographers.

http://www.jstor.org

Reality,Symbolism, Time, and Space in Medieval World Maps

David Woodward Department ofGeography, University ofWisconsin, Madison, WI 53706
Abstract. Medievalmappaemundi carry levelsof meaning thathave been widely misunderstood. Theircompilers have beenjudgedon their to showgeographical ability reality structured according to a coordinate system, buttheprimary function of thesemapswas to provide illustrated histories or moralized, didactic displays in a geographical thinkers' of setting. Thatmedieval understanding thephysical worldhas also been underestimated is reflected in thefrequently repeated viewsthat mostmedieval scholars was flator that be shown at itscenter. thought theearth Jerusalem should Thispaperchallenges thesecommonly heldviewsinthelight ofrecent reinterpretations inarthistory andthehistory ofcartography. Severalthemes areexplored, including thetype ofreality represented by themaps,thewaythemapcenter as theMiddleAges developed, changed and therelationship between oftheearth's concepts sphericity andthegraphic constraints on themappaemundi. Finally, thestudy suggests waysin which we can learnfrom thisgenre of maps.We need to evaluate the of the MiddleAges on their achievements own terms and in the context of their purpose.More forcartography, showthat ofhistorical specifically themappaemundi mapsmayalso consist aggreor cumulative gations ofevents in addition to representing that inventories objects existcosynchronouslyin space. Key Words:medieval world ofcartography, mappaemundi, maps,MiddleAges,history mapfunctions, cosmographical concepts.

MEDIEVAL worldmaps,or mappaemundi such as longitudeand latitude.These nineand writers also oversimplified as they arefrequently a well- teenth-century called,form understandthinkers' medieval defined genreof mapsthathave received only underestimated inthe world; thisis reflected spasmodicattention from geographers. Some ingofthephysical repeatedviews thatmostmedieval inmanuscript 1,100 maps,mostly codicesofthe frequently Jerutheearth was flat or that thought eighth to thefifteenth centuries, stillsurvive. scholars Some of Theyareusually schematic inform, andfallinto salemshouldbe shownat its center. a tendency arosefrom several subcategories depending on their histor- thesemisunderstandings icalorigin andtheir graphic structure. Rootedin to regardthe cultureof the Middle Ages as gathwerethus Mappaemundi static. boththeHellenistic and Romantraditions, they essentially model, theT-in-O wereadaptedby theearlyleadersand scholars eredintoonlyone category, ofgeogingeneral histories reproduced oftheChristian Church. To theextent that they so often to this Also contributing andcartography. embody both scriptural and classical sources, raphy cartography of medieval their meaning reflects thechanging emphases of lack of understanding to thetechnical obliviousness was theapparent medieval thought.' andartists on scribes constraints The mappaemundi carrylevels of meaning andconceptual themedia, tools,and techniques, thathave been completely misunderstood. In oftheperiod: concepts to relatesuchstructural thenineteenth century, whensystematic studies and a failure andprojection to themapmaking of thesemapsfirst appeared,theywereinter- as perspective in light preted oftheviewthatmaps(to be true oftheperiod. of maps)wereintended to showgeographical realthe validity This paper seeks to examine in heldviewsofmappaemundi itystructured according to a coordinate system, thesecommonly
AnnalsoftheAssociation ofAmerican Geographers, 75(4), 1985,pp. 510-521 of American 1985 by Association X Copyright Geographers

510

Medieval World Maps

511

thelight ofrecent reinterpretations madein art history and thehistory of cartography. Its aim is to showthat theintention ofthecompilers of thesemapswas as much historical as geographical and thatthe resulting documents blended concepts ofboth andspace as a context time for understanding the Christian life.By examining t ~~A S I A the development of such conceptsas the flat earth with Jerusalem at itscenter, thepaperalso N S seeksto demonstrate that thesemapscannot be considered as a single category a thouspanning EUROPE AFRICA sandyears ofmedieval I suggest history. Finally, inthelight that, ofthesedocuments, ourmodern viewofmapsmayneedadjusting. It is nowfully acceptedthatmapsneed notnecessarily show onlyEuclideanspace. Perhaps we need also to considerthe idea thata map does not by its nature haveto represent a cosynchronous scene drawing 1: T-O map. Schematic butmaybe a many-layered cumulation of his- Figure1. Category of Noah. among the sons of divided world the tripartite as wellas objectsin geographical torical events thehomeof the Semiticpeoples, Asia represented space. theHamitic. andAfrica EuropetheJaphetic, Since mappaemundi are betterunderstood whentheir variation is recognized, it is conve- of Mallos (c. 168 B.c.) and, ultimately froma nient to describe four main categories: tripartite,Pythagoreanconcept. and transitional; zonal, quadripartite, typical and the Intermediatebetween the tripartite inFigures ofeachareillustrated examples 1_4.2 zonal categoriesof mappaemundiis a thirdcatThe tripartite map typeconsists of a diskrep- egory, the quadripartite, which contains maps resenting theinhabited which bearing the characteristics of each. Though world(0), within is a tripartite schema(T) oriented to the east these are not numerous, they are sufficiently with Asia taking up theupperhalf ofthecircle, distinctiveand influentialto warrantseparate Europethe lowerleftquarter, and Africa the treatment.Withintheir circular,oval, or rectlowerright quarter (Fig. 1). The partsof theT represent thethree major hydrographic features N known to dividethe threepartsof the earth: Tanais (the Don River) dividingEurope and RIGID Asia,theNile dividing Africa andAsia, andthe Mediterranean Sea dividing Europeand Africa. TEMPERATE The genre in a classicaltradition, is rooted and ZONE itis conjectured that theearliest tripartite maps accompanied manuscripts of theDe Bello JurTORRID of Gaius SallustiusCrispus(Sallust) gurthino (86-34 B.C.). The zonal category is characterized by oriZONE to the north entation or southand by therepresentation infiveclimatic oftheGreekclimata TEMPERATE zones thatfollow parallelsof latitude (Fig. 2). ZONE Itsprototype is derived from thecosmographical section (ch. 5-8 ofBook 2) inMacrobius's early FRGDZONE A.D. commentary on Cicero's fifth-century DreamofScipio(51 B.C.), which inturn derived S itscosmography from Eratosthenes (c. 275-194 Figure drawing 2. Category 2: Zonalmap.Schematic B.C.), Posidonius (c. 151-35 B.C.), Serapionof of the system from a of fiveclimatic zones,derived Antiochea (secondor first oftheancient Greek geographers. century B.C.), Crates model

