George Kubler - The Shape of Time

F
GEORGE KUBLER
THE SHAPE OF TIME
REMARKS
ON
THE
HISTORY
OF
THINGS
NEW HAVEN AND LONDON: YALE UNIVERSITY PRESS
N 66 .K8 c.?
'Kubler, George, 1912:The shape of time
The title-page and chapter-head illustrations

show pottery figurine fragments made in the
Valley of Mexico between 300 B.c. and about
A.D. 900. All are from Teotihuad.n.and its
vicinity. (Courtesy American Museum of Natural
History.)
Originally published with assistance from the foundation

established in memory of Rutherford Trowbridge.
Copyright 1962 by Yale University

Sixth printing, 1970
Set in Aldine Bembo type and
printed in the United States of America by

The Carl Purington Rollins Printing-Office of
the Yale University Press, New Haven, Connecticut.
All rights reserved. This book may not be
reproduced, in whole or in part, in any form
(except by reviewers for the public press),

without written permission from the publishers.
Library of Congress catalog card number: 62-82 50
Distributed in Great Britain, Europe, Asia,
and Africa by Yale University Press Ltd.,
London; in Canada by MeGill-Queen's University
Press, Montreal; and in Mexico by Centro
lnteramericano de Libros Academicos,
Mexico City.
''"''J ~~ULU:GE LIBHAR't

~01\il.Ml\l, OREGOI't
TO MARTIN HEINEMANN
Preamble
SYMBOL,
FoRM,
AND DunA'fiON
Cassircr' s partial defmition of art as symbolic language has

dominated art studies in our century. A new history of culture
anchored upon the work of art as a symbolic expression thus came
into being. By these means art has been n1ade to com1ect with
the rest of history.

Bnt the price has been high, for while studies of meaning received all our attention, another dcfmition of art, as a system of
formal relations, thereby suffered neglect. This other defmition
matters more than meaning. In the same sense speech matters
more than writing, because speech precedes wnting, and because
writing is but a special case of speech.
The other defmition of art as form remains unfashionable,
although every thinking person will accept it as a truism that no
meaning can be conveyed without form. Every meaning requires a support, or a vehicle, or a holder. These are the bearers
of meaning, and without them no meaning would cross from
me to you, or from you to me, or indeed from any part of nature
to any other part.
!'he forms of communication arc easily separable from any
meaningful transmission. In linguistics the forms are speech
sounds (phonemes) and grammatical units (morphemes). In
music they are notes and intervals; in architecture and sculpture
they are solids and voids; in painting they are tones and areas.
The structural forms can be sensed independently of meaning.
We know from linguistics in particular that the structural cle-
viii
PREAMBLE
ments undergo more or less regular evolutions in time without

relation to meaning, as when certain phonetic shifts in the history
of cognate languages can be explained only by a hypothesis of
regular change. Thus phoneme a, occurring in an early stage of
a language, becomes phoneme b at a later stage, independently
of meaning, and only under the rules governing the phonetic
structure of the language. The regularity of these changes is such
that the phonemic changes can even be used to measure durations
between recorded but undated examples of speech.
Similar regularities probably govern the formal infrastructure
of every art. Whenever symbolic clusters appear, however, we
see interferences that may disrupt the regular evolution of the
formal systetn, An interference fr01n visual images is present in
almost all art. Even architecture, which is commonly thought to
lack figural intention, is guided from one utterance to the next
by the images of the admired buildings of the past, both far and
near in time.
The purpose of these pages is to draw attention to some of the
morphological problems of duration in series and sequence. These
problems arise independently of meaning and image. They are
problems that have gone unworked for more than forty years,
since the time when students turned away from "mere formalism" to the historical reconstruction of symbolic cmnplexes.
The main framework of these ideas was set down at Gaylord

Farm in Wallingford during November and December in 1959.
I am grateful to my family and friends, and to the staff at Gaylord,
and to my associates at Yale University, for their many thoughtful attentions to the demands of a restless patient. I wrote most
of the text early in 1960 in Naples, and the finished manuscript
was submitted to Yale University Press in November of that
year. For their perceptive readings and valuable suggestions on
its improvement I am indebted to my colleagues at Yale, Professors Charles Seymour, Jr., George H. Hamilton, Sumner McK.
Crosby, G. E. Hutchinson, Margaret Collier, George Hersey,
SYMBOL, FORM, AND DURATION
ix
and to Professor James Ackerman of Harvard, whom I taught

twenty years ago at Yale. For aid toward publication I am grateful to the Mature Scholars Fund of Yale University.
G. K.
New Haven
15 May 1961
r
!
Contents
PREAMBLE
Symbol, Fvrm, and Duration

1.
Tnn Hrs'roRY OF
THINGS
vii
r
The Limitations of Biography: Individual emrances. Talent

and genius. Biological and physical metaphors. Scientists and artists
The Historian's Commitment: The divisions of the arts

The Nature of Actuality: Of arts and stars. Signals. Relays
Self-signals and Adherent Signals: Iconographic studies. Configurational analysis. The taxonomy of meaning
2. THE CLASSING OF THINGS
Formal Sequences: Linked solutions. Open and dosed sequences. Fashions

Prime O~iects and Replications: Mutants. Diagnostic difficul-
5
12
r6
24
31
33
ties. Serial appreciation. Technical renewals. The invisi-
ble chain. Solitary and gregarious artists

Serial Position 1 Age, and Change: The rule of series. Systematic age. A Mexican paradigm. Linguistic change
3 THE PROPAGATION OF THINGS
39
53
62
Invention and Variation: Artistic invention. Convention and

invention
63
Replication: Permanence and change. The anatomy of routine. Historical drift
71
Discard and Retention: Obsolescence and ritual. Aesthetic

fatigue
77
CONTENTS
4. SoME KINDS OF DuRATION
xii
83
Fast and Slow Happening: The typology of artists' lives.

Tribes, courts, and cities
84
The Shapes of Time: Positional values. Periods and their

lengths. The indiction as module. Intermittent classes.
Arrested classes. Extended series. Wandering series.
Simultaneous series, Lenses vs. f1bers of duration
CONCLUSION
INDEX
1.
The History of Things
96
123
Finite Invention: The purist reduction of knowledge. Widening the gate. The fmite world
The Equivalence of Form and Expression: Iconological di-
123
minutions. The deficiencies of style. The plural present
127
131
Let us suppose that the idea of art can be expanded to embrace the whole range of man-made things, including all tools
and writing in addition to the useless, beautiful, and poetic things
of the world. By this view the universe of man-made things
simply coincides with the history of art. It then becomes an urgent
requirement to devise better ways of considering everything men
have made. Tins we may achieve sooner by proceeding from art
rather than from use, for if we depart from use alone, all useless
things are overlooked, but if we take the desirableness of things
as our point of departure, then useful objects are properly seen
as things we value more or less dearly.
In effect, the only tokens of history continually available to
our senses are the desirable things made by men. Of course, to
say that man-made tl:Ungs are desirable is redundant, because
man's native inertia is overcome only by desire, and nothing
gets made tmless it is desirable.
Such things mark the passage of time with far greater accuracy
than we know, and they fill time with shapes of a limited variety.
Like crustaceans we depend for survival upon an outer skeleton,
upon a shell of historic cities and houses filled with things belonging to definable portions of the past. Our ways of describing this
visible past are still most awkward. The systemaric study of
things is less than five hundred years old, beginning with the
description of works of art in the artists' biographies of the
Italian Renaissance. The method was extended to the description
of all kinds of tl:Ungs only after 1750. Today archaeology and
etlmology treat of material culture in general. The l:Ustory of art
treats of the least useful and most expressive products of human
'J'!"'C
THE HISTORY OF THINGS
industry. The family of things begins to look like a smaller family

than people once thought.
The oldest surviving things made by men are stone tools. A
continuous series runs from them to the things of today. The
series has branched many times, and it has often run out into dead
ends. Whole sequences of course ceased when families of artisans
died out or when civilizations collapsed, but the stream of things
never was completely stilled. Everything made now is either a
replica or a variant of something made a little time ago and so
on back without break to the first morning of human time. This
continuous connection in time must contain lesser divisions.
The narrative historian always has the privilege of deciding
that continuity cuts better into certain lengths than into others.
He never is required to defend his cut, because history cuts anywhere with equal ease, and a good story can begin anywhere the
teller chooses.
For others who aim beyond narration the question is to fmd
cleavages in history where a cut will separate different types of
happening.' Many have thought that to make the inventory
would lead toward such an enlarged understanding. The archae1.
I owe my first concern with the problems set forth here to the works and
ologists and anthropologists classify things by their nses, having

first separated material and mental culture, or thit1gs and ideas.
The historians of art, who separate useful and aesthetic products,
classify these latter by types, by schools, and by styles.
Schools and styles arc the products of the long stock-taking
of the nineteenth-century historians of art. This stock-taking,
however, cannot go on endlessly; in theory it comes to an end
with irreproachable and irrefutable lists and tables.
In practice certain words, when they are abused by too common use, suffer in their meaning as if with cancer or inflation.
Style is one of these. Its innumerable shades of meaning seem to
span all experience. At one extreme is the sense defined by Henri
Focillon, of style as the ligne des hauteurs, the Himalayan range
composed of the greatest monuments of all time, the touchstone
and standard of artistic value. At the other extreme is the commercial jtmgle of advertising copy, where gasolines and toilet
papers have "style," and another zone where annual fashions in
clothes are purveyed as "styles." In between lies the familiar
terrain of "historic" styles: cultures, nations, dynasties, reigns,
regions, periods, crafts, persons, and objects all have styles. An
unsystematic naming on binomial principles (Middle Minoan
style, style FrallfOis I") allows an illusion of classed order.
person of the late A. L. Kroeber. Our correspondence began in 1938 soon after I
read his remarkable study (with A. H. Gayton) on the Nazca pottery of southern
coastal Peru, "The Uhle Pottery Collections from Nasca," University of California Publications in American Archaeology and Ethnology, 24 (1927). It is a statistical
analysis based upon the assumption that undated items belonging to the same
form-class can be arranged in correct chronological order by shape-design correlations on the postulate that in one form-class simple formulatiom are replaced
by complex ones. See also A. L. Kroeber, ''Toward Defmition of the Na7ca Style,''
ibid., 43 (1956), and my review in American Antrquity, 22 (1957), 319-20. Professor Kroeber's later volume entitled Configuratl'ons of Culture Growth (Berkeley,
1944) explored more general historic patterns, especially the clustered bursts of
achievement marking the history of all civilizations. These themes continued as
Kroebcr's principal interest in the book oflcctures entitled Style and Civilizrttiolls
(Ithaca, 1956).
In an arresting review G. E. Hutchinson, the biologist, compared Kroeber's
Configurations to internal or free oscillations in animal populations by subjecting
Kroeber's work to mathematical expressions like those used in population studies.
The review is reprinted in The Itl'nmmt Ivory Tower (New Haven, 1953), pp.
74-77, from which I quote: "The great man, born to the period where dN/ dt is
maximal [where N is the degree of pattern saturation] can do much. His precursors
have provided the initial technical inspiration; much still remains to be done.
If he were born to the tradition later he would, with the same native ability, appear less remarkable, for there is less to do. Earlier the work would have been
harder; he would perhaps be highly esteemed by a small body of highly educated
critics, but would never attain the same popular following as if he had worked
at the time of maximum growth of the tradition. The rising and falling that we
see in retrospect is thus to be regarded as a movement to and from a maximum
in a derived curve. The integral curve giving the total amount of material produced seems to depend little on individual achievement, being additive, and therefore is less easily appreciated. We are less likely to think of 1616 as the date by
which most Elizabethan drama has been written than as the date of Shakespeare's
death."
But the whole arrangement is unstable: the key word has

different meanings even in our limited binomial context, signifying at times the common denon1inator among a group of objects,
and at others the impress of an individual ruler or artist. In the
firSt of these senses style is chronologically unrestricted: the common denominator may appear at widely separated places and
times, leading to "Gothic Mannerist'' and "Hellenistic Baroque."
In the second sense style is restricted in time but not in content. Since the lifetime of one artist often embraces many
''styles,'' the individual and ''style'' are by no means coterminous
entities. The "style Louis XVI" embraces the decades before
1789, but the term fails to specify the variety aud the transformations of artistic practice under that monarch's rule.
The immense literature of art is rooted in the labyrinthine
network of the notion of style: its ambiguities aud its inconsistencies mirror aesthetic activity as a whole. Style describes a
specific ftgnre in space better than a type of existence in time.2
In the twentieth century, under the impulse of the symbolic
interpretation of experience, another direction of study has taken
form. It is the study of iconographical types as symbolic expressions of historical change, appearing under a revived seventeenth-century rubric as "iconology." More recently still, the
historians of science have conjoined ideas and things in an inquest upon the conditions of discovery. Their method is to reconstruct the heuristic mmnents of the history of science, and
thus to describe happening at its point of inception.
The moment of discovery and its successive transformations as
traditional behavior are recovered as a matter of program by the
history of science and by iconological studies. But these steps only
outline the beginnings and the main articulations of historical
substance. Many other possible topics crowd upon the attention
2. Meyer Schapiro, "Style," Anthropology Today (Chicago, 1953), pp. 287-312,
reviews the principal current theories about style, concluding dispiritedly that
"A theory of style adequate to the psychological and historical problems has
still to be created."
LIMITATIONS OF BIOGRAPHY
as soon as we admit the idea that this substance possesses a

structure of which the divisions are not merely the inventions
of the narrator.
Although inanimate things remain onr most tangible evidence
that the old human past really existed, the conventional metaphors used to describe tbis visible past are mainly biological. We
speal< without hesitation of the "birth of an art," of the "life of a
I" o f"1 owermg,
. "" matunty,
. "
stye,
I " and t he "death o f a scI100,
and "fading" when we describe the powers of au artist. The
customary mode of arranging the evidence is biographical, as if
the single biographical unit were the true unit of study. The assembled biographies then are grouped regionally (e.g. "Umbrian
School'') or by style and place (''Roman Baroque''), in a manner
vaguely patterned upon biological classifiCations by typology,
morphology, aud distribution.
THE LIMITATIONS OF BIOGRAPHY
The lives of the artists have been a genre in the literature of

art ever since Filippo Villani collected anecdotes in 1381-2.
Artistic biography was expanded in this century with documents
and texts to gigantic proportions as a necessary stage in the
preparation of the grand catalogue of persons aud works. People
writing the history of art as biography assume that the final aims
of the historian are to reconstruct the evolution of the person
of the artist, to authenticate attributed works, at1d to discuss
their meaning. Bruno Zevi, for instance, praises artistic biography
as an indispensable instrument in the training of young artists. 3
J. A leading figure among younger architectural historians in Italy, Zevi observes in hi.~ book-length article on "Architecture" (as published in the American
edition of the Italian Encyclopedia of World Art, 1 [New York, 1959], cols. 683-84)
that the recent union between the history of art and studio or drafting-room
work, whlch is now being accomplished in European art education, could not
begin until the historians escaped from old misconceptions about art itself: and
until historians were ready to "give critical support to the creative experience of
'
THE HISTORY OF THINGS i
The history of an artistic problem, and the history of the in genius" of the Renaissance more simply as a qualified individual
dividual artist's resolution of such a problem, thus fmd a practical , bestriding many new tracks of development at a fortunate mojustification, which, however, confmes the value of the history , ment in that great renovation of Western civilization, and travelof art to matters of mere pedagogical utility. In the long view, , ing his distance in several systems without the burdens of rigorous
proof or extensive demonstration required in later periods.
biographies and catalogues are only way stations where it is easy
to overlook the continuous nature of artistic traditions. These
"Good" or "bad" entrances arc more than matters of position
traditions cannot be treated properly in biographical segments.

Biography is a provisional way of scanning artistic substance
but it does not alone treat the historical question in artists' lives,
which is always the question of their relation to what has preceded and to what will follow them.
Individual entrances. The life of an artist is rightly a unit of
study in any biographical series. Bnt to make it the main unit of
study in the history of art is like discussing the railroads of a
in the sequence. They also depend upon the union of tempera-
country in terms of the experiences of a single traveler on several
mental endowments with specific positions. Every position is

keyed, as it were, to the action of a certain range of temperaments. When a spccif1c temperament interlocks with a favorable
position, the fortunate individual can extract from the situation
a wealth of previously unimagined consequences. This achievement may be denied to other persons, as well as to the san1e
person at a different time. Thus every birth can be imagined as

set into play on two wheels of fortune, one governing the allotment of its temperament, and the other ruling its entrance into
of them. To describe railroads accurately, we are obliged to disregard persons and states, for the railroads themselves are the
a sequence.
elements of continuity, and not the travelers or the functionaries

thereon.
artists are not so n1uch those of talent as of entrance and position
The analogy of the track yields a useful formulation in the dis-
in sequence. Talent is a predisposition: a talented pupil begins
cussion of artists. Each man's lifework is also a work in a series
younger; he masters the tradition more quick1y; his inventions
extending beyond him in eitl1er or both directions, depending

upon his position in the track he occupies. To the usual coordinates f1xing the individual's position-his temperament and his
training-there is also the moment of his entrance, this being the
moment in the tradition-early, middle, or late-with which
his biological opportunity coincides. Of course, one person can
and does shift traditions, especially in the modern world, in order
come more fluently than those of his untalrnted fellows. But

undiscovered talents abound as well among people whose schooling failed to gear with their abilities, as among people whose
to fmd a better entrance. Without a good entrance, he is in danger
less than its presence.
of wasting his time as a copyist regardless of temperament and
It is meaningless to debate whether Leonardo was more

talented than Raphael. Both were talented. Bernardino Luini and
Giulio Romano also were talented. But the followers had bad
luck. They came late when the feast was over through no fault
of their own. The mechanics of fame are such that their prede-
training. From this point of view we can see the "universal

contemporary artists.'' He regards the problem of constructing an artistic education on historical principles as "one of the most vigorous cultural battles of the
195o's" in Europe and America. That battle is still f.1.r from over in America.
Its continuation appears as foolish now as it was wasteful in the past.
Talent and genius. By this view, the great differences between
abilities were unrequited in spite of their talent. Predispositions

are probably much 1nore numerous than actual vocations allow
us to suppose. The quality talented people share is a matter of

kind more than degree, because the gradations of talent signify
cessors' talent is magnified, and their own is diminished, when
~
;
8
talent itself is only a relatively common predisposition for visual

order, without a wide range of differentiation. Times and op,
porttmities differ more than the degree of talent.
Of course many other conditions must reinforce talent: physi,
cal energy, durable health, powers of concentration, are a few of
the gifts of forttme with which the artist is best endowed. But
our conceptions of artistic genius tmderwent such fantastic trans..
formations in the romantic agony of the nineteenth century that
we still today unthinkingly identify "genius" as a congenital disposition and as an inborn difference ofkind.anwng men, instead
of as a fortuitous keying together of disposition and situation

into an exceptionally efficient entity. There is no clear evidence
that "genius" is inheritable. Its incidence under nurture, in situations favorable to craft learning, as with adopted children reared
in the families of professional musicians, marks '"genius" as a
phenomenon of learning rather than of genetics.
Purpose has no place in biology, but history has no meaning
without it. In that earlier transfer of biological ideas to historical
events, of which so many traces survive in the historian's diction,
both typology (which is the study of kinds and varieties) and

morphology (the study of forms) were misunderstood. Because
these modes of biological description ca1mot be made to account
for purpose, the historian working with biological ideas avoided
the principal aim of history, which usually has been to identify
and reconstruct the particular problem to which any action or
thing must correspond as a solution. Sometimes the problem is
a rational one, and sometimes it is an artistic one: we always 1nay
be sure that every man-made thing arises from a problem as a

purposeful solution.
Biological and physical metaphors. However useful it is for
pedagogical purposes, the biological metaphor of style as a sequence of life-stages was historically misleading, for it bestowed
upon the flux of events the shapes and the behavior of organisms.
By the metaphor of the life-cycle a style behaves like a plant. Its
first leaves arc small and tentatively shaped; the leaves of its
1uiddle life are fnlly formed; and the last leaves it puts forth are
small again but intricately shaped. All are sustained by one unchanging principle of organization common to all members of
that species, with variants of race occurring in different environ-
tnents. By the biological metaphor of art and history, style is the

species, and historical styles are its taxonomic varieties. -As an
approximation, nevertheless, this metaphor recognized the recurrence of certain kinds of events, and it offered at least a provisional explanation of them, instead of treating each event as an
unprecedented, never-to-be-repeated unicum.
The biological model was not the most appropriate one for a
history of things. Perhaps a system of metaphors drawn from
physical science would have clothed the situation of art more
adequately than the prevailing biological metaphors: especially
if we are dealing in art with the transmission of son1e kind of
energy; with impulses, generating centers, and relay points; with
increments and losses in transit; with resistances and transformers
in the circuit. In short, the language of electrodynamics might

have suited us better than the language of botany; and Michael
Faraday might have been a better mentor than Linnaeus for the
study of material culture.
Our choice of the "history of things" is more than a euphemism to replace the bristling ugliness of "material culture." This
term is used by anthropologists to distinguish ideas, or "mental

culture," from artifacts. But the '~history of things" is intended
to reunite ideas and objects under the rubric of visual forms: the
term includes both artifacts and works of art, both replicas and
uuique examples, both tools and expressions-in short all materials worked by human hands under the guidance of connected
ideas developed in temporal sequence. From all these things a
shape in time emerges. A visible portrait of the collective identity,
whether tribe, class, or nation, comes into being. This self-image
reflected in things is a guide and a point of reference to the group

for the fnture, and it eventually becomes the portrait given to
posterity.
r~
10
'!
TIIB IIISTORY OF 'fH!"GS

~
i,'.
Although both the history of art and the history of science;

have the same recent origins in the eighteenth-century learning:
of the European Enlightenment, our inherited habit of separating
art frmn science goes back to the ancient division between liberal
and mechanical arts. The separation has had most regrettable
consequences. A principal one is our long reluctance to view the
processes comn1on to both art and science in the satne historical
perspective.
Scientists and artists. Today it is often remarked that two
painters who belong to different schools not only have nothing
to learn from each other but are incapable of any generous con1~
munication with one another about their work. The san1_e thing
is said to be true of chemists or biologists with different specialties.
If such a 1ncasurc of reciprocal occlusion prevails between mem..
bcrs of the same profession, how shall we conceive of communi..
cation between a painter and a physicist? Of course very little
occurs. The value of any rapprochement between the history of
art and the history of science is to display the common traits of
invention, change, and obsolescence that the 1naterial works of
artists and scientists both share in time. The tnost obvious ex..
amples in the history of energy, such as steam, electricity, and
internal combustion engines, point to rhythms of production and
desuetude with which students of the history of art also are
familiar. Science and art both deal with needs satisf1ed by the
mind and the hands in the manufacture of things. Tools and instruments, symbols and expressions all correspond to needs, and
all must pass through design into matter.
Early experimental science had intimate connections with the
studios and workshops of the Renaissance, although artists then
aspired to equal status with the princes and prelates whose tastes
they shaped. Today it is again apparent that the artist is an
artisan, that he belongs to a distinct human grouping as homo
Jaber, whose calling is to evoke a perpetual renewal of form in
matter, and that scientists and artists are more like one another as
artisans than they are like anyone else. For our purposes of dis-
L[MI'tATIONS OF BIOGRAPHY
II
cussing the nature of happening in the world of things, the

differences between science and art are nevertheless irreducible,
uite as much so as the differences between reason and feeling,
fetween necessity and freedom. Although a common gradient
connects use and heauty, the two are irreducibly different: no
tool can be fully explained as a work of art, nor vice versa. A
tool is always intrinsically simple, however elaborate its mechanisms may be, but a work of art, which is a complex of nuny
stages and levels of crisscrossed intentions, is always intrinsically
complicated, however simple its effect may seem.
A recent phenomenon in Europe and America, perhaps not
antedating 1950, is the approaching exhaustion of the possibility
of new discoveries of major types in the history of art. Each
generation since Winckelmann was able to mark out its own
preserve in the history of art. Today there are no such restricted
preserves left. First it was classic art that commanded all admiration at the expense of other expressions. The romantic gener.ation
again elevated Gothic art to the pedestal. Some fin de silcle
architects and decorators reinstated Roman Imperial art. Others
generated the languors and the botanical clahorations of art
nouveau on the one hand, or the rebels among them turned to
primitivism and archaic art. By a kind of rule of the alternation
of generations between tutelary styles of civilized and rude aspect,
the next generation adverted to baroque and rococo-the generation that was decimated by the First World War. The revival of
interest in sixteenth-century Matmerism which flared during the
1930's not only coincided with great social disorders but it indicated a historical resonance between the men of the Reformation and those of a time of depression and demagogy.' After
that, nothing was left to discover unless it was contemporary
4. Writing in 1935 (" 'Stilgeschichte' und 'Sprachgeschichte' dcr bildenden
Kunst," Sitzungsberichte der bayrischen Akademie der Wissenschaften, 31), Julius von
Schlosser referred with feeling to the modern Mannerism being propagated by
the "Kopisten, Nachahmer, Industriellen . , . von unserem heutigen Dekadententum.''
~..
_l-
I2
THE HISTORY OP THIJ:>.TQs!
art. The last cupboards and closets of the history of art have now
been turned out and catalogued by government ministries of.
Education and Tourism.
Seen in this perspective of approaching completion, the anna],
of the craft of the history of art, though brief, contain recurrent
situations. At one extreme the practitioners feel oppressed by the
fullness of the record. At the other extreme we have works of
rhapsodical expression like those dissected by Plato in the
Socratic dialogue with Ion. When Ion, the vain rhapsodist,
parades his boredom with all poets other than Homer, Socrat~
says, " ... your auditor is the last link of that chain which I have
described as held together by the power of the magnet. Yon
rhapsodists and actors are the middle links, of which the poet is
'l'lfll HISTORIAN'S COMMITMENT
I3
and the curious searcher much as the composer of new 1nusic
differs from the concert performer. The historian composes a

eaning from a tradition, while the antiquarian only re-creates,
111
performs, or re-enacts an obscure portion of past ti':'e in already
familiar shapes. Unless he 1S an annahst or a chromcler the hiStorian conununicates a pattern which was invisible to his sub-
jects when they lived it, and unknown to his contemporaries

before he detected it.
For the shapes of time, we need a criterion that is not a mere
transfer by analogy from biological science. Biological time consists of uninterrupted durations of statistically predictable lengths:
each organism exists from birth to death upon an "expected"
life-span. Historical time, however, is intermittent and variable.

Every action is more intermittent than it is continuous, and the
If the fullness of history is forever indigestible, the beauty of intervals between actions are infinitely variable in duration and
the first.'' s
art is ordinarily incommunicable. The rhapsodist can suggest , content. The end of an action and its beginning are indeterminate.
few clues to the experience of a work of art, if he himself ha, Clusters of actions here and there thin out or thicken sufficiently
indeed experienced it. He may hope that these hints will assist to allow us with some objectivity to mark beginnings and endthe hearer to reproduce his own sensations and ntental processes. ings. Events and the intervals between them are the elements of
He can communicate nothing to persons not ready to travel the the patterning of historical time. Biological time contains the
same path with him, nor can he obey any f1eld of attraction b"' unbroken events called lives; it also contains social organizations
yond his own direct experience. But historians are not middle by species and groups of species, but in biology the intervals of
links, and their mission lies in another quarter.
time between events are disregarded, while in historical thne the
web of happening that laces throughout the intervals between

THE
HrsronrAN's CoMMITMENT
existences attracts our interest.
Time, like mind, is not knowable as such. We know time only

The historian's special contribution is the discovery of the
indirectly by what happens in it: by observing change and permanifold shapes of time. The aim of the historian, regardless of
manence; by marking the succession of events among stable
his specialty in erudition, is to portray time. He is cmnmitted to
settings; and by noting the contrast of varying rates of change.
the detection and description of the shape of time. He transposes,
Written documents give us a thin recent record for only a few
reduces, composes, and colors a facsimile, like a painter, who in
parts of the world. In the main our knowledge of older times is
his search for the identity of the subject, must discover a patterned
based upon visual evidence of physical and biological duration.
set of properties that will elicit recognition all while conveying
Technological sedations of all sorts and sequences of works of
a new perception of the subject. He differs from the antiquarian
art in every grade of distinction yield a finer time scale overlapping with the written record.
5 Ion, tr. Jowett, Dialogues of Plato (New York, 1892), vol. I.
r~
!'
14
THE HISTORY OP THINGS: r!IE HISTORIAN's COMMITMENT
Now that absolute confirmations by tree-rings and earth.'

clocks are at hand, it is astonishing in retrospect to discover how.
very accurate were the older guesses of relative age based upott
sedations and their comparisons. The cultural clock preceded ali
the physical methods. It is nearly as exact, and it is a more search.
ing method of measurement than the new absolute clocks, which
often still require confirmation by cultural means, especially
when the evidence itself is of mixed sorts.
The cultural clock, however, rm1s mainly upon ruined frag.
tnents of matter recovered from refuse heaps and graveyards,
from abandoned cities and buried villages. Only the arts of ma.
terial nature have survived: of music and dance, of talk and ritual,
of all the arts of temporal expression practically nothing is known
elsewhere than in the Mediterranean world, save through traditional survivals among remote groups. Hence our working proof
of the existence of nearly all older peoples is in the visual order,
and it exists in matter and space rather than in time and sound.
We depend for our extended knowledge of the human past
mainly upon the visible products of man's industry. Let us suppose a gradient between absolute utility and absolute art: the
pure extremes are only in our imagination; human products
always incorporate both utility and art in varying mixtures, and
no object is conceivable without the admixture of both. Archaeo
logical studies generally extract utility for the sake of information
about the civilization: art studies stress qualitative matters for
the sake of the intrinsic meaning of the generic human experience.
The divisions of the arts. The seventeenth-century academic
separation between fine and useful arts first fell out of fashion
nearly a century ago. From about 1880 the conception of "fine
art" was called a bourgeois label. After 1900 folk arts, provincial
styles, and rustic crafts were thought to deserve equal ranking
with court styles and metropolitan schools under the democratic
valuation of twentieth-century political thought. By another
line of attack, "fine art" was driven out of use about 1920 by the
exponents of industrial design, who preached the requirement of
15
universal good design, and who opposed a double standard of

judgment for works of art and for useful objects. Thus an idea
of aesthetic unity carrte to embrace all artifacts, instead of ennobling some at the expense of others.
This egalitarian doctrine of the arts nevertheless crases many
important differences of substance. Architecture and packaging
tend in the modern schools of design to gravitate together under
the rubric of envelopes; sculpture absorbs the design of all sorts
0 [ small solids and containers; painting extends to include flat
shapes and pianos of all sorts, like those of weaving and printing.
By this geometric system, all visible art can be classed as envelopes, solids, and planes, regar<Uess of any relation to use, in a
classing which ignores the traditional distinction by "fine'' and
"minor," or "useless" and "useful" arts.
For our purposes two urgent distinctions should be added. In
the first place a great difference separates traditional craft educati\m from the work of artistic invention. The former requires
only repetitious actions, but the latter depends upon departures
from all routine. C1 aft education i~ the activity of groups of
learuers performing identical actk'llS 1 but artistic invention requires the solitary efforts of individu.tl persuns. The distinctiun
is worth retaining because artists working in different crafts
cannot communicate with one another in technical tnatters but
only in matters of design. A weaver learns nothing about his
luom and threads from study of the potter's wheel and kiln;
his education in a craft must be upon the instruments of that craft.
Only when he pos<>esscs technical control of his instruments can
the qualities and etfCcts of Jcsign in other crafts stimulate him to
new solutions in his own.
The second, related distinction touche::. the utilitarian and the
aesthetic nature of each of the branchrs of artistic practice. In
anhitecture and the ,11lied crafts, structure pertains to traditional
technical trainmg and it is inherently ratio11al and utilitarian, however daringly it\ drviccs may he applied to expressive ends. In
sculpture and painting likewise, every work has its technical
r
l
"j''
I
'
I6
THE liiSTORY OF THiliGS ~
cookery offormulas and craft practices upon which the expressive i

and formal combinations are carried. In addition, sculpture and '
painting convey distinct messages more clearly than architecture, .
These communications or iconographic themes make the utili-
tarian and rational substructure of any aesthetic achievement.
Thus structure, technique, and iconography all belong to the
non-artistic underpinning of the "fmen arts.
The main point is that works of art are not tools, although
many tools may share qualities of fmc design with works of art.
We are in the presence of a work of art ouly when it has no
preponderant instrumental use, and when its technical and rational foundations are not pre-eminent. When the technical organization or the rational order of a thing overwhelms our
attention, it is an object of use. On this point Lodoli anticipated
the doctrinaire functionalists of our century when he declared in
the eighteenth century that only the necessary is beautiful. Kant,
however, more correctly said on the same point that the necessary
cannot be judged beautiful, but only right or consistent.' In
short, a work of art is as useless as a tool is useful. Works of art
are as unique and irreplaceable as tools are common and expendable.
THE NATURE OF ACTUALITY
"Le passe ne sert qu'a connaitre 1'actualite. Mais 1' actualite

m'6chappe. Qu'est-ce que c'est done que l'actualitl?'' For years thi1
question-the fmal and capital question of his life-obsessed my
teacher Henri Focillon, especially during the black days from
I940 to 1943 when he died in New Haven. The question hm
6. See Emil KaufmaiUl, Architecture in the Age of Reason (Cambridge, Mass.,

1955), pp. 95-100.
7 Paul Menzer, "Kants Asthetik in ihrer Entwicklung," Abhandlungen dll
deutschen Akademie der Wlssenschaflen zu Berlin, Kl. fiir Gesellschaflswissenschcifte~,