512

Woodward
tion no different fromotherclasses of maps in the Middle Ages or any otherperiod. The survivingcorpus ofmedievalworldmaps have been seen therefore as a marked retrogression from an expected gradual improvement in the representationof the earth's featureson maps. Thus Charles Beazley, in his otherwise fundamentally useful work, was able to write: "The non-scientific maps of the laterMiddle Ages . . . are of . . . that a bare allusion such complete futility to the monstrositiesof Hereford and Ebstorf should suffice"(Beazley 1897-1906, 3:528). A comparison between two medieval maps made withina few decades of each otherlends superficial supportto Beazley's view. The mappamundi known as the HerefordMap (c. 1290) and the earliest known dated portolanchartby Petrus Vesconte (1311) both featurethe Mediterranean Sea prominently, but the positional accuracy of theVesconte map is clearlysuperior to that of the HerefordMap, as Figure 5 demonstrates.This assessment should be modified in several significant ways, however. The first the mappamundion its own involves studying terms,accordingto its intendedfunction and in the context of the scriptorium in which it was compiled.This view was forcefully expressedby John K. Wright,who developed the themeechoed by Kimble-that the lack of geometrical accuracy in the mappaemundidid not necessarwas not ily warrantcriticism,as this attribute E

4~~~~~~~~~~~~0. 4~~~~~~~0

Figure3. Category 3: Quadripartite map. Schematic ofthecombination ofthetripartite drawing world with thefourth part-inhabitedby Antipodeans-separated byan "ocean river."

angularshapes orientedto the east, there is an "ocean river" that divides the known tripartite worldfrom thefourth part,unknownon account of the sun's heat, but inhabitedby the Antipodeans (Fig. 3). The maps are believed to stem fromone lost eighth-century prototypeof Beatus of Liebana in his Commentaty on theApocalypse of St. Johnin whichhe stressedthe mandate of the Apostles to travelin all parts of the earthto preach the Gospel. The fourthcategory, which is transitional betweenthe medieval and Renaissance periods, reflectsthe profoundchange in mappaemundi that took place in the fourteenth and fifteenth centuries.These maps differ from fundamentally the zonal or tripartite models of the late Roman world,and belong in manyways to the spiritof the Renaissance, having as theirbasis the configuration of the MediterraneanSea commonly N foundin the portolanchartsand relying in some degree on the contemporary of explorecording ration,especially the Portuguesevoyages to the Atlantic islands and along the west coast of Africa(Fig. 4).

XS

Realism and Mappaemundi

4: Transitional Figure4. Category map. Schematic ofa fifteenth-century The outdrawing mappamundi. From the late-nineteenth centuryon, several linesoftheMediterranean Sea are basedon portolan authorshave viewed the mappaemundiprimar- charts, while therest ofthemapretains thetraditional ily as bearers of locational information, a func- circular frame.

Medieval WorldMaps
HEREFORD MAP ca. 1290 VESCONTE 1311

513

Figure5. Diagram Sea as shown on theHereford theeastern comparing partof theMediterranean Map (c. 1290)(left) and a portolan chart Vesconte dated1311(right). byPietro the primarygoal of this genre of map (Wright 1925,248; Kimble 1938, 181). We should not assume, however, that there was no interestin geographicallocation by the compilersof these maps but thattherewere certain prescribed constraintswithin which they had to work. The rare account of how to make a mappamundiby Hugh of St. Victor in his On the Mystical Noah's Ark provides an example of the kind of framework into which geographical information could be fitted: Theperfect arkis circumscribed with an oblong circle, whichtoucheseach of its corners,and the space which thecircumference includes represents theearth. In thisspace,a world mapis depicted in this fashion: thefront ofthearkfacestheeast,and therearfacesthewest.... In theapex to theeast formed between thecircleand thehead of theark is Paradise.... In theother apex, which juts out to thewest,is theLast Judgment with thechosen to theright, and thereprobates to theleft.In the northern cornerof thisapex is Hell, wherethe damned arethrown with theapostate spirits (Migne vol. 176,col. 700). 1844-64, this whole island would have been longer" (Vaughan 1968, 243). This does not mean that the importanceof current geographical content was always Map as an example, ignored.Using theHereford Crone has demonstrated that lists of placenames from writtenitinerarieswere incorpo-