Jahrgang 1950 (1952).
tllB N.ATURB OF ACTUALITY
17
been with me ever since, and I am now no closer to the solution

of the riddle, unless it be to suggest that the answer is a negation.
Actuality is when the lighthouse is dark between flashes: it is
the instant between the ticks of the watch: it is a void interval
slipping forever through time: the rupture between past and
future: the gap at the poles of the revolving magnetic f1eld, infuriteshnally small but ultimately real. It is the interchronic pause
when nothing is happening. It is tlte void between events.
Yet the instant of actuality is all we ever can know directly.
The rest of time emerges ouly in signals relayed to us at this
instant by innumerable stages and by unexpected bearers. These
signals are like kinetic energy stored until the moment of notice
when the mass descends along some portion of its path to the
center of the gravitational system. One may ask why these
old signals are not actual. The nature of a signal is that its message
is neither here nor now, but there and then. If it is a signal it is
a past action, no longer embraced by the "now" of present being.
The perception of a signal happens "now," but its impulse and
its transmission happened "then." In any event, the present instant is the plane upon which the signals of all being are projected.
No other plane of duration gathers us up universally into the
same instant of becoming.
Our signals from the past are very weak, and our means for
recovering their meaning still are most imperfect. W eal<est and
least clear of all are those signals coming from the initial and
terminal moments of any sequence in happening, for we are unsure about our ideas of a coherent portion of time. The beginnings are much hazier than the endings, where at least the
catastrophic action of external events can be determined. The
segmentation of history is still an arbitrary and conventional
matter, govemed by no verifiable conception of historical entities and their durations. Now and in the past, most of the time
the majoriry of people live by borrowed ideas and upon traditional accmuulations, yet at every moment the fabric is being
undone and a new one is woven to replace the old, while from
r8
'i
THE HISTORY OF TRINes~
time to time the whole pattern shakes and quivers, settling into l
new shapes and figures. These processes of change are all mys.:
terious uncharted regions where the traveler soon loses directio!)
and stumbles in darkness. The clues to guide us are very few in.
deed: perhaps the jottings and sketches of architects and artists,
put down in the heat of imagining a form, or the manuscript
brouillons of poets and musicians, crisscrossed with erasures and
corrections, are the hazy coast lines of this clark continent of the
''now," where the impress of the future is received by the past.
To other animals who live more by instinct than do humm~,
the instant of actuality must seem far less brief. The rule ofinstinct
'!'fill NATURE OF ACTUALITY
19
timated by the possibilities that fail to attain realization in any

instant: only when they are few can actuality seem full.
Of arts and stars. Knowing the past is as astonishing a performance as knowing the stars. Astronomers look only at old
light. There is no other light for them to look at. This old light
of dead or distant stars was emitted long ago and it reaches us
only
in the present. Many historical events, like astronomical
bodies, also occur long before they appear, such as secret treaties;
aide-m6moircs, or important works of art made for ruling per-
sonages. The physical substance of these documents often reaches

qualified observers only centuries or millewia after the event.
is automatic, offering fewer choices than intelligence, with ch.. Hence astronomers and historians have this in c01nmon: both are
cuits that close and open unselectively. In this duration choice is

so rarely present that the trf\iectory from past to future describes
a straight line rather than the infinitely bifurcating system of
human experience. The ruminant or the insect must live time
more as an extended present which endures as long as the inclividuallife, while for us, the single life contains an infmity of
present instants, each with its innumerable open choices in vo.
lition and in action.
Why should actuality forever escape our grasp1 The universe
has a fmite velocity which limits not only the spread of its events,
but also the speed of our perceptions. The moment of actuality
slips too fast by the slow, coarse net of our senses. The galaxy
whose light I see now may have ceased to exist millennia ago, and
by the same token men cannot fully sense any event until after
it has happened, until it is history, until it is the dust and ash of
that cosmic storm which we call the present, and which per
petually rages throughout creation.
In 1ny own present, a thousand concerns of active business lie
unattended while I write these words. The instant admits only
one action while the rest of possibility lies unrealized. Actuality
is the eye of the storm: it is a diamond with an infmitesimal per
foration through which the ingots and billets of present possibility
arc drawn into past events. The emptiness of actuality can be es
concerned with appearances noted in the present but occurring

in the past.
The analogies between stars and works of art can profitably
be pursued. However fragmentary its condition, any work of
art is actually a portion of arrested happening, or an emanation
of past time. It is a graph of an activity now stilled, but a graph
made visible like an astronomical body, by a light that originated
with the activity. When an important work of art has utterly
disappeared by demolition and dispersal, we still can detect its
perturbations upon other bodies in the field of influence. By the
same token works of art resemble gravitational fields in their
clustering by "schools." And if we admit that works of art can
be arranged in a te1nporal series as connected expressions, their
sequence will resemble an orbit in the fewness, the regularity, and
the necessity of the "motions" involved.
Like the astronomer, the historian is engaged upon the portrayal of time. The scales are different: historic time is very short,
but the historian and the astronomer both transpose, reduce,
compose, and color a f.Ksimile which describes the shape of
time. Historical time indeed may occupy a situation near the
center of the proportional scale of the possible magnitudes of
time, just as man himself is a physical magnitude midway between the sun and the atom at the proportional center of the
~~
I .
20
THB HISTORY OP 'riiUHJ:sf
'f.fl:S NATURE OF ACTUALITY
2I
solar system, both in grams of mass and in centimeters of di,' with the past therefore originated as signals whiclt become comameter.s
signals in an 1mbroken alternating semotions emitting further
.
.
Both astronomers and historians collect ancient signals intQ ~ uence of event, s1gnal, recreated event, renewed s1gnal, etc.
compelling theories about distance and composition. The 6elebrated events have undergone the cycle millions of times
tronomer' s position is the historian's date; his velocity is Ollt: each instant throughout their history, as when the life ofJesus is
sequence; orbits are like durations; perturbations are analogoUs connnemotated in the utmumbered daily prayers of Christians.
to causality. The astronomer and the historian both deal with To reach us, the original event must undergo the cycle at least
past events perceived in the present. Here the parallels diverge, once, in the original event, its signal, and our consequent agitafor the astronomer's future events arc physical and recurrellt tion. The irreducible minimum of historical happening thus reones, while the historian's are human and unpredictable one1 quires only an event together with its signals and a person capable
The foregoing analogies are nevertheless useful in prompting us of reproducing the signals.
Reconstituted initial events extracted from the signals are the
to look again at the nature of historical evidence, so that we may
be sure of our ground when considering various ways of classh1g principal product of historical research. It is the scholar's task to
it.
verify and test all the evidence. He is not concerned primarily
Signals. Past events may be regarded as categorical commo. with the signals other than as evidences, or with the commotions
tions of vatying magnitudes of which the occurrence is declared they produce. The different commotions in turn are the proper
by inbuilt signals analogous to those kinetic energies impounded territory of psychology and aesthetics. Here we are interested
in masses prevented from falling. These energies undergo variow mainly in the signals and their transformations, for it is in this
transformations between the original event and the present. Th, domain that the traditional problems arise which lace together
present interpretation of any past event is of course only anothe1 the history of things. For instance, a work of art is not only the
stage in the perpetuation of the original impulse. Our particulat residue of an event but it is its own signal, directly moving other
interest is in the category of substantial events: events of which makers to repeat or to improve its solution. In visual art, the
the signal is carried by matter arranged in a pattern still sensib!,

today. In this category we are interested less in the natural signah
of physical and biological science, than in the artifact signals oi
history, and among artifact signals we are concerned less wit1
documents and instruments than with the least useful of artifact
-works of art.
All substantial signals can be regarded both as transmissiom
entire historical series is conveyed by such tangible things, tmlike

written history, which concerns irretrievable events beyond physical recovery and signaled only indirectly by texts.
Relays. Historical knowledge consists of transmissions in
which the sender, the signal, and the receiver all are variable elements affecting the stability of the message. Since the receiver of
a signal becomes its sender in the normal course of historical

and as initial commotions. For instance, a work of art transmi~ transmission (e.g. the discoverer of a docU1nent usually is its
a kind of behavior by the artist, and it also serves, like a relay,' editor), we may treat receivers and senders together under the
the point of departure for impulses that often attain extraordina~ heading of relays. Each relay is the occasion of some deformation
magnitudes in later transmission. Our lines of communicatim in the original signal. Certain details seem insignificant and they
are dropped in the relay; others have an importance conferred by
8. Harlow Shapley, Of Stars and Men (New York, 1958), p. 48.
their relationship to events occurring in the moment of the relay,
~
22
1
j
Tlfii NATURE OF ACTUALITY
23
and so they are exaggerated. One relay may wish for reasons of!
temperament to stress the traditional aspects of the signal; another .
will emphasize their novelty. Even the historian subjects his evi-
dence to these strains, although he strives to recover the pristine
signal.
Each relay willingly or unwittingly deforms the signal according to his own historical position. The relay transmits a
composite signal, composed only in part of the message as it was
received, and in part of impulses contributed by the relay itself.
Historical recall never can be complete nor can it be even entirely
of transmitted duration disintegrates more and n1ore the farther
correct, because of the successive relays that deform the n1essage.

The conditions of transmission nevertheless are not so defective
suited to tl1e purpose in hand. For example, the judges and

counsel in a law court may expend upon tbe determination of
that historical knowledge is impossible. Actual events always

excite strong feelings, which the initial message usually records.
A series of relays may result in the gradual disappearance of the
animus excited by the event. The most hated despot is the live
despot: the ancient despot is only a case history. In addition,
many objective residues or tools of the historian's activity, such
as chronological tables of events, cam10t easily be deformed.
the sequence of events leading to a murder, an amount of effort
we go back. Though fmite, the total number of historical signals

greatly exceeds tbe capacity of any individual or group to interpret all the signals in all their meaning. A principal aim of the
historian therefore is to condense the multiplicity and the redundancy of his signals by using various schemes of cltSsification
that will spare us the tedium of reliving the sequence in all its
instantaneous confusion.
Of course, the writing of history has many extremely practical

uses, each of which imposes upon the historian a viewing range
vastly greater than the events tl1emselves required for their

happe11ing. At the other extreme, when I wish to mention
Columbus' first voyage to America, I do not need to collect all
the signals, such as documents, archaeological indications, earthclock measurements, etc., to prove the date 1492.: I can refer to
credible secondary signals derived from firsthand sources. In
Other examples are the persistence of certain religious expressions
between these extremes, an archaeologist tracing a buried floor
through long periods and uuder great deforming pressures. The
level with his assistants spends about the same energy upon reading the signal as tbc original builders put into tbe floor in the
rejuvenation of myths is a case in point: when an ancient version

becomes mllntelligibly obsolete a new version, recast in contem-
porary terms, performs the same old explanatory purposes.'

The essential condition of historical knowledge is that the
event should be within range, that some signal should prove past
existence. Ancient time contains vast durations without signals
of any kind that we can now receive. Even the events of the past
few hours arc sparsely documented, when we consider the ratio
of events to their documentation. Prior to 3000 B.c. the texture
9 H. Hubert and M. Mauss, "La Representation du temps dans la religion,"
Milanges d'histoire des religions (Paris, Alcan, 2nd ed., 1929), pp. 189-229, On
mythopoetic transformations of historical personages, see for example V. Burch,
Myth and Coustantine the Great (Oxford, 1927).
first instance.
Hence a primary signal-meaning the evidence closest to the

event itself-may require a great expense of energy for its detection and interpretation, but once the signal has been brought
in it can be repeated at a fraction of the cost of the original detection. In this way the fundamental determinations of history
relate to detecting and receiving primary signals from the past,
and they usually concern simple matters of date, place, and agent.
For the most part the craft of history is concerned with the
elaboration of credible messages upon the simple foundations
afforded by primary signals. More complex messages have widely
varying degrees of credibility. Some are fantasies existing in the
minds of the interpreters alone. Others are rough approxima-
r
i
THE HISTORY OF THINGs
j SllLp...SIGNALS
AND ADHERENT SIGNALS
25
tions to historical truth, such as those reasonable explanations or\ exhausted the possibilities of such an apparently simple category
myths called euhemerist.lO
' of sensation. Yet this self-signal is the least honored and the most
Still otl1er complex messages are probably stimulated by special: overlooked of the dense stream of signals issuing from the picture.
primary signals of which our understanding is incomplete. These; Jn the consideration of painting, architecture, sculpture, and all
arise from extended durations and from the larger units of geog- their allied arts, the adherent signals crowd in upon most persons'
raphy and population; they are complex, dimly perceived sig11als attention at the expense of the autogenous ones. In a painting,
which have little to do with historical narrative. Only certain for example, the dark foreground figures resemble persons and
new statistical methods come near to their detection, such as the animals; a light is depicted as if emanating from the body of an
remarkable lexicostatistical discoveries made in glottochronology, infant in a ruined shelter; the narrative bond connecting all these
the study of the rate of change oflanguages (pp. 6o-61).
. shapes must be the Nativity according to St. Luke; and a painted
scrap of paper in one corner of the picture bears the name of the
SELF-SIGNALS AND ADHERENT SiGNALS
These remarks so far pertain mainly to one class of historical signals, to distinguish them frmn the more obvious messages
of another kind which we have not yet discussed. These other

signals, including writing, are added to the self-signal, and they
are quite different from it, being adherent rather than autogenous.
The self-signal can be paraphrased as the mute existential declaration of things. For example, the hammer upon the workbench
signals that its handle is for grasping and that the peen is an extension of the user's f!St ready to drive the nail between the fibres
of the plank to a flrm and durable seat. The adherent signal, diestamped on the hammer, says only that the design is patented
tmder a protected trade-mark and manufactured at a commercial
address.
A fine painting also issues a self-signal. Its colors and their
distribution on the plane of the framed canvas signal that by
making certain optical concessions the viewer will enjoy the
simultaneous experience of real surfaces blended with illusions of
deep space occupied by solid shapes. This reciprocal relation of
real surface and deep illusion is apparently inexhaustible. Part of
the self-signal is that thousands of years of painting still have not
IO.
Anne Hersman, Studies in Greek Allegorical Interpretation (Chicago, 1906).
painter and the year of the work. All these are adherent signals
composing an intricate 1ncssage in the symbolic order rather
than in an existential dimension. Adherent signals of course are
essential to our study, but their relations with one another and
with the self-signals make up part, and only part of the game, or
the scheme, or the problem that confronted the painter, to which
the picture is the resolution in actual experience.
The existential value of the work of art, as a declaration about
being, cannot be extracted from the adherent signals alone, nor
from the self-signals alone. The self-signals taken alone prove
only existence; adherent signals taken in isolation prove only the
presence of meaning. But existence without 1neaning seems terrible in the same degree as meaning without existence seems
trivial.
Recent movements in artistic practice stress self-signals alone,
as in abstract expressionism; conversely, recent art scholarship
has stressed adherent signals alone, as in the study of iconography.
The result is a reciprocal misunderstanding between historians
and artists: the unprepared historian regards progressive contemporary painting as a terrifying and senseless adventure; and
the painter regards most art scholarship as a vacant ritual exercise.
This type of divergence is as old as art and history. It recurs in
every generation, with the artist demanding from the scholar
the approval of history for his work before the pattern is com-
'J!!'!":.
1,-> ,'c -
THE HISTORY OF
I
THI~G~s~LPSIGNALS
AND ADHERENT SIGNALS
27
'
plete, and the scholar mistaking his position as an observer a!1\cultural symbols.'' lconology is a variety of cultural history, in
historian for that of a critic, by pronouncing upon matters ~which the study of works of art is devotee\ to the extraction of
contemporary significance when his perceptive skill and l1i coJlclusions concerning culture. Because of its dependence upon
equipment arc less suited to that task than to the study of who\Jong-lived literary traditions, iconology so far has been repast configurations which are no longer in the condition of actiw suicted to the study of the Greco-Roman tradition and its surchange. To be sure, certain historians possess the sensibiliry an; vivals. Continuities of theme are its principal substance: the
the precision that characterize the best critics, but their numb~ breaks and ruptures of the tradition lie beyond the iconologist' s
is small, and it is not as historians but as critics that they manif~ scope, like all the expressions of civilizations without abundant
literary documentation.
these qualities.
The most ,valuable critic of contemporary work is anoth~ Configurational analysis. Certain classical archaeologists in
artist engaged in the same game. Yet few misunderstandings ea. turn also have been much concerned with similar questions about
ceed those between two painters engaged upon different kind meaning, especially in respect to the relations between poetry
of things. Only long after can an observer resolve the differenc~ and visual art. The late Guido v. Kaschuitz-Weinbcrg and
between such painters, when their games are all out, and full) friedrich Matz12 are the principal representatives of this group,
available for comparison.
who engage in the study of meaning by the method of StrukturTools and instruments are recognized by the operational char. analyse, or configurational analysis, in an effort to determine the
acter of their self-signal. It is usually a single signal rather than, premises underlying the literature and art of the same generation
multiple one, saying that a specific act is to be performed in at in one place, as for example, in the case of Homeric poetry and
indicated way. Works of art are distinguished from tools ani the coeval geometric vase painting of the eighth century B.C.
instruments by richly clustered adherent meanings. Works of ao Thus Strukturforschung presupposes that the poets and artists of
specify no immediate action or limited use. They are like gam one place and time are the joint bearers of a central pattern of
ways, where the visitor can enter the space of the painter, or th sensibility from which their various efforts all flow like radial
time of the poet, to experience whatever rich domain the arti~ expressions. This position agrees with the iconologist' s, to whom
has fashioned. But the visitor must come prepared: if he bring literature and art seem approximately interchangeable. But the
a vacant mind or a deficient sensibility, he will see nothing. Ad archaeologists are more perplexed by the discontinuities between
herent meaning is therefore largely a matter of conventiolll painting and poetry than the iconologists are: they still find it
shared experience, which it is the artist's privilege to rearran~ difficult to equate the Homeric epic with Dipylon vases. This
and enrich under certain limitations.
perplexity reappears among students of modern art, to whom
Iconographic studies. Iconography is the study of the forms "literature and painting appear sharply divergent in content and
sumed by adherent meaning on three levels, natural, conve technique. Erudition and pornography are exalted and conjoined
tiona!, and intrinsic. Natural meaning concerns primary identifi in present-day literature, but they are both avoided in painting,
cations of things and persons. Conventional meanings occu
when actions or allegories are depicted which can be explaine( n. Erwin Panofsky, Studies in Iconology (New York, 1938).
.
by reference to hterary sources.
12, Friedrich Matz, Gescltichte der griechischen Kunst (Frankfurt, 1950), .2 vols.
th The introduction is an exposition of Struktuiforschung, for