~P
ha C I d e a

er

ia

:.a

up~p

The bounding shape of the mappamundi, whethercircular,oval, or rectangular, was thus Mdd. MS. n004, o. rrGalil predetermined by biblical or classical tradition. Into this space were arranged the features deemed significant forthe reader. As we might expect, the scale forthatpart of the map of pri~ egean maryinterestwas frequently larger: for exam0Se ple, the Jeromemap of Asia exaggerates Asia Minor to the point that it is almost as large as the representation of the rest of Asia (Fig. 6). Likewise, Matthew Paris's well-knownexplanationof distortion on his map of Britaindemonstratesthattherewas a conscious awareness from the 6. The Jerome mapofAsia. Adapted Figure thatscale could be conveniently adjusted within in the British tefhcentury manuscript Library, certainconstraints:"if the page had allowed it, Ad.M.10049, fol.64r.

514

Woodward

rated into themap.Though itssize(163 x 135cm) dimension of time (Pdcht 1962, 2; Hindman clearly precluded its use as a vade mecum for 1977, 38). equivactualjourneys(forwhichwritten itineraries, The mappaemundiare thecartographic pilgrim guides, and verbaldirections might well alent of narrativemedieval pictures. The thirauthorof the HerefordMap even haveserved), it is perhaps likely that theHere- teenth-century refersto his graphicworkas a "history" on the ford mapand theother largethirteenth-century wallmapscouldhavefulfilled a practical didac- map itself: "All who have this history. of ticpurpose indeveloping theconfidence orstim- This usage is stillsupportedby one definition ofan event as a "pictorialrepresentation ulating theimagination ofintended travelers, for history which recognizable content was desirable or series of incidents," supportedby the use of how- figures" (OxfordEnglish Dictionary, 1961 ed., For thispurpose of spiritual education, ever,morewas requiredof the mappamundi s.v. "historiated"). This themehas been develthan a modicum ofcurrent geography. The maps oped extensively by von den Brinckenwherethe neededalso to be imbued withtherichness of mappaemundi are seen as syncreticpictorial theChristian historical tradition. It is thusalso chroniclesparallelto the textualchroniclesfrom important torealizethat themappaemundi were St. Jerome to Hartmann Schedel (von den Schulz has applied not snapshotsof the world's geography at a Brincken1968, 118). Juergen given point intime, buta blending ofhistory and this idea of the didactic moralized map to art a projection on cityviews in the geography, ofhistorical events on a history, focusing particularly geographical framework. This pointwas made late-medievalperiod and the Renaissance. by Bevan and Phillott (1873) in a preliminary historians are agreedthatthemain Cartographic way,buta fullillustration of thistheme was to of of thesemaps was therepresentation function awaitthecareful studies of G. R. Crone(1954, rather thanthe and history, religiousmysteries 1965), againwith reference totheHereford Map. facts. ofprecise Theynever geographical recording to explain thisaspectofthemediCrone demonstrated with minute documentation pause,however, totheir main is irrelevant concern, that theHereford Map was a complex blending evalmap,which For thehisthe of knowledge. geographical growth of Greek,Roman,and Christian sources and this didactic ofart, itis precisely torian bycontrast, and tracedits pedigree back to thelate fourth thatis of its rangein timeand content, tradition, early fifth centuries A.D. As theinterest in mapinterest 446). (Schulz1978, primary has grown in recent paemundi years-aided in The sources of historical and geographical partby thepublication of a catalogue of these documents (Destombes1964)-several histori- informationon the mappaemundi were both their ansandarthistorians haveturned attention classical and biblical-the commemorationof to themaps,particularly in the context of the famous events and places being sometimes ratherthan in inseparable,withthe Old Testament textandgraphic relationship between images represented.Though medievalwriting and painting (Ruberg1980, the New more frequently assumethattheborder early Judaismemphasized the importanceof an 551-92). Mostpainters oftheir workenclosesa cosynchronous space, event's location, the earliestChristiansshowed in which butnarrative is illus- little interest-with some important exceppainting, a story trated by showing severalstagesof a narrative tions-in the exact location of even theirmost side by side within the sameframe, has a long sacred events(North1979,76). The teachingsof In themonumental history. artofancient Egypt Christ emphasized the spiritual and not the sto- physical world; forexample, in response to the and Mesopotamia, forexample, sequential ries of battles and othernotable events are question of whetherto build a shrineat Gerizim within recorded a singleframe (Groenewegen- or Jerusalem, Christ's answer was to be less Frankfort 1951).In medieval popularnarrative concernedwiththe location thanwiththe motisev- vation of the proposed action (John4, 19-24). ofbiblical illustrations, particularly stories, in There are, moreover,few allusions to global or eraleventsseparated by timeare portrayed thesame scene,notin sequenceas in a frieze cosmological matters:the words sphere,globe, or cartoon, butplacedin their appropriate posi- or hemisphere in the geographical sense are thus nowherefoundin its pages.3 tionsin the scene. The chiefcharacters The early leaders of the Church, in reaction reappearin a staticlandscapeto expressthe
(Crone 1965). theword "historiated"to mean "decorated with