study called iconology, and they pertain to the explanation aapproximation might be "studies of form-field relations."
lntrmslc meamngs const1tute
which an English
r
28
'
I
THE HISTORY OF THINGS j SiltFSIGNALS AND ADHERENT SIGNALS
29
where the quest for non-representational form has been thei 1ental changes generation by generation, but the outlines of
principal ahn in om century.
' r.rge, coarse changes are clearly evident, such as tl1e differences
The difficulty can be removed by modifying the postulate ofi of iconographic system before and after A.D. 1400 in Western
a central pattern of sensibility among poets and artists of the same. civilization.
place and time. It is unnecessary to reject the idea of central
In tl1e middle ages or during antiquity, all experience fotmd
pattern altogether, because the quest for erudite expression, for its visual forms in a single metaphorical system. In antiquity the
instance, was shared by poets and painters alike in seventeenth~ gesta deorum enveloped the representation of present happening.
century Europe. It is enough to temper the conception of the The Greeks preferred to dtscuss contemporary events under a
governing configuration (Gestalt) with the conception of the mythological metaphor, like that of the labors of Hercules, or
formal sequence set forth here on pp. 33 f. Formal sequences pre- in tenus of the epic situations of Homeric poetry. The Roman
suppose independent systems of expression that may occasionally emperors adopted biographical archetypes among the gods, asconverge. Their survival and convergence correspond to a sumis1g the names, the attributes, and the cults of the deities. In
shared purpose which alone defines the field of force. By this the middle ages the lives of the saints fulfilled the same function,
view the cross-section of the instant, taken across the full face of as when the regional histories ofReims or Amiens found their exthe 1noment in a given place, resembles a mosaic of pieces iu pression in the statues of local saints standis1g is1 the cathedral
different developmental states, and of different ages, rather than embrasures. Other variations on the principal narratives of
a radial design conferring its meaning upon all the pieces.
Scripture conveyed further details oflocal history and sentiment.
The taxonomy of meaning. Adherent meanings vary categori- This preference for reducis1g all experience to the template set
cally according to the entities they clothe. The messages that can by a few master themes resembles a funnel. It channels experience
be conveyed in Meissen porcelain differ from those of large into a more powerful flow; the then1es and patterns are few in
bronze sculpture. Architectural messages are unlike those of number but their intensity of meaning is thereby increased.
painting. The discussion of iconography or iconology imAbout A.D. 1400 many technical discoveries in the pictorial
mediately raises taxonomic questions, analogous to those of dis-- representation of optical space allowed, or more probably, actinguishing the fur, feather, hair, and scales of the biological companied, the appearance of a different scheme of stating exorders: all are integuments, but they differ from one another in perience. This new scheme was more like a cornucopia than a
ftmction, in structure, and in composition. Meanings undergo funnel, and from it tumbled an immense new variety of types
transformations by mere transfer, which are mistaken for changes and themes, more directly related to daily sensation than the
in content.
preceding modes of representation. The classical tradition and
Another difficulty arising from the treatment of iconography its reawakening formed only one current in the torrent of new
as a homogeneous and uniform entity is the presence of large forms embracing all experience. It has been at flood height and
historical groupings within the body of adherent meaning. These steadily rising ever sislCe the fifteenth century.
are related more to the mental habits of different periods than to
The survival of antiquity has perhaps commanded the attention
incorporation as architecture, sculpture, or painting. Our his- of historians mainly because the classical tradition has been supertorical discriminations still are too imprecise to document these
seded; because it is no longer a live water; because we are now
~ntside
I
l
THE HISTORY OF
TH!~J
it, and not inside it." We are no longer borne by it

m a current upon the sea: it is visible to us from a distance and~
perspective only. as a major part of the topography of histo~
By the same token we cannot clearly descry the contours of~~
great currents of our own time: we are too much inside cl1;
streams of contemporary happening to chart their flow and vo[:
ume. We are confronted with inner and outer historical surface:
(p. 54). Of these only the outer surfaces of the completed p
are accessible to historicallmowledge.
<I
2.
The Classing of Things
only a few art historians have sought to discover valid

ways to generalize upon the immense domain of the experience
IJ. E. Panofsky, Renaissance and Renascences (Stockholm, 1960), has commeutt:

at some length on the end of the modern age in the present century.
' 0 [ art.
These few have tried to establish principles for architecture,

sculpture, and painting upon an intermediate ground partly in
the objects and partly in our experience of them, by categorizing
the types of organization we perceive in all works of art.
One strategy requires enlarging the unit of historical happening. At the begi1111ing of the century F. Wickhoff and A. Riegl
moved in this direction when they replaced the earlier moralizing
judgment of "degeneracy" that had been passed upon Late
Roman art, with the hypothesis that one system or organization
was being replaced by a new and different system of equal value.
' l 's tern1s, one "ill
r " gave way to anot h
In Rteg
w -to-rorm
cr.t
Such a division of history, along the structural lines marked by
the frontiers between types of formal organization, has had reserved approval from almost all twentieth-century students of
art and archaeology.
These proposals differed altogether from the notions of necessary sequence first advanced by the Swiss historian of art, Heinrich Wolffiin, whose work was classed with the "theory of pure
visibility" by Benedetto Croce.z Wolffiin compared fifteenthA. Riegl, Dfe Spiitriimische Kunstindustrie (Vienna, 1901-23), 2 vols.
The term itself arose according to Croce ("La teoria dell'artc come pura
visibilita," Nuovi saggi di estetica [Bari, 1926], pp. 233-58) in conversation among
Hans von Marees, Conrad Fiedler, and Adolf Hildebrand in Munich. The date
was probably about 1875 WOlfflin's most celebrated work is KunstgeschichtUche
Gnmdbegriffe (Munich, I9IS)
I.
2.
rI
THE CLASSING OB THiNG!!
I
I
i
century aud seventeenth-century Italian art. By pointing out fiv \
polar opposites in the realizati011 of form (linear-painter!':
sur~ace-depth; closed-open; mrdtiplicity-ru1ity; absolute-relati~;.
clanty) he usefrdly characterized some fundamental differenc,
of morphol~gy in me two periods. Oilier writers soon extend,;
the conceptton to both Greco-Roman and medieval art in
three-part division of each by archaic, classic, and baroque :cages'
A fourili stage called mannerism (the sixteenth century) was u1
serted about 1930 between classic and baroque. Occasionally
wnte~s e~en have promoted the rococo and neo-classic styles to
me d1grury of stages in the life-cycle. Wolffiin's categories had
great influence in affecting the historical scholarship both of
music and literature, without ever reaching unquestioned acceptance among art historians themselves.
'
,,
'
There, the archive-minded specialists naturally found ue~

documents more usefrd than stylistic opinions spWl from Wolf.
flin' s Grundbegriffe, and at the other extreme, the rigorous his.
torians condemned Wolffiin because he neglected the individual
qualities of things and artists in trying to frame general observa.
tions concerning ilieir classes. The boldest and most poetic
affirmatiOn of a biological, co~ception of the nature of the history
of art was Hcun Focillon s Vte des Formes (1934). The biological
n~etaphor was of course only one among many pedagogical devices used by that marvelous teacher who turned everything to
good account in firing the common clay of his hearers. Unin.
formed critics have misunderstood his extraordinary inventiveness of rnind as rhetorical display, while the more ponderous
ones have dismissed him as another formalist.
The shapes of time are the prey we want to capture. The time
of history is too coarse and brief to be an everdy granular duration such as the physicists suppose for natural time; it is more
like a sea occupied by innumerable forms of a fmite number of
types. A net of another mesh is required, different from any now
muse. The notiOn of style has no more mesh than wrapping paper
or storage boxes. Biography cuts and shreds a frozen historic
33
pORJIIAL SEQUENCES
substance. Conventional histories of architecture, sculpture, paint-
ing, and the cognate crafts miss both me minute and main details
of artistic activity. The monograph upon .a single work of art is
like a shaped stone ready for position in a masonry wall, but that
wall itself is built without purpose or plan.
FORMAL SEQUENCES
Every important work of art can be regarded both as a

historical event and as a hard-won solution to some problem. It
is irrelevant now whether the event was original or conventional,
accidental or willed, awkward or skillfrd. The important clue is

that any solution points to the existence of some problem to
which there have been other solutions, and that other solutions
to this same problem will most likely be invented to follow the
one now in view. As the solutions accumulate, the problem
alters. The chain of solutions nevertheless discloses the problem.

Linked solutions. The problem disclosed by any sequence of
artifacts may be regarded as its mental form, and me linked solutions as its class of being. The entity composed by the problem
and its solutions constitutes a form-class. Historically only those
solutions related to one anomer by the bonds of tradition and
influence are linlced as a sequence.
Linked solutions occupy time in a great variety of ways, discussed in the remainder of this book. They disclose a finite yet
uncharted domain of mental forms. Most of these are still open
to further elaboration
by new solutions. Some are closed, cmn-
pleted series belonging to the past.

fu mathematical usage a series is the indicated sum of a set of
terms, but a sequence is any ordered set of quantities like the
positive integers.' A series therefore implies a closed grouping,

3 Mathematics Dictionary, ed. Glenn James and R. C. James (Princeton, 1959),
pp. 349-50. Professor Oystein Ore, of Yale University, to whom I showed this
chapter after he told me about the work he is completing on graph theory, wrote
the following assessment: "In attempting to give a systematic presentation of so
~.
.. '..
-~~j---
THE CLASSING OF THINGS
34
It'
and a sequence suggests an open-ended, expanding class. The !

1
mathematical distinction is worth keeping in t!Jis discussion.
In general, formal sequences exceed the ability of any indi,
vidual to exhaust their possibilities. An occasional person, born
by chance into a favoring time, may contribute beyond the usual
measure of a single life-span, but he cannot alone simulate in his
life the corporate activity of a whole artistic tradition.
The mathematical analogy for our study is topology, the
geometry of relationships without magnitudes or dimensions,
having only surfaces and directions. The biological analogy is
speciation, where form is manifested by a large number of individuals undergoing genetic changes.
Where are the bonndaries of a formal sequence1 Because his.
complex a subject matter one would be inclined, as in the natural sciences, to

look to the mathematicians for some pattern to serve as a descriptive principle.
The mathematical concepts of series and sequences came to mind but after some
thought these appear to be too special for the problem at hand, However, the
less known field of networks or directed graphs seems to be considerably more
suitable.
"We are concerned with the variety of stages in the creativity of the human
race. From one stage one moves to another in the development, There are a
variety of directions which may be selected. Some represent actual happenings.
Others are only possible steps among many available ones. Similarly each stage
may have occurred among several possible steps leading to the same result.
"This one may picture in a general way by the mathematical concept of a
j )(
_/ direct.ed graph ~r network. Such a graph consists of a number
~/ ~ /
ofpomts or vertlCCS or stages. Some of these are connected by
a directed line, an edge or a step. At each stage there is thereof
fore a number of alternative edges which may be followed, and
/
also a number of incoming edges from which this stage could
/
have resulted. The actual development corresponds to a
J~
(directed) path in the graph and it is only one among many
't
possible ones.
"One may ask whether the graphs we should like to consider are of a special
type among the many directed graphs which can be constructed. There seems
to be one essential restriction, that the graphs shall be acyclic, that is, there exists
no cyclic directed path returning to its original stage. This essentially corresponds
to the observation about human progress that it never returns to the previous
conditions."
FORMAL SEQUENCES
35
wry is unfrnishcd business, the boundaries of its divisions continually move, and will continue to move for as long as men
make history. T. S. Eliot was perhaps the first to note this relationship when he observed that every major work of art forces
upon us a reassessment of all previous works. 4 Thus the advent
of Rodin alters the transmitted identity of Michelangelo by enlarging our understanding of sculpture and permitting us a new
objective vision oflus work.s
For our purposes here, the boundaries of a sequence are marked
out by the linked solutions describing early and late stages of
effort upon a problem. With a sequence having many stages,
there was a time when it had fewer. More new ones ntay be
added in the future. The sequence can continue only when the
problem is given greater scope by new needs. As the problem
expands, both the sequence and its early portions lengthen.
Open and closed sequences. When problems cease to command
active attention as deserving of new solutions, the sequence of
solutions is stable during the period of inaction. But any past
problem is capable of reactivation under new conditions. Aboriginal Australian bark-painting is an open sequence in the twentieth
century, because its possibilities are still being expanded by living
artists, but Greek vase-painting is an arrested sequence (p. 109)
because the modern painter needed to renew his art at "primitive" sources rather than among the images of the Hellenic
world. The transparent animals and humans of Australian painting, and the rhythmic figures of African tribal sculpture correspond more closely to contemporary theories of reality than
to the opaque and unequivocal body forms of Greek art.
The method imposed by such considerations is analytical and
4 T. S, Eliot, "Tradition and the Individual Talent," Selected Essays, 1917-32
(New York, 1932), p. 5 Also, Points of View (London, 1941), pp. 25-26.
5 Andre Malraux has appropriated the "Eliot effect" in several passages of The
Voices of Silence (New York, 1954), pp. 67, 317, 367, where major artists are
represented as altering their respective traditions retroactively by their own
novel contributions.
'
~ ~"mo on-r:~ ow~=
divisive rather than synthetic. It discards any idea of regular!

cyclical happening on the pattern of "necessary" stylistic serie1'
by the biological metaphor of archaic, classic, and baroque stage,.:
Sequence classing stresses the internal coherence of events, all'
while it shows the sporadic, unpredictable, and irregular natu1,
of their occurrence. The field of history contains mmy circuits
which never close. The presence of the conditions for an event
does not guarantee the occnrrence of that event in a domain
where man can conte1nplate an action without cmnmitting it.
Simple biographical narration in the history of art tends to
display the entire historical situation in terms of an individual's
development. Such biography is a necessary stage of recon.
struction, but a formal sequence designates chains oflinl<ed event!
by an analysis which requires us to do the opposite: to perceive
the individual in terms of his situation.
Sequence classing allows us to bridge the gap between biog.
raphy and the history of style with a conception less protean than
biological or dialectic theories of the dynamics of style, and more
powerfully descriptive than biography. Its dangers and its limitations will be readily apparent. In the long run, the conception
of a sequence may serve as a scaffolding which it may be con-
"
enerations of architects struggled to coordinate the regular seguence of the vaulted nave bays with the great weights of the
fa~ade towers. These had to rest in part upon nave supports
ideally no thicker one than any other. A solution gradually was
perfected, by thickening the bearing walls beneath the tower
periphery, by augmenting the buttressing, and by sacrificing excessive slenderness in the proportioning of the nave supports.'
At Mantes Cathedral the west fa~ade was among the earliest
to show this perfected solution by compromise. The architect
wanted a uniform rhythm of equivalent supports under evenly

distributed light. He wanted a formal value toward which technical solutions could help him. The volume of the interior is
uninterrupted, yet the mass of the fa~ade is immense. Both abjectives have been brought to a close fit, however inconsistent
they may have seemed in the first place. This solution in Gothic
cathedral architecture implicated a cluster of subordinate traits
such as colunrns, buttresses, and windows, whose further alteration was governed by the fa~ade solution. The fa~ade solution
was itself subordinate to another system of changes respecting
the composition of the towers.
venient to discard later on, after it has given access to previously
Hence "cathedrals" are not a true form-class but an ecclesiastical category and an administrative conception in canon law.
invisible portions of the historical ediftce.

It is disturbing to those who value the individuality of a thing
to have that individuality diminished by classiftcations and gen
eralizations. We are caught between difficulties: single things
Among the ediftces thus designated, there is a handful of closely

related designs built in norrl1ern Europe between II40 and 1350.
These are part of a sequence of forms that also includes some
abbeys and some parish churches. The formal sequence is not
are extremely complicated entities, so complicated that we can
"cathedrals." It is more like "segmented structures with rib
pretend to understand them only by generalizing about them

(p. 96). One way out is frankly to accept the complexity of
single things. Once their difficulty is conceded, it is possible to
vaults," and it excludes barrel-vaulted catheclrals.

The closest defmition of a formal sequence that we now can
venture is to affirm it as a historical network of gradually altered
find aspects that can be used in comparisons. No such trait now
repetitions of the same trait. The sequence might therefore be
known is unitary or fundamental: every trait of a thing is both
of
described as having an armature. In cross section let us say that

it shows a network, a mesh, or a cluster of subordinate traits;
An example from French Gothic cathedral architecture: several
6. Hans Kunze, Das Fassadenproblem der franzOsischcn Friift-und Hochgotik (Leipzig, 1912; dissertation, Strassburg).
a cluster of subordinate traits as well as a subordinate part

another cluster.
F"
''
THE CLASSING OF THINGI!
~lUME O:BJECTS AND REPLICATIONS
39
and in long section that it has a fiber-like structure of temporal\ e impervious. A fashion is the projection of a single image of
stages, all recognizably similar, yet altering in their mesh fro 111f ~utward being, resistant to change during its brief life, ephemerbeginning to end.
al, expendable, receptive only to ~opying but ~tot to f~mdamental
Two questions immediately arise: in the first place, are foru~[. variation. Fashions touch the litmt of credtbihty by vtolatmg the
sequences not indefmitely numerouS? No, because each cor.: precedent and by grazing the edge of the ridiculous. T~ey belong
responds to a conscious problem requiring the serious attentio11
of many persons for its successful resolution, There are no linkeq

solutions without there having been a corresponding problent
There is no proble1n where there is no awareness. The contoui!
of human activity as a whole are therefore congruent with tho~

of the totality of formal sequences. Each class of forms consists of
a real difficulty and of real solutions. In the course of time, mos1
of these solutions may have been destroyed, but that difficulty
is only an apparent one, for our determinations of sequence ca~.1
ot to a connected chain of solutions, but they consutute, each
~asbian in turn,
classes of only one member each. A fashion is a

duration without substantial change: an apparition, a flicker, forotten with the round of the seasons. It is like a class, but it
~iffers from
a sequence by having no appreciable dimension in
time.
PRIME OBJECTS AND REPLICATIONS
if necessary be founded upon only one surviving solution 01
If tools on the one haud, and fashions on the other mark
example. Such determinations are of course provisional and in.

complete. Yet every object attests to the existence of a require.
out our provisional boundaries, we need now to set further
ment for which it is the solution, even when that object is only a
late copy in a long series of coarsened products far removed frmn
the clarity and sharpness of an original.
In the second place, are we going to consider all man-made
objects or only a selection of them1 Where is the minimal
boundary I We arc concerned mainly with works of art rather
as well as the spectator's and the artist's views of the situation of
than with tools, and we are interested more in long durations
than in brief ones, which tell us less about our subject. Tools and
instruments commonly have extremely long durations. Upon
occasion these extend so far that it is difficult to note great
changes, as, for example, in the minor inflections that record the
passage of civilizations in the cooking pots of a deep refuse

midden. A working rule is that the simpler tools record very
large durations, and that the more complicated tools record brief
episodes of special needs and inventions.
Fashions, The minimal boundary perhaps lies near the limits
of fashion. Fashions in dress are among our briefest durations.
A fashion obeys special demands to which the longer evolutiom
divisions inside the domain. There are prime objects and replicas
the work of art in time.
Prime objects and replications denote principal inventions,
and the entire system of replicas, reproductions, copies, reductions, transfers, and derivations, floating in the wake of an
important work of art. The replica-mass resembles certain habits

of popular speech, as when a phrase spoken upon the stage or in
a f1lm, and repeated in millions of utterances, becomes a part of
the language of a generation and finally a dated cliche.

It will be useful first to examine the nature of the prime object
from which a replica-mass derives. Prime objects resemble the
prime numbers of mathematics because no conclusive rule is
known to govern the appearance of either, although such a rule

may someday be found. The two phenomena now escape regulation. Prime numbers have no divisors other than themselves
and unity; prime objects likewise resist decomposition in being

original entities. Their character as primes is not explained by
their antecedents, and their order in history is enigmatic.
40
THE CLASSING OF THltiGs
pRI.ME OBJECTS AND REPLICATIONS
The annals of art, like those of bravery, directly record only aj

handful of the many great moments that have occurred. When
we consider the class of these great moments, we are usually co11.~!
fronted witl1 dead stars. Even their light has ceased to reach tts.!
We know of their existence only indirectly, by their perturb,,!
tions, and by the immense detritus of derivative stuff left in their;
paths. We shall never know the names of the painters ofBonatn_,
pak, nor those of Ajanta: indeed, the wall paintings at Bonampak
and Ajanta, like the Etruscan tomb murals, are probably only'
pale reflections of a lost art that gracecl the more urban halls of'
living princes. The history of art in this sense resembles a broken,
but much-repaired chain made of string and wire to connect the,
occasional jeweled links surviving as physical evidences of the
invisible original sequence of prime objects.
Mutants. Although biological metaphors are avoided throughout this essay, their occasional use for clarifying a difficult di~
tinction is justified when we are talking about prime objects. A
prime object differs from an ordinary object much as the individual bearer of a mutant gene differs from the standard example
of that species. The mutant gene may be infinitesimally small but'
the behavioral differences which it occasions can be very great
indeed.
In addition, the idea of a prime object requires a fundamental
adjustment in our ideas of the integrity and unity of the work of
art. The mutant fraction imposes consequences upon the progeny
of the thing. But altogether different is the field of action assumed for the whole object. These differences are of the same
order as between an act of procreation and an act of moral ex-
ample. A possibility for change appears with the mutant-bearing

prime object, while a generally beautiful or distasteful object
merely calls for ritual repetition or avoidance.
Our interest therefore centers upon minute portions of things
rather than upon the whole mosaic of traits that constitutes any
object. The effect of the mutant fraction, or prime trait, is dy-
4'
natnlc in provoking change while that of the whole

. .object is
simply exemplary, exciting feelings of approval or dtslike more
than any active study of new possibilities.
.
Diagnostic dijficulties. Strictly considered, a form-class extsts
only as an idea. It is incompletely manifested by prime objects,
or things of great generating power, in the category of the
Parthenon, or of the portal statues at Reims, or of th~ frescoes
by Raphael in the Vatican. Their physical presence ts always

dimmed by the accidents of time, but their prime status ts unquestionable. It is guaranteed by direct comparisons with other
things of lesser quality, and by a vanety of testlmomals from
artists in many generations. Yet the Parthenon 1s bmlt upon an
archaic formula surviving into Periclean time. ~he portal stat~es
of Reims e1nbracc the work of several generatmns, and all, m-
duding the frescoes by Raphael, have undergone damaging wear

and diminution. In the biologists' language, they are all three
phenotypes, from which we have to deduce the originating genotypes.
These three examples, however, are extremely special ones
illustrating the phenomenon of tl1e climactic entrance. Such entrances occur at moments when the combinations and pennutacions of a game are all in evidence to the artist; at a n1oment when
enough of the game has been played for him to behold its full
potential; at a moment before he is constrained b~ the exhaustmn
of the possibilities of the game to adopt any of 1ts extreme terminal positions.
Every stage of the game, whether early or late, contains
prime objects variously qualified according to their entrances.
But the number of surviving prime objects is astonishingly
small: it is now gathered in the museums of the world and in a
few private collections; and it includes a large proportion of
celebrated buildings. It is likely that buildings constitute the
majority of our prime objects, being immobile and often indestructible objects. It is also likely that a large proportion of
rc
42
THB CLASSING OF
THINGs~~~
prime objects was made of perishable substances like cloth and

paper, and that another large proportion was made of precious
metal and melted down when needed.
The classic example is the gigantic statue of Athena Parthenos
made by Phidi... in gold and ivory for the Parthenon, known
only by mean replicas made for pilgrims and tourists. Another
example: many splendors of the fourth-century world under
Constantine the Great are known to us only by the terse and
gray descriptions of the "Bordeaux pilgrim" who traveled
through southern Europe and North Africa in A.D. 333. 7 This
unique tourist's chronicle is a part of the replica-mass of the Early
Christian world, and its accidental survival has justifred the devoted labors of several generations of scholars in establishing its '
text and annotating its contents. Still another example of the
difference between prime objects and replica-masses is the daily
crossword puzzle. The manuscript draft by the puzzle-maker is
a prime object (which no one conserves); all the solutions in
subways and on the desks of people who "kill time" compose
the replica-mass.
Since a formal sequence can be deduced only from things, our
knowledge of it depends upon prime objects and their replicas.
But the number of prime objects is distressingly small, and as
most of our evidence consists of copies or other derivative things,
these inferior expressions, which often are very far removed
from the original impress of the responsible mind, therefore must

occupy much of the historian's time.
An important question arises at once. If, in a given sequence
an initial prilne object begins the connected series, why are the
subsequent prime objects of the series not to be regarded as

replicaSI The question is particularly urgent when we recall that
prime objects surely known to arise from prime traits are very
few. In many places and periods it is impossible to identify them

among the accumulated collections of replicas. The question ex-
7. 0, Cuntz, ltineraria romana (Leipzig, 1929), vol.
I.
pRIME OBJECTS AND REPLICATIONS
43
tends in all directions: are we ever in the unmistakable presence

of an initial prime object< Can such an entity be isolated< Have
prime objects any real existence< Or are we simply conferring
upon some leading examples of their class an additional symbolic
distinction of ilnaginary priority? The questions caution us
against misplaced concreteness, but they should not weaken the

main thesis. Man-made objects of all kinds correspond to human
intentions in historical sequence. Prime objects correspond to
prime traits, or to mutant intentions, while replicas merely
multiply the prime objects. Though its type is extremely traditional, the Parthenon is recognizable as prime by many refinements lacking in other temples of its series. But the copies of the
Athena Parthenos statue in the National Museum at Athens,
or the Strangford shield in the British Museum, only coarsen and
reduce the original without increments of any kind.
Many sorts of replicas reproduce the prime object so completely that the most sensitive historical method cannot separate
them. In another kind of seriation, each replica differs slightly
from all the preceding ones. These accumulated variations may
originate without design, merely for relief from monotonous
repetition. In time their drift is perceived and brought to order
by an artist, who imposes upon the mass of replicas a new scheme

manifested by a prime object not categorically different from
the preceding prime object, yet historically different in that it
corresponds to a different age of the formal sequence to which
the prime objects both belong. Neither of these models of historical happening makes the discovery of prime objects easy:
we can maintain only that they must have existed in numbers
corresponding to the critical moments of change in all the classes

of sequence. They may have existed ouly as random notes or
sketches. Their flfSt full appearances may be indistinguishable
in many instances from the immediately subsequent replicas.
The shelves of every museum of archaeology are organized to
display this conception of the sequence of artifacts. One group
of replicas arranged by type adjoins another group on another
~""'"'"'
rI . .
o rn,,.
"
leaded fences between color areas, is a case illustrating the manner

in which an entire older tradition can become a point of deparrure when technical novelties require its reactivation. For long
intervening periods a formal sequence may nevertheless seem inactive, simply because the technical conditions for its revival are
chapter; here it is hnportant to bring out once n1ore the elusive l
not yet present. Such inactive classes deserve to be ranked in
llature of prime objects. Signatures and dates inscribed up 011

works of art by their authors in no way assure us that they are i
prime. Most works of art, moreover, are anonymous, and they
fall naturally into brge groups. Under most circumstances the ;
prime objects have disappeared into the mass of replicas, where
their discovery is most diffictut and problematic, akin to the!
greater difficulty of discovering the first recognizable examples;
of the biological species. In reality our knowledge of sequences.
is for the most part based upo11 replicas.
Our disti11ction between prime objects and replicas also illustrates a capital difference between European and non-European;
arts. With European objects we often can approach closer to the,
order of apparent completion, that is, by the length of time since

any innovation enlarged the class. By this criterion all the classes
of form are still open sequences, and it is only by an artificial
convention that we may call any class a historically closed series.
Serial appreciation. A pleasure shared by artists, collectors, and
historians alike is the discovery that an old and interesting work
of art is not unique, but that its type exists in a variety of examples
spread early and late in time, as well as high and low upon a
scale of quality, in versions which are antetypes and derivatives,
originals and copies, tramformations and variants. Much of our
hot moment of invention than in non-European ones, where!
completion in the presence of a shape in time.
our knowledge is so often based orUy upon replicas of uniform i

or debased quality. A long tradition of collecting and connoisseur.:
ship appears ocly among Chinese, Japanese, aud European:
peoples; elsewhere in the world the continual accumulation of.
things was never systematically ordered by the efforts of col-
lectors and critics, so that the prime objects virtually all have'
been lost from view.
No formal sequence is ever really closed out by the exhaustion
of all its possibilities in a connected series of solutions. The re,
validation of old problems in new circumstances is always possible and sometimes actual, like the renewal of stained-glass
technique as the gemmeau glass newly invented in France since
the war.' The use of fracture to modulate the light rather than
As the linked solutions accumulate, the contours of a quest by

several persons are disclosed, a quest in search of forms enlarging
the domain of aesthetic discourse. That domain concenlS affective
states of being, and its true boundaries are rarely if ever disclosed
by objects or pictures or buildings taken in isolation. The continuum of connected effort makes the single work more pleasurable and more intelligible than in isolation. Rebecca West perceived this rule of appreciation by series when she declared in
The Strange Necessity that the main justification for much ordinary literature is the stimulation which it provides for critics.
Appreciation of the series nevertheless runs counter to the
i
i
'
~ =~""''
shelf. The two masses are alike yet different. They are on good
evidence assigned to different periods. They correspond either to
different ages of the same sequence or to its different region~
varieties.
1
We shall revert to these questions at more length in the next)
''
"''=
8. W. Pach, "The Gemmeaux of Jean Crotti: A Pioneer Art Form," Magazine

of Art, 40 (1947), 68-69. The novelty consists in laminating different colors of'
glass without leading. The glass sandwich then is cracked for graduated scintillation.
satisfaction in these circumstances arises from the contetnplation
of a formal sequence, from an intuitive sense of enlargement and
main drift of modern literary criticism, which has been set since
about
1920
against the "intentional fallacy,"10 or judgment by
9 R. West, The Strange Necessity (New York, 1928), esp. "The Long Chain of
Criticism,''
10. W. K. Wimsatt, Jr. and M. C. Beardsley, "The Intentional Fallacy," The
Sewanee Review, 54 (1946), 468-88.
46
THB CLASSING OF
setting instea,d
THINGSrp:I~B
~fby_intrinsic merit. Here the "new critics" held I
that the poets mtent10n does not extenuate Ius performance, and \
OBJECTS AND RBPLICATIONS
quences wh?se territorial
47
defi~ition _we are seeking. A trivial
case is the lustory of buttons: 1t 1s tnv1al because there are few
events in the history of the button; there are only variants in
historical and biographical conditions.

!
A literary work, however, consists only of verbal meanings: I
shape, size, and decoration; there is no duration in respect to

difficulties encountered and overcotne. An exa1nplc of excessive
such principles of criticism would fail of application in the visual

arts, where verbal symbols are incidental, and where an elementary problem arises. It is a familiar problem in literature,
which experts usually have settled before the poem or the play
reaches the student. This is the matter of "establishing the text,"
of compiling and comparing all the versions and variant readings
to find the clearest, most consistent sequence. The study of the
history of things is actually only at this stage of"establishing the
text," of discovering the "scripts" of the principal themes.
An unmistakable erosion wears down the contours of every
work of art, botl1 in its physical form, which is gradually ob!iterated by dirt and wear, and by the disappearance of so many
steps in the artist's elaboration of his conceptions. Often he makes
no record whatever, and we can only conjecture what the stages
of design may have been. If there are diagrams, sketches, and
drawings, their number is always small because the artist was
more generous than today in the direction of the wastebasket,
and he did not hoard the chips and shavings of his day's labors
for the art market. In Greek and Roman sculpture, this kind of
erosion has eliminated virtually all traces of the individual's
factuality is the textbook on the whole history of world art,

which must embrace all the principal events from paleolithic
painting to the present. It can set forth only some names and
dates, and some general principles for the understanding of works
of art in their historical setting. It is a work of reference, and it
cannot relate or resolve problematic matters.
Still other historical themes describe convergent happenings
rather than linked events: an example is the conventional discussion of paleolithic cave painting together with South African
Bushman rock paintings of recent date, made probably after
A.D. 1700. There is no demonstrable link between the two
groups, however much they may look alike. 111ey are sixteen
thousand years apart without connecting historical tissue. Indeed,
paleolithic cave painting cannot be shown to connect with anything until its discovery in the nineteenth century. Its own internal history is still vague, lacking detailed articulation by
periods and groupings. Both paleolithic and Bushman paintings
can profitably be considered, nevertheless, as elements of a formal
sequence including contemporary art, together with the efforts
of various individuals to assimilate and transform the prehistoric
processes of elaboration: we usually have only replicas and poor
"tradition" which entered the stream of modern consciousness
copies to give us an abraded and coarsened conception of remote

originals altogether lost to view, "Establishing the text" under
these circumstances is a laborious necessity of study, and the editors who perform it deserve more reward than they get. Without it we should have no sequence in time, no measure of the
distances between versions, and no conception whatever of the
authority and power of the lost originals.
only late in the nineteenth century.