Medieval World Maps

515

is also known as a manwerealso anxious shapeoftheaura,which to theclassicalphilosophers, dorla(Italian for almond), an extremely common of information to pointout thattheknowledge in Christian impor- symbol iconography from thefifth secondary was ofstrictly abouttheearth should be on century on, is also reflected in theshapeofcerwhosemind tanceto theChristian, themselves, suchas a numspiritual plane.Thus,in characterizingtainmappaemundi a higher maps by the was able to say, ber of mid-fourteenth-century St. Augustine a truebeliever, RanulfHigdenor the map man .. . although he knowsnotthe Englishchronicler "a faithful than knownas the Genoese map of 1457now precirclesof the GreatBear, is muchbetter in theBiblioteca and served Nazionalein Florence. who can weighout the elements another numberthe stars and measure the skies" Thiscommon symbolic shapethus maystand for theworld as Christ's domain. 1965,5:4). (Augustine With all theseconsiderations, can The functions of medieval mappaemundi itis inadvisable plane to comparethe geographical as beingon a different accuracyof the thusbe regarded charts or estatemaps mappaemundi withthatof contemporary from thoseoftheportolan or than loca- latermaptypes, suchas theportolan charts rather or ofthesameperiod. As teaching sym- thegreat wallmapsoftheRenaissance. The histionaldevices, theyreliedon mystical, ofcartography, likethehistory of science, imagery to a remarkable tory bolic,and allegorical history of the Christian is moving away from beingprimarily a search extent.The spiritual to theLast Judgment,forprecursors its Creation world, from to an attempt to understand the plannedeventsin developments witha sequenceof divinely in variousperiodson their own byJesusChrist, terms.For mappaemundi suchas theSalvation thismeansstudying between, portrayed-in moreor less themnotprimarily are all carefully as repositories of thencurrent detail-on the mappaemundi. geographical knowledge (although a modiThe threefold division of theworldin theT- cum of such information be may sometimes in obtained of theearth thepeopling O mapsrepresents from nowhere else) but as illustrated by each of Noah's sons. histories the threecontinents or moralized,didacticdisplaysin a the geographical The families ofShem(theeldestson having setting. share-or Asia-in the tripartite maximum and Japheth (Europe) scheme),Ham (Africa), or listed on thesemaps.This The Map Center depicted are often is thesourceforthenaming traditional division of peoplesin theOld of thethree The narrative great groups historical character ofthemapandJaphetic. theSemitic, is also revealed inthechoiceofa cenWorld: Hamitic, paemundi is also symbolized It is commonly by tralfeature. The Passionof Christ assumedthatthe that theT majority theT-O maps.Lanmanhas suggested ofmedieval world mapswerecentered a cruxcommissa on Jerusalem, in theT-O schemarepresents the biblicalpassage: following is lentcre- "This cityof Jerusalem I have set amongthe or tau cross, and thisobservation whenthe ends of thecross- nations, withthe othercountries roundabout denceparticularly at an angle,as in Figure7 her" (Ezek. 5,5).4 Therewas also, as we have piece are trimmed inthe as in the Ebstorf seen,a definite revealed spatial sensitivity (Lanman 1981). Sometimes, the location of mappamundi Old Testament thatrendered map-a hugethirteenth-century in WorldWar II-the eventsin Jewish connected history intimately unfortunately destroyed on the with themselves is literally theevents superimposed (Cohn1981, 2). bodyof Christ We wouldexpect themappaemundi to reflect map of theworld.His head is at theeast,feet in the thisconcept, that three ofthebest to gather anditis true inthewest,andarmsoutspread and thesouthin a powerful gesture sym- knownand most oftenillustratedmappaenorth oftheworld. mundi-the Hereford,Ebstorf,and Psalter thesalvation bolizing centered on Jerusalem. the maps-are all precisely The third mainstageofChristian history, centhetime ofthese either is represented maps(thethirteenth Last Judgment, bythefig- From cenat tury) of thefifteenth to aboutthe middle in Gloryor of God theFather ure of Christ feature was typical. this itcan be saidthat intheeasterly loca- tury, thetopofthemappaemundi is byno meansrepperiod tionof Paradise.An aura of lightsymbolizing Butthistwo-century MiddleAges; it is but of theentire holiness may also surroundthe figure.The resentative

516

Woodward

:::::
ViSS >..ft: "X'0: A;ff;C~ t0 0 :XiffX0ii ; :fffti

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

N: X

'

t000Sw?t

00fE00X

;t

fXE0

:0V' t'2~~~~~~~~~.77
KM&. o: 4i,0' W" 0s

Sallust aneleventh-century From crossbar. a truncated manuscript 7. A T-Omappiamundi demonstrating Figure Plut.64.18,fol.63v. intheBiblioteca MediceaLaurenziana, Florence,