Technical renewals. When trying to understand the composition of formal sequences, we may with profit look briefly into
the question of craft techniques. The name of art is itself close to
technical cum1ing. The devices and artifices of planned illusion
are the working gear of the artist who is engaged in substituting
his own codes of time and space for our older and less interesting
that all criticism must be within the poem itself regardless of its
Some themes are trivial; others are excessively factual; but
somewhere between triviality and factuality lie the formal se-
ones.
Among craftsmen a technical innovation can often become the
=W" "' >m,.f"CBOBJECTS

AND
rniM
REPLICATIONS
49
point of departure for a new sequence where all the elements of four-part rib-vault construction in twelfth-century France."
the tradition are revised in the light of the possibilities opened to j '[hey too are examples of the stunted sequence. The chances of
view by the iunovation. An example is the displacement of : uch retarded classes being called into renewed being are unpreblack-fignred vase painting near the end of the sixth century j ~ictable, although obscure techuical failures have sometimes been
B.c. by red-figured technique." This amounted to a reversal of ! revived for further development after long periods of oblivion,
figure and ground in order to favor the figure and to convert especially in the !~story of science.
.
.
the ground from a decorative setting into an atmospheric dis. i The linked senes of solutions composmg a sequence 1s not
tance.
llecessarUy restricted to a single craft. On the contrary, it is more
The technical change in the potter's firing habits may of course likely to appear when different crafts come into play at the same
have been brought into being by the painter's specific demands time. Thus Greek vase painters probably took many suggestions
for such a renovation of the conditions of the craft, but the frotn the pictorial achievements of the wall painters, which in
probability is that the "new" technical habit was available long turn (at least in Etruscan tomb painting of Greek style) may have
before an artist seized upon it for his needs. To this general borrowed certain schemes from vase painting, such as procestopic-of invention in relation to change-we must return later; sional profile figures.
here, however, it is useful to show how one sequence may yield
The formal sequence thus may find its realization in several
to another when an item in the composition of the original se- i crafts simultaneously. An example is afforded by the abrupt conquence is significantly altered. We have chosen a technical ex- trasts of light and shadow in seventeenth-century chiaroscuro
ample in the history of vase-painting; other case histories might composition, allowing novel illusions of depth and movement.
be taken from innovations in the painter's subject tnatter, or in This uew organization of the surfaces rapidly spread throughout
expressive attitude, or in perspective convention. The point is : all the visual arts. No province of Europe escaped the dominion
that the formal sequence always corresponds to a distinct conof these forms: the contagion spread from city to court, or
ception of potential series of changes.
from court to city, as in Holland, where there were only cities,
Conversely, be it noted that many techuological innovations : and from thence to every cranny of the social structure, exemptprovoke no immediate development. Heron's aeolipile of the ing only the most isolated communities, or those too poor to
first century A.n.t2 was an oddity without consequences for renew their churches, houses, and pictures.
seventeen centuries until the sustaining economic, sociological, The invisible chain. An ancient tradition of representation
and mechanical conditions for the development of steam en- shows us the poet inspired by the muse.14 His posture with lifted
gines were at hand. The example points to abortive, retarded, pen betrays the greater presence as he receives the message from
or stunted sequences of which occurrences can also be identified another sphere of being. His whole body strains upward and the
in the arts. Henri Focillon used to speak of the "failures that lurk folds of his clothing flutter upon the breath of the spirit. The
in the shadow of every success" when he described the oddities . best-known versions show the Evangelists receiving the Gospels;
-such as eight-part vaults-strewn along the definitive line of
n. G. M. A. Richter, Attic Red-Figured Vases (New Haven, 1946), pp. 46-so.
12. A. G. Drachmann, Ktesibios, Philon and Heron. A Study in Ancient Pneumatic~
(Copenhagen, 1948).
IJ. H. Focillon, Art d'Occident (Paris, 1938), p. 188. S. Quiriace at Provins:

"une de ces experiences sans lendemain."
14. A. M. Friend, Jr., "The Portraits of the Evangelists in Greek and Latin
Manuscripts," Art Studies, 5 (1927), ns-so; 7 (1929), 3-29.
so
i:
'
I
I
'.:
THE CLASSING OF
others portray Church fathers or saintly scholars. The

of divine inspiration sometimes also appears in tl1e
tions of painters engaged upon sacred images: Saint Luke,
depicts the Virgin" in the altarpiece by Roger van der
appears as a Flemish burgher, suffused by the light of the
and poised with raised stylus like the Evangelists of
illuminations.
51
pR!r.ID OBJECTS AND REPLICATIONS
d he is obsessed with the urgency of its realization, in a solitary
~sture of intense effort, traditionally represented by tlle figures
f the poet or the muse.

Prior events and future possibilities within the sequence: these
dimensions govern the position of every work of art. The notebook of Villard de Honnecourt, the tllirteenth-century architect,
ontaius a sketch of one of the towers of Laon Cathedral, and
Like the Evangelists, like Saint Luke, the artist is not a : ~tnder it he wrote, "Nowhere have I ever seen a tower like this."
agent obeying only his own will. His situation is rigidly
; "The itinerant master-builder was not only praising the work of
by a chain of prior events. The chain is invisible to him,
j a predecessor; he was also challenging it, as if to say "this good
limits his motion. l-Ie is not aware of it as a chain, but only ! way has been done, and I can improve it."t6
vis a tergo, as tl1e force of events behind him. The conditions 1 Such possibilities seem actively to possess the people who
posed by these prior events require of him either that he
i explore them. Vasari recorded the obsession of Paolo Uccello
obediently in tlle path of tradition, or that he rebel against ! with the perspective construction of painted surfaces.17 Much
tradition. In either case, his decision is not a free one: it is .! ofC6zanne's work tells us of his long obsession with the ideal
tated by prior events of which he senses only dimly and landscape through its prior realizations in Romano-Campanian
rectly tlle overpowering urgency, and by his own
painting and in tlle art of Poussin, whose canvases Ccza1me
peculiarities of temperament.
, studied in the Louvre." Every major artist betrays such an obPrior events are more significant than temperament: the hir; session: its traces are legible in works by men whose historical
tory of art abounds in examples of misplaced temperamen~l identity has otherwise vanished, like the Maya architects and
like the romantics wrongly born in periods requiring classk sculptors at Palenque or Uxmal.
measure, or the innovators living in periods governed by rigil, Probably the individual is immunized against other unrelated
rule. Prior events exercise a selective action upon the spectrum: interests by his obsession, in the sense that he does not often
0
of temperaments, and each age has shaped a special temperamem. make major contributions in more than one formal sequence exto its own uses both in thought and in action. Among artis~ cept uuder special conditions. Such conditions occur at the end
tl1e prior events that determine the individual's actions constitu~ of a series of related forms, when the person privileged to make
the formal sequences we have been discussing. They are th1' a terminal statement must then shift his labors to another class
events composing the history of the quest that most closely cou of forms. Examples of this kind of shift commonly are discerns tlle individual. His position in that quest is a position lw guised itt biographical writing as different periods in one man's
cannot alter, but only realize. The tlleme of possession by th1.
work in hand is evident in many artistic biographies: the inJl: 16. H. R. Hahnloscr, Villard de Honnecourt (Vienna, 1935), pl. 19 and pp. 49vidual is driven in every action by forces of an intensity absent' jo. '']'a/ este en m[u]lt de tleres, si co[m] v[osJ pores trover en cest liv[r]e; etl aucun
from other lives; he is possessed by his vision of the possible.llll, onq[ue]s tel.tor nevi co[m] est cell de Loo[n]."
15. D. Klein, St. Lukas als Maler der Maria (Berlin, 1933).
17. G. Vasan, Lives, trans. G. de Vere (London, 1910-12), 2, IJI-40:

"!pcctive . . , kept him ever poor and depressed up to his death" [1475].
IS. Gertrude Berthold, Ctfzanne und die a/ten Meister (Stuttgart, 1958).
"
per-
52
THE CLASSING OF Tlllli!cs
sJIRIAL POSITION, AGE, AND CHANGE
work, corresponding to old and new circuits of influence. But

changes of period also concern sequences in different stages of
development, among which the individual is seeking the viable 1
choice, where he can still challenge the past with a notable iu1
.1
provement.
Solitary and gregarious artists. In this game of jockeying for

position-and it goes on in every generation-there are further
elements of variation afforded by the play of temperaments and
endowments. Guercino's change c.r621-23 fro1n baroque to
53
nd it is of a thinner quality. His non-professional contemporary

admirers are likely to know him in his person more than in his
work.
Usually the entire range and bearing of such a career can be
brought into focus only long after death, when we can place it
in relation to preceding and subsequent events. But by then the
shock of the innovation has faded. We may tell ourselves that
these pictures or buildings once broke with the .tradition, But in
our present they have entered the tradition as if by simple
chronological distance.
Probably all important artists belong to this functionally lonely
class. Only occasionally docs the artist appear as a rebel, as in
the sixteenth and in the nineteenth centuries. More commonly
he has been a courtier, a part of the household of the prince, an
entertainer, whose work was valued like that of any other en-
classic form was intended to satisfy his clients, and it displays an 1

interaction between the painter and his surroundings much more
than it relates to any m'\ior shift of formal substructure, as Denis
Mahon has shown in his Studies in Seicento Art and Theory (1947),
Some sequences require contributions from n1any different
kinds of sensibility. The presence of pairs of great rivals engaged
in contrasting ways upon the same proble1ns at the same time tertainer, and whose function was to amuse more than to disusually defines such situations: Poussin and Rubens, Bernini and I quiet the audience.
Today the artist is neither a rebel nor an entertainer. To be a
Borromini thus n1arked out the wide roads of seventeenth-cen. ~
tury painting and architecture. The contemporary revalidation rebel requires more effort away from his work than the artist
~f priniltive experience also has be~n carried by contrasted pain, wants to make. The entertainers have formed professional guilds
hke Ehot and Joyce or !Gee and Ptcasso. Such pans and groupr 1 in those many categories of public amusement from which the
emerge upon the scene of only the most ample investigations, 1 artist is now almost completely excluded. Only the playwright
and not in the narrowing galleries of more restricted intcrests still functions both as an artist and as an entertainer. More lonely
where we find Uccello and Cezamre, less as rivals than as isolated i than ever, the artist today is like Dedalus, the strange artificer of
prospectors alone with their obsessions.
i wonderfnl and frightening surprises for his immediate circle.
Each great branch of art calls upon a different temperament
Painting and poetry more than all others invite the solitary,
'I'
nature. Architecture and music summon the gregarious man to
whom work in company is a requirement of his nature best

satisfied by the division of labor and by concerted entrances.,
I
The outstanding innovator among artists, like Caravaggio,;
nevertheless is functionally lonely. His break with tradition may j
or may not be known to the multitude, but he is himself ofi
necessity aware of the isolation it brings. The work of the asso-j
dates and pupils immediately around him derives from his own,i
SERIAL POSITION, AGE, AND CHANGE
Iu mathematical usage (p. 33), series and sequence differ

as closed and open classes of events. In the foregoing pages we
assumed that most classes are open-ended sequences. Here, however, we will assume that most classes can be treated as closed
series. The difference between the two points of view depends

upon the viewer's position, whether he wishes to be inside or out-
54
THE CLASSING OF THING~
SJ!lll.AL POSITION, AGE, AND CHANGE
ss
side the events in question. From the inside, most classes look like rule of minimal durations is even more rigorous, marked by
open sequences; from outside they seem to be closed series. Ill collective attitudes of acceptance or rejection, which we shall
order to reconcile both positions, let us say that the conception of discuss later. These cannot be short-circuited by fortuitous disthe formal sequence outlined in the preceding section allows us to coveries as in the technological fields. For example, the discovery
assemble the ideas of things, with their first realizations and with of a lununous paint was no help to the painter engaged in an
the consequent mass of replicas, as events into connected finite attempt to record on canvas the play of light in nature: he was
bound to aclueve his unconventional aim with conventional
series.
The rule of series. Every succession may be stated in the fol. materials.
Our procedure is rather to recognize the recurrence of a need
lowing propositions: (1) in the course of an irreversible finite
series the use of any position reduces the number of remaining in differing stages of its gratification, and the persistence of a
positions; (2) each position in a series affords only a limited problem throughout various efforts to solve it. Every need evokes
number of possibilities of action; (3) the choice of an action a problem. The juncture of each need with successive solutions
commits the corresponding position; (4) taking a position both leads to the conception of sequence. It is a conception much
defines and reduces the range of possibilities in the succeeding, 11arrower yet more labile than that of any biological metaphor,
position.
[ for it considers only human needs and their satisfaction, in a oneStated differently: every new form limits the succeeding inno.l to-one correspondence between needs and things, without the
vations in the same series. Every such form is itself one of aI intermediary of any other irrelevant entity like "life-cycle." The
fmite number of possibilities open in any temporal situation.! main difficulty arises in the specification of uneeds," but we have
Hence every innovation reduces the duration of its class. The! carefully sidestepped that question by restricting the discusdon
boundaries of a class are fixed by the presence of a problem re-I to relationships rather than to magnitudes.
quiring linked solutions: classes may be small or large: we ate 1 Systematic age. We need here to study the nature of durations.
here concerned only with their internal relationships and not j To speak of sequences or series, that is~ of specified needs and
with their dimensions or magnitudes.
I their successive stages of satisfaction, is to mark a variety of duraOne more proposition allows us to qnalify the mode of dura- [ tions. No duration, however, can be discussed save in respect to
tion. Each series, originating in its own class of fonns, has its its beginning, middle, and end, or to its early and its late moown minimal duration for each position, depending upon the I ments. In one duration, we are agreed that "late" cannot precede
effort required. Small problems require small effort; large ones I "early." Hence we may speak of the systematic age of each item
demand more effort and so consume more time. Any effort to\ in a formal series according to its position in the duration.
Easily recognized visual properties mark the systematic age of
shortcut the circuit leads to fail~re. The rule of series requires,,
any
item, once we have identified its series. It is not the purpose
each posmon to be occupied for 1ts corresponding period before,
the next position can be taken. In purely technological domains i of this book to dwell upon the techniques and kinds of chronothis is self-evident: the steam engine was invented before the i logical discrimination, yet certain fundamental observations are
locomotive, but the mise au point of a locomotive required many' indispensable. Early solutions (promorphic) are technically
more parts, each consuming its portion in the economy of the simple, energetically inexpensive, expressively clear. Late solutime sequence, than a steam engine alone. In works of art the tions (neomorphic) are costly, difficult, intricate, recondite, and
'THE CLASSING OF TB:I~GS
SBttiAL POSITION, AGE, AND CHANGE
animated. Early solutions are integral in relatio11 to the proble111

they resolve. Late ones are partial in being addressed more to
the details of function or expression than to the totality of the
same problem. Our terms, promorphs and neomorphs, have cl!e
permission of the best dictionaries, and they avoid ilie promiscu.
! '
57
rrence systematic age differentiates things as sharply as any other

historical or geographical differences.
The presence of such a systematic age for the entire complex
of manufactures in a civilization has long been obscured by
several blinds. The temptation to interpret social processes from
ous conta1nination of terms such as ''prhnitive'' and ''decadent," potsherds and broken stones has also been irresistible, and we are
and "archaic" or "baroque." But these determinations depend presented with cycles of political revolution, based on evidence
upon defining the pertinent form-class, for otherwise the visual which really speal<S best of other things. Potsherds and broken
properties oflate solutions in one class may deceptively resemble stones are classes of effort that reflect political life from vast disthose of early solutions in another class. Late and early are per. tances, no more strongly than when we fainrly hear the dynastic
force relative to a defined starting point. They will h1evitably conflicts of medieval France in Proven1:al poetry.
remain as parameters until some absolute measure of visual posiEach poem in an anthology of related verse has its own systion in thne is devised.
tematic age, like each pot in a museum series, and each piece of
For the present, let us begin with the idea of simultaneity. The sculptured stone. Politics and civilizations probably can be ranked
simultaneous existence of old and new series occurs at every his. by their systematic ages. There is surely a systematic age indetori cal moment save the first. At that totally imaginary moment, I pendent of absolute age for the entire phenomenon of human
and only ilien, were all the efforts of men in the same systematic civilization, which has endured for far less time than humanity,
age. From the second instant of historical time, two kinds of be- in an open sequence that includes the entire history of things.
havior have been possible. Ever since then, most actions are
A Mexican paradigm. An instructive example of all these
ritual repetitions and very few are unprecedented.
matters comes from the Spanish colonization of the Indian
Things, which may be compared to fossil actions, display peoples of Mexico h1 the sixteenth century. The early churches
several kinds of simultaneity. A Visigothic relief carving from for two generations after the Conquest were built mostly by
Spain and a Maya stela were made during the same year in the 1 Indian workmen under the supervision of European Mendicant
sixth century A.D., and they are totally unrelated, but iliey are i friars." They can be classed readily in early and late groups
simultaneous. The Visigoiliic work belongs to a new series; the 1 within the era of Mendicant rule in Mexico until 1570. The early
Maya stela pertains to a very old one. At the other extreme of 1 churches have rib vaults, some of them resembling types disused
relatedness, Renoir and Picasso in Paris about 1908 knew each in Europe sh1ce the twelfth century, and they bear a decoration
other's work; Renoir's pictures then belonged to an old class and of late medieval antecedents. The late group of churches displays
Picasso's early Cubist pictures belonged to a new one. These I much technical refinement in the construction of domical vaults
pictures are simultaneous and related, but they have different like those then being built in Spain with a decoration of classisystematic ages, looking as different as the unrelated Maya and [ cizing origins. The beginning date and ilie initial conditions are
Visigothic sculptures. The similarities between Late Hellenistic unequivocal, like ilie terminal date, when secular clergy effecarchitecture at Baalbek and Roman Baroque churches long ago I tively displaced the friars in Indian towns.
captured the attention of historians, like iliose between different
19. Robert Ricard, La "Conquete spirituelle" du Mexique (Paris, 1933) and G.
kinds of archaic art and the different kinds of antique revival. I Kubler, Mexican Architecture of the Sixteenth Ce11tury (New Haven, 1948).
I
I
1'HE CLASSING OF
'I)
The need was clear to all the European colonists: the ,,,,.,,. ,,
conquest of the Mexican Indian required fme large ch1arches
convent-schools. The continuing problem presented by this
was to train and supervise Indian labor in European habits
work. The series embraced Mendicant churches with the
of subordioate classes therein involved. From the Indian
everythiog staJted as if from zero; the quarrymen had to
taught the use of metal tools; the masons had to be taught
principles and the technique of building arches and domes;
sculptors had to be taught Christian iconography aud the painteq
had to learn the principles of European one-point perspective
construction as well as the rendering of forms in graduated color
to simulate their appearances h1 light and shade. Any Indian
sense of need or problem surviving from pre-Conquest life
driven underground or out of existence. At the same time
evidence shows the Indian craftsmen eagerly turniog to

the superior techoiques and representational habits of their
pean teachers.
I'"
I
I
'
:I
Thus the Christian transformation of Mexican an;hitecllutt 1

contains at least three major patterns of change. Indian life
fests two of them: one is the abrupt abandonment of
habits and traditions, and the other is the gradual acquisition of
the new European modes of production. The discards and the
replacements happened at different speeds. The third pattern
governs the European colonists themselves. The conquistadon
were men of a generation already accustomed to eclectic variety
in Spaio: the architecture of their day was preponderantly late
medieval: Renaissance im10vations from Italy still were rare. The
leading fashion of the years 1500-1520 was the early mode
that harsh and energetic ornament later to be designated
Plateresque, which the men of the time io Spain thought of
a lo romano. By 1550 a new architecture based upon the art
Vignola began to displace Plateresque ornament. The Esco1:i~ I
is its principal expression, but contemporary with the Es,col:i~ j
and near it at Segovia, work continued on the late medieval
SBR[A
L POSITION~ AGE, AND CHANGE
59
't d cathedral begun only in 1525. Certain masters and workvatue