or (such as thosebased on thefourthphases in themedieval century one of severaldifferent Historia Adversum Paganos of can be identified fifth-century of whichat leastfour period, Paulus Orosius)the mapsdo notappearto be forthe mappaemundi.5 place. The eighthon any particular maps centered Nor do the threethirteenth-century "AngloAlbimapandthetenth-century above-which fallintothetripartitecentury mentioned cat- Saxon" map are examples.Otherinfluential theonlymajor category-constitute T-in-O thanJerusaoh places other repeated mapsare centered frequently an oversimplification egory, the Greektradition, an ancient categories lem. Following ofthefour In thefirst intheliterature. of Mainz(c. 1110)is cenof thispaper-the worldmapof Henry at thebeginning introduced as the teredon theCyclades,theislandssurrounding assumedthat, is commonly tripartite-it of theMediterraneanthesacredisle ofDelos. themeeting T represents a majorgroup The zonal maps,whichform axis, the withthe Don-BlackSea-Aegean-Nile centered areclearly ofover300maps, Holy Land is nearenoughto thatintersectionconsisting In thelater inAfrica. somewhere to be at the centerof the map. ontheequator, forJerusalem townof Aryn,the of zonal maps, the mythical examples notonlymany Thereare,however, on thecentral lying of theearth of Islamiccenter ofthestemand thecrossbar theintersection is promworld, theinhabited bisecting butalso ofthe meridian farabove thecenter, theT being notin the is clearly Jerusalem marked. thisintersection. inently farfrom of Jerusalem placing the maps in the intermeFurthermore, mapsJeru- center. oftheseschematic In most examples of theBeatus largely consisting diate category, in the non-scheat all, and is not marked salem noton anyidentificentered thethirteenthmaps,are usually before mapsfrom tripartite matic

Medieval World Maps

517

ableplace butsomewhere in theMediterraneanof their viewshas thusprobably been exaggerSea. atedon account oftheir An controversial nature. It was onlyafter thestrengthening oftheidea excellent example of thisis found in theundue of Jerusalem as the spiritual a natural spaceingeneral center, historical texts given to Cosmas outcome oftheCrusades, that a noticeable shift Indicopleustes' sixth-century concept of a flat, occurred toward themapson Jerusa- rectangular, centering four-cornered earth witha vaulted lem.This thenbecomesa characteristic of the heaven (Jones 1934, 305).Thisis nowpreserved fourth thetransitional, category, from theendof only in two manuscripts and was virtually thethirteenth century to aboutthemid-fifteenth ignored by medievalcommentators, withthe when new discoveriesextendedthe exceptionof Photiusof Constantinople, century, who margins oftheworld mapandthecenter moved statednotonlythat"the styleis poorand the accordingly. By then,AndreaBianco had to arrangement hardlyup to the ordinary stanstateexplicitly whyhe had broken with thetra- dard" butthat"he mayfairly be regarded as a dition ofcentering mapson Jerusalem, andeven fabulist rather thana trustworthy authority" tried to explainit away on his map of 1436on (Photius 1920, 36:31-32).Nevertheless, thelight whatmustbe one of the earliest references to of Cosmas's fameis keptburning by suchhisthe conceptof a center of population density: torians whocontinue that to write he "had great "Jerusalem is indeed thecenter oftheinhabited popularity amongeven the educated till the worldlatitudinally, though longitudinally it is twelfth century" (Randall1926, 23). somewhat to the west, but since the western The views of the common populaceon the is morethickly portion populated by reasonof issue are of coursebarelyrecorded. There is Europe,therefore Jerusalem is also the center someevidence from thecosmographical content longitudinally ifwe regard notempty space but of the vernacular epics and romancesof the thedensity ofpopulation" (Bianco 1436). twelfth andthirteenth centuries designed for the everyday person that many thought oftheearth as a disk(Tattersall 1981, 46). Buttheinterpreof theword"round" in thesepoemsas tation The Flat Earth either circular or spherical is fraught withconandthesimiles fusion, of"apple" and "ball"The question oftheextent oftheunderstand- frequently usedintheseworks-wouldseemto ingoftheearth'ssphericity in theMiddleAges lend morecredence to a common view of the is confounded by severalfactors, summarized earth'ssphericity, evenbefore 1300. by several recent authors (Randles1980;Tatter- The shapeoftheearth seemshowever tohave sall 1981).As we have discussed, themedieval beenofless interest tothemajority oftheclergy, consists period ofseveral different sub- letalonethegeneral entirely populace, than theconcept anditis unwise periods to assumethat theviews of human lifeexisting on theotherside of the of a few individualscan be extendedto the world. The ninth-century confrontation between periodas a whole.Even in theearliest ofSalzburg andPopeZacharias was about period, Virgil however, despite thevariousdifficulties of bib- theheretical doctrine of theAntipodes and not lical interpretation, mostearlyChurch leaders aboutthesphericity of theearth (Betten1924). held to the classical conceptof the spherical It was possibleto believein a spherical earth and Augustine earth, mentions it at without specifically whatwas on theother side of knowing least twice (Augustine 1965,5:51). The most it. The Antipodes concept,withthe apparent popularof the late Roman secular writers absurdityof people standingupside down, Pliny, and Martianus Macrobius, Capella-also causedan embarrassment for theChurch, as the made unequivocalreferences to its spherical baptismal statusof a race of peoplewho were shape. It was perhaps in overreaction to these not descended from the sons of Adamwas in worksthat Severianus and Lactantius,who question. Evenwellinto theAgeofDiscoveries, denied onprinciple that anything "pagan" could books werebeingwritten the doctrine, against be accurate, included thespherical earth intheir suchas theContra ofZacharia Antipodes Lilio, sweeping of the pagan writings a monkin the church condemnation of St. JohnLateranin (Jones1934).But it muststillbe said thattheir Rome(Randles1980,31). writings werenotto have the influence of the Whatever the confusion about the general greater Church Fathers andthat theimportance medieval viewsoftheearth's is no shape,there