.
.
.
al
l . 1
articipated m both edtfices, workmg ternate y m t 1e
)!lOll P
,
,
othicizing a!ld m the Vtgnolan manners.
.
g By this account the situation of sixt.eenth-century Mextcan
hitecture seems very cotnphcated; m another perspective,
;,:wever, no historical crisis shared by s~ many milli~ns of people

h ever displayed its structure more stmply or platnly. A great
~tural distance separated native behavior and Spanish behavior
~: the I szo' s. This distance is like that which separates Old
J{ingdom Egypt or Sumerian Mesopotamia from the Europe of
charles the Fifth. Such dizzying confrontations of altogether
different stages of cultural development upon a contioental scale
characterize every colonial phase of modern European lustory
with many different results, but the same mechanism of change
is repeated in every instance.
.
.
In the grammar of historical change, the Mextcan conquest Js
like a paradigm, displaying with unusual clarity all the principal
roperties of this fundamental mechanism. The traditional be~avior of a person or of a group is challenged and defeated. New
behavior is learned from the victors, but during the learning
period, the new behavior is itself ~hanging. This pattern recurs
in every crisis of existence. The mnnmal case prob~bly concer~s
our use of words in daily speech, when we are mvaded, as 1t
were, by new behavior symbolized in a word of questionable
propriety or authority, and must decide to use or reject the expression and the behavior for which it is the symbot On anoth~r
scale of tnagnitude the crisis returns each year m women s
fashions, and it happens in every generation, whenever an avantgarde successfully advances the new or tmtried soluti~ns to cur-
rent problems, and in every trivial daily confrontatmn of retarde<i\.and progressive solutions to the same problem. Here the
terms ''retarded'' and ''progressive" are descriptive: no judgment
of quality is intended, for the terms only record the antithetic
phases of any moment of change, by describing their anchorage
in time as backward or forward lookiog.
6o
l'HB CLASSING' OF l'Hl!IJGs
sJlRI.AL POSITION, AGE, AND CHANGE
!ated or communication will fail. ''Noise, is irregular and unex-
Linguistic change. Recent discoveries in the stndy oflanguage>o

have an important bearing upon onr estimates of change in th
history of things. Cognate languages which are not in clos:
contact, hke Portuguese and French, or Huastec and Maya, have
changed gradually or "drifted" since the peoples were separated.
Word changes in lists of selected basic meanings can be measured
by simple statistical means. The surprising result is that linguistit
change occurs at a fixed rate, and that the amount of linguistic
change can be used as a direct measure of the antiquiry of the
separation of the peoples. Many htmdreds of textual and archaeological verifications support the thesis of the regularity of lin.
guistic change.
Historians are accustomed to think of cultural change as an
irregular and unpredictable set of occurrences. Because language
is an integral part of culture, it should share the irregularity and
unpredictability of history. How then can we account for the
unexpected regularity of linguistic change, and how does it
affect our conception of historical change?
In the first place, the historian's idea of change is related to
to the linguist's idea of "drift," exemplified by the progressive
separation that widens between cognate languages. This ''drift,"
produced by cumulative changes in the articulation of sounds,
can be related in tum to the interferences that distort any audible
ected change. The efficacy oflanguage requires that it be reduced

; a minimum. This is achieved by transforming irregularity
0
nto steady pulses that accompany the communication as a hum
~f fnred pitch and loudness. The rate of change in language is
regular because commtmication fails if the instrument itself
varies erratically. The "noise" of history is altered in language
to the unobtrusive hum of steady change.
These recent developments in the historical theory of language
require us to reconsider the position of works of art as historical
evidence. Most kinds of historical happening are subject to incalculable interferences which deprive history of tl1e scope of
predictive science. Linguistic structures, however, admit only
those interferences whose regularity will not interfere with comtnunication. The history of things, in turn, admits more inter-
ferences than language, but fewer than institutional history, because things which must serve functions and convey n1essages
cannot be diverted from these fmalities without loss of identity.

Within the history of things we fmd the history of art. More
than tools, works of art resemble a system of symbolic com1nunication which must be free from excessive "noise" in the
many copies upon which communication depends, in order to
ensure some fidelity. Because of its intermediate position be-
tween general history and linguistic science, the history of art
communication. The telephone engineer calls such interferences

"noise." "Drift," "noise," and change are related by the presence
of interferences preventing the complete repetition of an earlier
may eventually prove to contain unexpected potentialities as a

predictive science, less productive than linguistics, but more so
set of conditions.
than can ever be possible in general history.
Historical change occurs when the expected renewal of con

ditions and circumstances from one moment to the next is not
!i'
6r
completed but altered. The pattern of renewal is recognizable

but it is distorted; it is changed. In history the interferences preventing the faithful repetition of any pattern are mostly beyond
human control, but in language the interferences must be regu~
20. D. H. Hymes, "Lexicostatistics So Far," Current Anthropology,
3-44, and "More on Lexicostatistics," ibid., 338-45.
(Ig6o),
ENT!ON AND VARIATION
iN V
we welcome every mechanical production in an industrial
r: when the conception of planned waste has acquired positive

3 The Propagation of
agora! value instead of being reprehensible, as it was during

~llennia of agrarian civilization.
iNVENTION AND VARIATION
It is a truism that the objects around us correspond to nee~

old and new. But this truism like others describes only a fragment
of the situation. In addition to these correspondences between
needs and things, other correspondences exist between things
and things. It is as if things generated other things in their OWlJ
images by human intermediaries captivated with those possi.
bilities of sequence and progression we have just described. 111e
sense of Henri Focillon' s Vie des Formes captures the illusion
reproductive powers appearing to reside in things, and
Malraux later amplified the perception upon a much
canvas in Voices of Silence. In this view the propagation of things
may obey rules which we are now obliged to consider.
The occurrence of things is governed by our changing attitudes
towards the processes ofinvention, repetition, and discard. With.
out invention there would be only stale routine. Without copy
ing there would never be enough of any man-made thing, and
without waste or discard too many things would oudast their
usefulness. Our attitudes towards these processes are themselves
in constant change, so that we confront the double difficulty of
charting changes in things, together with tracing the change in
ideas about change.
A signal trait of our own time is an ambivalence in everything
touching upon change. Our whole cultural tradition favors the
values of permanence, yet the conditions of present existence require an acceptance of continual change. We cultivate avant
gardisme together with the conservative reactions that radicru
iuuovation generates. By the same token the idea of copying i1
in disfavor as an educational process and as an artistic practice,
The antipodes of the human experience of time are exact

repetition, which is onerous, and unfettered variation, whic~ is
haotic. If deprived of customary behavior, as after the bombmg
~f cities, people enter a state of shock, unable to cope with the in..
V'ention of a new environment. At the other extreme of behavior,
actuality becomes hateful within instants of our being deprived
of the faculty of departing from past custom. The puuishment of
the prisoner lies in precisely this ironbound restriction to routine,
which denies him the freedom to change external motions as
he pleases.
Iuventions, which are commonly thought to mark great leaps
in development and to be extremely rare occurrences, are actually
one with the humble substance of everyday behavior, whereby
we exercise the freedom to vary our actions a litde. We are only
slowly emerging from the great romantic deformation of experience, when all the different skills and callings were costumed
in fairy-tale parts, such as seven-league boots that allowed some
to leap faster and farther than their plainshod contemporaries.
The iuuovator in any class rejoices in his familiarity with certain kinds of difficulties, and when he invents something, he is
the beneficiary of what we have called a good entrance (pp. 6-7)
as the first to perceive a connection among elements to which
the key piece had only just come into view. Another could have
done it as well as he, and another very often does, as in the celebrated coincidence of Charles Darwin and Alfred Wallace in respect to the theory of the origin of the species, because of similar
training, equal sense of problem, and parallel powers of perseverance.
THE PROPAGATION OF
In our terms each invention is a new serial position. The ac..

ceptance of an invention by many people blocks their con,
tinuing acceptance of the preceding position. These blocks pr~
vail only among closely related solutions in sequence, and they
do not arise outside their own field of relevance, as we kuo~
from the coexistence at any instant of great numbers of active
and simultaneous series (pp. II2 ff.). The products of prior posj,
tions become obsolete or m1fashionable. Yet prior positions are
part of the invention, because to attain the new position the
inventor must reassemble its cmnponents by an intuitive in.
sight transcending the preceding positions in the sequence.
its users or beneficiaries the new position also demands son1e
familiarity with prior positions in order that they may discover
the working range of the invention. The tcclmique of invention
thus has two distinct phases: the discovery of new positions fol.
lowed by their amalgamation with the existing body of know\.
edge.
The familiar delay between discovery and application reappean
in every field of kr10wledge: the invention of oil painting oc.
curred several times and at several places before fifteenth-century
painters were ready to use it in panel painting. The discovery
metal tools can be ascribed in ancient America to several independent centers: the central Andes, c. rooo B.c.,
Mexico, after A. D. IOoo, and the Great Lakes country of
America before A.D. 1000. Possibly southern Mexico learned
from the Andean peoples: certainly the North American Indians
discovered metals independently.'
When we imagine the transposition of the men of one age into
the material setting of another, we betray the nature of our ideas
about historical change. lu the nineteenth century Mark Twain's
Connecticut Yankee was imagined as a superior person successfully enlightening the Middle Ages. Today we would view him
only as a stray spark swiftly extinguished without further notice.
jti'/ENTION AND VAlUATION
6s
p,n iustructive fantasy is to imagine the exploration of a historical

,nauifold of dimensions in which all times could coexist, If we
,night exchange information with men four hundred centuries

before us, it is likely that the venture would result only in the
cnriclm1ent of our lmowledge of paleolithic life. Our skill at
bunting large auimals would appear worthless, and paleolithic
nsen would have no use for anything of ours unrelated to his
usual needs. If, on the other hand, we should ever have the misfortune really to encounter the future, as tl1e Indians of sixteenthcentury America encountered it, we would have to abandon all
our own positions to accept all those of the conqueror.
b1 other words, our ability at any moment to accept new
knowledge is narrowly delimited by the existing state of knowledge. A fixed ratio may mark the two kinds of knowledge. 111e
!Uore we know, the more new lmowledge we can accept. Inventions lie in this penumbra between actuality and the future, where
the dim shapes of possible events are perceived. These narrow
limits confine originality at any moment so that no invention
overreaches the potential of its epoch. An invention may appear
to meet the edge of possibility, but if it exceeds the penumbra, it
remains a curious toy or it disappears into fantasy.
Artistic invention. The discoveries and inventions of the past
three centuries outnumber those of the entire previous history
of mankind. Their pace and nun1ber continue to increase as if
asymptotically towards a limit which may be inherent to human
perception of the cosmos. How does artistic invention differ
from useful inventio111 It differs as human sensibility differs from
the rest of the tmiverse. Artistic inventions alter the sensibility of
mankind. They all emerge from and return to human perception,
unlike useful inventions, which are keyed to the physical and
biological environment. Useful inventions alter mankind only
indirectly by altering his environment; aesthetic inventions enlarge human awareness directly with new ways of experiencing
r. P. Rivet and H. Arsandaux, La mitallurgie en Amirique pdco!ombienne (Paris, the universe, rather than with new objective interpretations.
1946).
Psychological science is more concerned with human faculties
66
THE PROPAGATION OF TH!Nct . tNV~N'f!ON AND VARIATION
as separate objects of study than ~ ~ single historically changing

instrmnent of awareness. Aesthetic mventtons are focused up011
.i:.
'
'I'
rtists learn a theory of limits from aesthetics, but aestheticians

;,,rn little from art, being concerned more with philosophical
individual awareness: they have no therapeutic or explanatory questions than with artistic ones.
Indeed the work of many artists often comes closer to philopurpose; they only expand the range of human perceptions by
enlarging the channels of emotional discourse.
sophical speculation than most aesthetic writings, which retrace
The late medieval invention of a pictorial language capable of rbe same ground over and over, sometimes systematically and
representi?g all space o~ two dimensions~ in northern Europe sometimes historically, but rarely with originality. It is as if
(Pol de Lrmbourg) and m Italy, was an aesthetic invention dif, aesthetics had ceased since Croce and Bergson to be an active
ferent from the technical invention of oil painting. A related branch of philosophical speculation. The major artistic inventions,
effort in the twentieth century is still in progress to invent a on the other hand, resemble modern mathematical systems in the
language of expression that will be independent of images (ab. freedom with which their creators discarded certain conventional
assumptions, at1d replaced them with others. There has not been
stract expressionism) and of fixed intervals (atonal1nusic).
It does not follow that aesthetic invention is less important and there could not be any separation between theory and apthan useful invention because it concerns only. an infinitesim~ plications, or between inventor and user: both had to be one and
portion of the universe altered by useful invention. Human sen. clte same person, so that at the beginning of these achievements
sibility is our only channel to the universe. If the capacity of that cl1e painter or the musician performed single-handedly all the
channel can be increased, knowledge of the universe will expand functions elsewhere distributed among several persons in the
accordingly. The charmel may of course be augmented by many search for new knowledge.
useful inventions; ultilnately, however, every rational structure
Returning now to the place of invention in the history of
can be razed by adverse feeling. Emotions ftmction like a main things, we confront once again the paradox which arose earlier
valve in the circuit between us and the tmiverse. They may be in this discussion. It is the paradox of generalization concerning
artificially regulated by chemistry and psychiatry, but neither unique events. Since no two things or events can occupy the
of these sciences can enlarge the instruments of aesthetic ex~ same coordinates of space and time, every act differs from its
perience. Such enlargements always remain the prerogative of predecessors and its successors. No two things or acts can be
artistic invention.
accepted as identical. Every act is an invention. Yet the entire
In rough terms, artistic invention is one among many ways of organization of thought and language denies this simple affirmaaltering the set of the mind, while useful invention marks out tion of non-identity. We can grasp the universe only by simplifythe scope of the knowledge the instrument was previously de- ing it with ideas of identity by classes, types, and categories and
signed to encompass.
by rearranging the infmite continuation of non-identical events
We have been using the terms of art and aesthetics almost as into a finite system of similitudes. It is in the nature of being that
synonyxns, for the reason that a clear separation such as the one no event ever repeats, but it is in the nature of thought that we
between pure and applied science fails to appear in aesthetics. understand events only by the identities we imagine among them .
Convention and invention. In the daily behavior of Everyman,
The significant gains are registered as art rather than as aesthetics.
The separation between contemplative and practical labors that this paradox fmds a constant expression. Since each act partakes
science requires has little meaning where art is concerned. Some of invention by departing from the preceding action in its class,
68
THE PROPAGATION OF TllllllGs
invention is open to everyone all the thne. One result is that in..
vcntion is misunderstood in two ways: both as a dangerous de..

parture from routine, or as an unconsidered lapse into the
unknown. For most persons inventive behavior is a lapse of
propriety surrounded by the frightening aura of a violation of
the sanctity of routine. They are so carefully schooled to con.
vention that it is nearly impossible for them to fall even by
accident into the unlmown.
Many societies have accordingly proscribed all recognition of
inventive behavior, preferring to reward ritual repetition, rather
than to permit inventive variations. On the other hand no form
of society ever can be devised to allow each person the liberty to
vary his actions indefinitely. Every society functions like a gyro.
scope to hold the course despite the random private forces of
deflection. In the absence of society and instinct, existence would
float as if unbound by gravitation in a world without friction
inventive behavior. The more common category of inventions

embraces all discoveries arising from the intersection or con-
frontation of previously unrelated bodies of knowledge. The

intersection may bring a principle together with traditional
practices. Or the confrontation of several unrelated positions may
evoke a new interpretation clarifying them singly and together.
Whether extensions of a principle, or new explanatory principles
be the results of such intersections, they all arise from confrontations among elements already given to tl1e observer.
A much rarer category of "radical" inventions discards these
ready-made positions. The investigator constructs his own sys-
tem of postulates and sets forth to discover the tmiverse they

alone can disclose. The confrontation is between the new, un-
tried coordinates and the whole of experience: between untested

guide-lines and tl1e evidence of the senses: between the unknown
and tl1e familiar, the assumed and tl1e given. It is the method of
radical invention, which differs from the otl1er method of discovery by confrontation, much as a new mathematical topic
differs from its applications.
If tl1e difference between useful inventions and artistic inventions corresponds to the difference between changing the
or from commercial advertisements.
Both in science and in art the inventive behavior rejected by

the mass of people has become more and more the prerogative
of a handful who live at the crumbling edge of convention. Only
exceptionally can any of these have an entrance allowing him
to wander very far. A few in each generation arrive at new posiI
rearranging the furniture, frmn new confrontations rather than
from fresh questions aimed at the center of being.