518

Woodward

form of reasonto believethatthe best-known the T-O map, withits tripartite mappamundi, divisionof the inhabitedworld and the surfacieevidence was prima rounding ocean river, medieval beliefin a flatearth.It foruniversal solid a three-dimensional was noteasyto depict Medievalconmedium. on a two-dimensional were not and perspective cepts of projection In thebroadest sense,itis poswelldeveloped. from a sible to arguethatany transformation projecsphereto a plane involvesan implied tion.6Thus, even the simpleMacrobiandiagrams (forexamplein Fig. 2), whichshowthe lineson a cirparallel climata as zonesbetween couldbe said cle representing thehemisphere, to be on a projectionthat approximatesan was aspect);theauthor orthographic (equatorial quite aware thatthese climata were circular edition From a printed five circles. 8. Isidore's as Figure zones on the globe,yettheyare portrayed Kimble from 1472.Reproduced theEtymologiae, straight parallellines on the map. Despitethe of (1938),opp. p. 36. philosoofthethirteenth-century achievements pherRogerBacon, whohad a clearidea of the system lightand uses the word globus to express this coordinate value of usinga systematic the positionsof (Fontaine 1960, 333). and inventory to transform On the other hand, several passages have a systematic projecplaces and who described used to indicate that Isidore thoughtthe been in tion,therewas no generalunderstanding a world was flat. For example, he describes the Christian Europe of the way to represent surface on a plane untilthatknowl- earthas a "circle of lands" (orbis) like a wheel spherical of Claudius (Migne 1850,82:495). Isidore also seems to have edge-preservedin theGeography at confused the Greek zonal concept, which he translation Ptolemy-was revealedthrough gleaned from the Poeticon Astronomicon of century. thebeginning ofthefifteenth can perhapsbest be illus- Hyginus(readingitin Latin), and misunderstood The controversy thezones ofthewritings from theexample (partic- its mainpoint:thatthelines separating trated of Isidoreof Seville,one of the most Reriun) of the MiddleAges. influential encyclopedists of theuniverse Isidoreacceptedthe sphericity inhiswording on that andis quiteexplicit question(Brehaut 1912).Isidorealso uses theword
globus several times in the De Natura Rerurn ularly the Etymnologiae and the De Natura

indescribing themoonortheplanets butalways 1960, 223,231,239,277). Buthe cau(Fontaine of on the sphericity tiously avoidedcomments exceptin a passage wherehe the earthitself of the theconfines theocean bathing describes ofthe 1960, 325).His awareness globe(Fontaine is further indicated ofa spherical earth concept
in the Epistula Sisebuti, an astronomicalpoem

to Isidore in theform by Siseofa letter written theDe Natura for whom but,KingoftheGoths, hadbeenwritten Rerurn 1960,151).In (Fontaine aboutthe an eclipse,Sisebutwrites explaining terrestrial globebeingan obstacleto the sun's

were only circles when they were drawn on a globe. When applying the five zones to his they emerged as five circles world, therefore, placed on a flat disk as in Figure 8 (Fontaine 1960, 208-11). This could hardly be taken as evidence of Isidore's beliefin a flatearth,howto grasp the geomever, but ratherhis inability in the Greek climateconcept. etryinherent whereas Isidore someIn summary, therefore, times appears confused about the shape of the earth,the weightof the evidence tends toward When viewed in his acceptance of its sphericity. the general context of his time, in which most Christianand secular writers of the influential had also propoundedthisview, it is not surprising thatthis was the case. Later Christianwritersnot only specifically of the earth, but proaccepted the sphericity vided explicit reasons for theirview. Thus the Venerable Bede (672/3-735) explained that the

Medieval World Maps

519

appearance in cause of theunequallength of the days lay in world.Nor can thedisk-shaped theglobular ofthesemapsbe taken literally to reflect shapeof theearth, it as many describing notonly"roundlikea shield, ina flat earth. As we haveseen, butalso in every a general belief direction, not the prevailing opinion like a playground ball" (Bede 1843- this was probably 44, 210). amongthose (admittedly not a greatnumber) In thelaterMiddleAges,themostinfluentialwhocaredabouttheissue. the spreadof time commentators agreedthattheearth was spher- Second, by exaggerating ical. The writings of Aristotle, withhis elegant depicted within their themappaemundi borders, three-part proofof the sphericity that of the earth, also demonstrate mapsin general neednot or the astronomical and and geographical onlyspatialrealities workof be seen as reflecting Ptolemy, forwhichthe conceptwas essential, perceptions or pictures of theearthstopped at hadfound their wayintotheWestas thechan- a givenmoment in time. Theymayalso consist nels of classical and Arabiclearning became of historical or cumulative invenaggregations openedafter the twelfth In the four- toriesof eventsthatoccur in space. For the century. teenth andfifteenth a curious centuries-despite themyth mappaemundi, thismeant melange of still in someschoolhistory perpetuated texts of bothChristian and secularlegendary history. of placingJerusalem at the Columbus,the commonnavigator, valiantly Even the practice defending the idea of the globe before the center of the mappaemundiovercame the learned clergy7-there wouldhaveremained lit- weight of theclassicalseculartradition onlyat tle doubtin the mindof the medievalscholar thevery endofthemedieval period. The logical in Gautier that, de Metz's words, "a mancould inseparability of history and geography is thus theworldas a flymakesthetourof vividly illustrated by the mappaemundi. go around an apple" (Caxton1913,52). in which Because theyprovide a dual context The limitationsof the two-dimensional human bothtime and space have events unfold, medium of themappaemundi and thelack of a been of prime interest to thephilosopher. Berclear conceptof projective geometry and per- trand Russell calledthestudy unified conofthis spective appearto havecontributed to thebelief text "chronogeography" (Russell 1927,283), thatthey represented a flat, disk-shaped earth. and there seemsto be a growing awareness of But since the overridingpurpose of these the importance of integrating bothgeography maps-as narrative histories-wasnotto con- andhistory as providing a unified context for our veyfactsaboutthesize and shapeof theearth understanding of the processesthatshape the where except they boredirectly on theChristian world. mission, theycan hardly be criticized fornot thishas significant For cartography, implicareflecting thephilosophical thought ofthetime. tions.No map represents features on its face the same time.Even in observedat precisely modern topographic maps,thedatesof compilationof its severalpartsmayvaryby several Conclusion themodern years.Nevertheless, mapoften prothatthe objectsin the landWhatcan we learnfrom theoft- vides the illusion challenging repeated medieval scape it portrays are cosynchronous. Thereis misconceptions concerning the perhaps a case to be madefora modification of world maps?In thefirst place,itunderlines to thisidea, of providing a new kindof reference need,nowrecognized bymedieval historians, ofhistorievaluatethe achievements thelayers explicitly of the MiddleAges mapthatreveals on their thathave and in thecontext of their cal events,processes,and artifacts own terms was shaped the present The function landscape,as well as the ofthemappaemundi purpose. it. This of course a visualnarrative ofChris- objectsthatnowexistwithin to provide primarily tianhistory cast in a geographical framework,has been attemptedin historicalmaps and not to communicate geographicalor cosmo- atlaseswithvarying degreesof success,but it to an entirely has not yet been extendedsystematically facts.Theythusrepresent graphical reference different large-scale maps.It is notthe cartographictraditionand should modern that purposeof thispaper to explorethe ways in on the grounds therefore not be ridiculed of geography and history to an ever- whichthisblending theyappeared as retrogressions But it literalgeographical improving cartographically. pictureof the could be accomplished

520

Woodward
6. This issue was discussed by Tobler(1966). 7. This mythwas greatlyexpanded by Washington Irvingin his biography of Columbus. See Morison (1942, 1:117).

may be fruitful to expand the agenda of cartographersto studyways in which the large-scale reference map can represent thehistoricalmeaningof the landscape. Maps will thencome to be seen as artifactsthat can portraythe location not only of objects and condiand distribution tions but also of events and processes.

References
Arentzen,Jorg-Geerd. 1984.Imago mundicartographica. Studien zur Bildlichkeit mittelalterlicher Welt- und Okumenekarten unter besonderer von Text Beracksichtigung des Zusammenwirkens 53. und Bild. Munstersche Mittelalter-Schriften Munich: WilhelmFink Verlag. Augustine, St. 1965. De civitatedei of St. Augustine, 7 vols., trans.Eva MatthewsSanfordand William McAllen Green. Cambridge,Mass.: Harvard UniversityPress. Beazley, C. Raymond. 1897-1906. The dawn of modof explorationand geoern geography:A history graphical science from the conversion of the Roman Empire to A.D. 900. 3 vols. London: J. Murray. Bede, the Venerable. 1943-44. The complete works of VenerableBede. Vol. 6, De temporum ratione, trans. JohnAllen Giles. 12 vols. London: Whittakerand Co. Betten, Francis J. 1924. Knowledge of the sphericity the earlierMiddle Ages. Cathof the earthduring olic Historical Review 3:74-90. Bevan, W. L., and Phillott, H. W. 1873. Medieval of the Heregeography:An essay in illustration ford mappa mundi.London: E. Stanford. Bianco, Andrea. 1436. MS. Fondo Ant. It. Z76. Biblioteca Marciana, Venice, Italy. Brehaut, Ernest. 1912. An encyclopedistof the Dark Ages: Isidore of Seville. New York: Columbia Press. University Caxton, William. 1913. Mirrourof the world. Early English Text Society, Extra Series 110. London: and Co. Kegan, Paul, Trench,Trubner Cohn, Robert L. 1981. The shape of sacred space: Four biblical studies. Chico, Calif.: Scholars Press. Crone, Gerald R. 1954. The worldmap byRichard of Haldingham in Hereford Cathedral. Reproductionsof Early Manuscript Maps 3. London: Royal GeographicalSociety. . 1965. New lighton the Herefordmap. Geographical Journal 131:447-62. Destombes, Marcel, ed. 1964. Mappemondes A.D. 1200-1500: catalogue preparepar la Commission des Cartes Anciennes de l Union Geographique Internationale.Amsterdam:N. Israel. de Fontaine, Jacques. 1960. Isidore de Seville, TraitW la Nature. Bordeaux: Ecole des hautes etudes hispanique. Groenewegen-Frankfort,Henrietta Antonia. 1951. Arrest and movement: An essay on space and time in the representational art of the ancient of Chicago Press. Near East. Chicago: University in biblical lands. Har-El, Menashe. 1981. Orientation Biblical Archeologist44 (1): 19-20. Harley, J. B., and Woodward, David, eds. ForthVol. 1, Carcoming. The history of cartography.

Acknowledgments
I should liketo thank AnneGodlewska, Brian Harley, Robert Sack,andYi-FuTuanfor their helpful reading oftheearlier drafts ofthemanuscript. Thanks arealso due to thestaff of the University of Wisconsin Cartography Laboratory, especially Onno Brouwer, for preparing themaps.

Notes
1. This paper reports preliminaryresearch for a chapteron medieval mappaemundi prepared for a larger work edited by Harley and Woodward Other issues involving the map(forthcoming). and fullreferences will be foundin that paemnundi work. For recentlypublishedgeneral studies, see Ruberg (1980) and Arentzen(1984). 2. There have been several attemptsat classifying mappaemundi, as in the standard catalogue of medievalworldmaps editedby Destombes (1964). The classification proposed here is explained furtherin Harley and Woodward (forthcoming). 3. The phrasein Psalm 83,11, "surface of theglobe," is now regardedto have no geographical significance (Mahood 1968, 275, n.11). In addition,the frequentreferencesto "the round world" in the original (sixteenth-century)Book of Common Prayer, as in Psalm 89:12, 96:10, 98:8 (Pss. 88, 95, and 97 respectively in theBible) are an expresfromthe sion of circularity ratherthan sphericity, Latin of the Vulgate orbis terrae. The only specificmention of a "map" (or at least a townview) that I have been able to find in the Bible is in Ezekiel 4:1: "Man, take a tile [Vulgate: lateremn] and set it beforeyou. Draw a city on it, the city of Jerusalem"(New EnglishBible). It is also possible, as Menashe Har-El believes, that maps were in use for the extensive survey (or "register" in the New English Bible) dividingthe tribes of Israel, foundin Joshua 13-19, especially 18:5 (Har-El 1981, 19-20). 4. This oversimplification is also repeated in some general books on the historyof geography,most recently in Holt-Jensen (1980, 11) where it is statedthatin the Middle Ages "the worldbecame a flatdisk withJerusalemat its center." 5. Four stages are proposed in Harley and Woodward(forthcoming). Macrobius to Isidore: thelate Greco-Romanand PatristicPeriod, (c. 400-700); Bede to Lambert of St. Omer (c. 700-1100); Henry of Mainz to the Herefordmap (c. 11001300); and Pietro Vesconte to Fra Mauro, (c. 1300-1460).

Medieval World Maps

tographyin prehistoric, ancient, and medieval Europe and theMediterranean.Chicago: Universityof Chicago Press. Hindman, Sandra. 1977. Textand image infifteenthcenturyillustratedDutch Bibles. Leiden: E. J. Brill. Holt-Jensen, A. 1980. Geography:Its history and concepts. London: Harper and Row. Jones, Charles W. 1934. The flat earth. Thought 9:296-307. Kimble, George H. T. 1938. Geographyin theMiddle Ages. New York: Russell and Russell. Lanman, JonathanT. 1981. The religioussymbolism of the T in T-O Maps. Cartographica 18:18-22. Mahood, Mitchell. 1968. Psalms II 51-100. New York: Doubleday. Migne, Jacques Paul. 1844-64. Patrologia latina. 221 vols. Paris. Morison, Samuel Eliot. 1942. Admiral of the Ocean Sea: A lifeof Christopher Columbus. Boston: Little, Brown and Co. North, Robert. 1979. A history of biblical map making. Beiheftezum TtibingerAtlas des Vorderen Orients,ser. B, no. 3. Wiesbaden: Ludwig Reichert Verlag. Pacht, Otto. 1962. The rise of pictorial narrativein twelfth-century England. Oxford: Clarendon Press. Photius of Constantinople. 1920. The libraryof Photius, trans. J. H. Freese. London: Macmillan. Randall, John Herman, Jr. 1926. The makingof the modernmind: A surveyof the intellectualbackground of the present age. Boston: Houghton Mifflin Co.

521

Randles, W. G. L. 1980. De la terreplate au globe terrestre:Une mutationepistemologique rapide (1480-1520). Cahiers des Annales 38. Paris: LibrairieArmandColin. Russell, Bertrand. 1927. Philosophy. New York: W. W. Norton. Ruberg, Uwe. 1980. Mappae Mundi des Mittelalters in Zusammenwirken von Text und Bild. In Text und Bild: Aspekte des Zusammenwirkens zweier Kunste in Mittelalter und frtiherNeuzeit, ed. Christel Meier and Uwe Ruberg. Wiesbaden: Ludwig ReichertVerlag. Schulz, Juergen. 1978. Jacopo de' Barbari's view of Venice: Map making,city views, and moralized geography before the year 1500. Art Bulletin 60:425-74. Tattersall, Jill. 1981. Sphere or disc? Allusions to the shape of the earth in some twelfth-century and thirteenth-century vernacular French works. Modern Language Review 76:31-46. Tobler, W. R. 1966. Medieval distortions:The projections of ancient maps. Annals of the Association of American Geographers56:351-60. Vaughan, Richard. 1968. MatthewParis. Cambridge: Press. CambridgeUniversity von den Brincken, Anna-Dorothee. 1968. Mappa mundiund Chronographia. Deutsches Archivfur die Etforschung des Mittelalters 24:118-86. Wright,John Kirtland. 1925. The geographical lore of the timeof the Crusades: A studyin the history of medieval science and tradition in western Europe. New York: AmericanGeographicalSociety.