Indeed these alternatives mark the outlines of a typology of
the channeled pathways of tradition. Every act would be a free

. Thus the human situation admits invention only as a very
d1fficult tour de force. Even in industrial societies which depend
upon constant renewal by novelty, the very act of invention is
distasteful to the majority. The rarity of invention in modern
life corresponds to fear of change. The spread of literacy today
is manifested not by a happy attention to new actions and new
thoughts, but by stereotypes drawn from political propaganda
in much shorter lengths. Most of their inventions arise, like
from precedent, without the attraction of example, and without

invention.
JNVBNTION AND VARIATION
tions requiring the gradual revision of older opinions. The great

mathematicians and artists, who stray farthest from usual notions,
lead the procession. The chains of other inventors are measured
environment and changing our perceptions of the environment,

then we must account for artistic inventions in the terms of perception. A special character of major artistic inventions resides
in their apparent remoteness from what has gone before them.

Useful inventions, when seen in historical sequence, show no
such great leaps or discontinuities. Every stage follows its predecessors in a close-meshed order. Artistic inventions, however,
seem to cohere by distinct levels between which the transitions
are so difficult to identify that their existence may be questioned.
An important component in historical sequences of artistic
70
i
':1!::
: ',
'i
;,,;
'ji i
'>i
THE PROPAGATION OF Tiil:N"Gg
events is an abrupt change of content and expression at intervals

when an entire language of form suddenly falls into disuse, beillg
replaced by a new language of different components and an
unfamiliar grammar. An example is the sudden transformatio11
of occidental art and architecture about 19!0. The fabric of
society manifested no rupture, and the texture of useful inventions
continned step by step in closely linked order, bnt the system of
artistic invention was abruptly transformed, as if large numbers
of men had suddenly become aware that the inherited repertory
of forms no longer corresponded to the actual meaning of existence. The new outer expression wit!1 which we are universally
familiar today in all the figurative and structural arts is an expression corresponding to new interpretations of the psyche, to
a new attitude of society, and to new conceptions of nature.
All these separate renovations of thought came slowly, but in

art the transformation was as if lnstantancous, with the total con~
figuration of what we now recognize as modern art coming all
at once into being without many firm links to the preceding

system of expression. The cumulative transformation of all being
by useful inventions was a gradual process, but its recognition
in perception by a corresponding mode of expression in the arts
was discontinuous, abrupt, and shocking.
The nature of artistic invention therefore relates more closely
'i:'
to invention by new postulates than to that invention by simple

confrontation which characterizes the useful sciences. Here we
touch again upon the fundamental differences that separate prime

objects from replicas. A prime object concerns radical invention,
while the replicas vary from their archetypes by small discoveries
based upon simple confrontations of what has already been done.
Hence radical invention is most likely to occur at the beginning
of a series; it is marked by many prime objects; and it resembles
artistic creation more than it resembles scientific proo As a
series "ages," its primes are less numerous than at the beginning.
We may imagine the present instant as a smooth gradation
between before and after, excepting when radical inventions and
~npUCAT!ON
71
rime objects make their presence known, like the infmitesimally

pmall amonnts of new matter created from time W time, with the
s emendons energies of physical science. Their appearance in the
::xture of actuality forces the revision of all human decisions,
not instantaneously but gradually until the new particle of knowledge has been woven into every individual existence.
REPLICATION
This age dedicated to change for its own sake has also discovered the simple hierarchy of the replicas that f!II the world.
We shall merely mention the staggering replication present in
energy with only thirty some particles, or in matter which consists of about a hundred atomic weights, or in the genetic transmission of life, which since the beginning now comprises only
about two million described species of animals.
The replication that fills history actually prolongs the stability
of many past moments, allowing sense and pattern to emerge
for us wherever we look. This stability, however, is imperfect.
Evety man-made replica varies from its model by minute, unplat1lled divergences, of which the accumulated effects are like
a slow drift away from the archetype.
The term "replication" is a respectable old-fashioned word
long in disuse, and we revive it here not only to avoid the
negative judgment that adheres to the idea of "copying" but
also to include by definition that essential trait of repeating
events which is trivial variation. Since sustained repetition of any
sort is impossible without the drift occasioned by tiny unwanted

variations, this slow historical motion engages our 1nain interest
here.
Permanence and change. Let us imagine a duration without any

regt!lar pattern. Nothing in it would ever be recognizable, for
nothing would ever recur. It would be a duration without
measures of any sort, without entities, without properties, without events-a void duration, a timeless chaos.
Our aetna! perception of time depends upon regularly recur-
72
THE PROPAGATION OF THl.NGs
rent events, unlike our awareness of history, which depend,

upon unforeseeable change and variety. Without change there
is no history; without regularity there is no time. Time and hi,..
tory are related as rnle and variation: time is the regular settu1g
for the vagaries of history. The replica and the invention are
related in the same way: a series of true inventions excluding all
intervening replicas would approach chaos, and an all-embracing
infinity of replicas without variation wonld approach formle.,.
ness. The replica relates to regularity and to time; the invention
relates to va~iation and to history.
Human desires in every present instant are torn between the
replica and the invention, between the desire to return to t11e
known pattern, and the desire to escape it by a new variation,
Generally the wish to repeat the past has prevailed over the ;,11
pulses to depart from it. No act ever is completely novel, and no
act can ever be quite accontplished without variation. In every
act, fidelity to the model and departure from it are inextricably
mingled, in proportions that ensure recognizable repetition, together with such minor variations as the moment and the circum~
stances allow. Indeed, when variation from the model exceeds
the amount of faithful copying, then we have an invention.
Probably the absolute ammmt of replication in the universe exceeds the variations, for if it were otherwise, the universe would
appear more changeable than it does.
The anatomy of routine. Replication is similar to cohesion.
Every copy has adhesive properties, in holding together the
present and the past. The universe keeps its form by being perpetuated in self-resembling shapes. Unlimited variation is a synonym for cl1aos. The number of ritual acts in Everyman's life
greatly exceeds the few variant or divergent actions permitted
in his daily round. Indeed the cage of routine binds him so
closely that it is almost impossible for him to stumble into an
inventive act: he is like a tight-rope walker whom vast forces
so bind to the cable that he cannot fall, even if he wishes, into
the unknown.
]l.IiPLICATION
73
Every society binds and shelters the individual within an invisible many-layered structure of routine. As a single entity he is
urrounded by tl1e ceremonial duties of physical existence.
~nother less dense shell of routine binds him and protects him,
as a participant in the life of a family. The group of families
n1akes a district; the districts make a city; cities form a county;
counties nuke a state; states compose a civilization. Each suc-
cessively more open layer of routine further shelters the individual

from disruptive originality. The total structure, while remaining
many-centered, protects each person with many layers. Some
persons have fewer shells of.routine, ~ut non~ is entirely exet~pt.
The entire system of these mterlockmg, reciprocally supportmg

routines of course drifts and sways, and swells and shrinks, in
response to many conditions, not the least of which is inherent
in duration itself, as we observe whenever a variant introduced
for variety's sake and for no other reason appears in the texture
of a repetitive act: the bored scribe introduces secret variations
into the tnany copies of the invitation he must pen.
The foregoing remarks describe only a cross-section of the

habitual actions holding societies together. There is also another
dimension, composed of successive repetitions of the same acts.
Each act varies slightly from the preceding. The successive versions-with gradual changes of which some are independent of
external causes, occasioned only by the makers' craving for
change, drroughout long periods of repetition-describe distinct

patterns in time of which we shall here attempt a rough description and a classing (p. 96). The identification of these
fibrous durations of behavior is not easy: beginnings, endings,
and botmdaries are most elusive, and their dctennination perhaps
requires a special geometry of which the rules still arc not available to historians.
Recognizable behavior is recurrent behavior, but any study
of behavior immediately brings us to the unresolved question:
what are the fundamental units of behavior, and how numerous
are they1 Because of the restriction to things, our f1eld is greatly
74
THE PROPAGATION OF 'rl-IIJ:iG$
simplified in being reduced to the physical products of behavio

which allow us to substitute things for actions. The field rf
study still remains unmanageably complex, yet we are noo
enabled to view fashion, eclectic revival, and Renaissance as r~
lated phenomena of duration. Each of them requires large nu ,
111
be."' of ri~al gestures, which ensure participation with ;u1 ad.
tmred soctety, but each occupies a different typical duration
(p.
99 ).
Our concepti~n of the copy now includes both acts and things.
Under the rubnc of actions we examined repetition in general
including habits, routines, and rituals. Among things (whic!'
differ as approximate and exact duplicates) our attention tur~
to copies and replicas. Symbolic associations, however, attach
both to things and to actions. A symbol exists by virtue of

repetitions. Its identity among its users depends upon their
shared ability to attach the same meaning to a given form. The
person using a symbol does so in the expectation that others will
enlarge the association as he does, and that the resemblances bctweett people's interpretations of symbols will ontweigh clte
differences. Indeed it is unlikely that any copy ever passes as such
without a great deal of assistance from symbolic associations.
For example, in 1949 when I produced a photograph of him for
a Peruvian tnountain shepherd at Vicos, who had never seen a
photograph, he was unable to orient the flat, spotty paper or to

read it as a self-portrait, for want of those complicated habits of
translation which the rest of ns perform from two to three ditnensions without effort.
.!,i'
,!
In this perspective all things and acts and symbols-or cite

whole ofhnman experience-consist of replicas, gradually chmging by minute alterations more than by abrupt leaps of invention.
People have long assumed that only the large changes were
significant, like those represented by great discoveries, as of gravitation or the circulation of the blood. The little changes separated
by infinitesimal alterations like the chmges appeating in copies
of the same document prepared by different scribes, are dismissed
jBPL!CAT!ON
5
75
trivial. By the interpretation offered here, large-interval
~banges are similar to small-interval changes. Furthermore, many

~nges assumed to be large are really small when seen in fnll
context. Thus a historian collecting the information gathered by

others, is enabled to reinterpret the entire body with a more
satisfying explanation, for which the entire credit attaches to his
narne, although his personal contribution is only of tlte same
order of magnitude as that of the single units of information on
which the theory itself rests. Hence a difference traditionally assurned to be one of a kind may be a difference only of degree.
The limits to our organic situation are very narrow. Too cold,
or too hot, too little air, or too much bad air, and we die. Our
tolermces for many other modes of variation are probably also
small, with the present tnoment serving as an extremely minute
valve regulating the amount of change adtnitted into reality.
Since the universe remains recognizable from one moment to
the next, each instant is nearly an exact copy of the one immediately preceding it. The changes which occur are small
relative to the whole, and they are proportional to the magnitnde
we assume for momentary duration. The idea corresponds to
our direct experience: from this instant to the next the motions
of the tmiverse will not change greatly, bnt during the next year
the course of happening will have altered direction many times.
Large historical changes occupy large durations. If the accountancy is properly conducted, no great event can be assigned to
brief times, althongh the traditional terms of speaking about the
past often require ns to think and behave as if history consisted
only of great, brief instants separated by wastes of empty duration.
Historical drift. Certain types of motion appear when we look
at time as an accumulation ofnearly identical successive moments,
drifting by minute chmges towards large differences amassed
over long periods. Motion is perhaps a misnomer for the changes
occurring between early and later members of a series of replicas.
Yet a series of objects, each made at a different time, and all
THE PROPAGATION OF THINGS
''i
'!.
related as replicas based upon the same original form, describe

through time an appearance of motion like that of the frames
of a film, recording the successive instants of an action, which
produce the illusion of movement as they flicker past the beam of
light.
Replication obeys two contrasting kinds of motion. They may
be described as motions towards and away from quality. Aug.
mented quality is likely when the maker of a replica enriches
upon hls model by adding to its excellences, as when a talented
pupil in1proves upon his 1nediocre teacher's exercises, or when
Beethoven enriches Scotch songs. Diminished quality becomes
apparent when the maker reduces the excellence of the replica,
either because of economic pressures or because of his inability
to comprehend the full scope or import of the model. Peasant
replicas of court furniture and dress; paintings which are copies
by untalented pupils; provincial series of constantly coarsened
replicas of replicas are all examples of diminishing quality. Jn.
dustrial examples are n1orc cotnmon. When a mass-produced
article of good design begins to have a wider market and more
intense competition, the manufacturers simplify its design to get
the price down until the product is reduced to the fewest possible
parts in a construction no more durable than necessary. Hence
loss of quality has at least two different velocities: one provincial,
leading to coarseness, and the other commercial, leading to
tawdriness. Provincialism and commercialism thus are related
as kinds of qualitative degradation.
Some five thousand years ago there were no great cities and
no far-flung trade, but only farming or fishing villages. The
human community achieved both provincial and commercial
distinctions only much later. The record of these earlier periods
shows much less gradation in quality than the works of today.
The pottery objects of one village differed little from one
another within the span of a generation, and it is only across
longer periods that we begin to see the interplay of regional
traditions with positions upon the scale between early and late
p!SCARD AND RETENTION
77
JUO!lUfactures in the same series. The idea of metropolitan or

centrally influential events, as well as the idea of an excellence
depending upon old and stable craft traditions, prob~bly entered
human consciousness only with the appearance of nttes and wtth
economic differences allowing the creation ofluxuries by special
artisans. Provincial degradation is surely older than commercial
vulgarization. But village monotony remains the oldest of all
the qualitative grades of civilized life.
Drs CARD
AND RETENTION
Change is occasioned by the many ways in which entities

join and separate. Actuality concerns their instantaneous arrangement, and history treats of their successjve positions and rela-
tionships. The different positions of human entities suggest a

category of forces for whlch we have no other name than needs.
Discard and retention belong with other processes to this field
of forces-a field where appetite and repletion, pleasure and revttlsion, are the poles.
The decision to discard somethlng is far from being a simple
decision. Like each fundamental type of action, it appears in the
experience of every day. It is a reversal of values. Though the
thlng once was necessary, discarded it becomes litter or scrap.
What once was valuable now is worthless; the desirable now
offends; the beautiful now is seen as ugly. When to discard and
what to discard are questions to which the answers are governed
by many considerations.
Obsolescence and ritual. The usual view in our age is that obsolescence is merely an economic phenomenon occasioned by
technical advances and by pricing. The cost of the maintenance
of old equipment outruns the cost of its replacement with new
and more efficient items. The incompleteness of this view is
apparent when we consider the decision not to discard.
For example, one important way of retaining the products of
the past occurs in the elaborate tomb furniture of some peoples,
THE PROPAGATION OF THI:tiGS
whose ftrm belief in an after-life was of course a main motive

for the immense funeral deposits in Egyptian, Etruscan, Chinese
or Peruvian burial grotmds. We are at liberty to suppose tha;
the craftsmen in these societies were excessively numerous in.
relation to princely demands upon their skill. The corollary is
that the production of precious things far exceeded the rate of
their disappearance by normal wear and breakage.
The idea of elaborate tomb furnishings then would have
achieved several objectives: to preserve past remains with due
piery while avoiding the contagion of handling the things of
dead people, as well as to withdraw old-fashioned objects from
daily use and thereby to stimulate new manufactures for livhtg
use. Tomb furniture achieved apparently contradictory ends, in
discardh1g old things all while retaining them, mnch as in out
storage warehouses, and museum deposits, and antiquarians'
storerooms. There, huge acCumulations await a return to use
,,,
I
which depends both upon increasing rarity and upon future

changes of taste.
Here is a possible reason why the rate of technological change
was so slow until a century or so ago. When the making of things
required a very large effort, as in pre-industrial life, it was easier
to repair them than to discard them. The opportunities for change
were correspondingly fewer than in au industrial economy,
where mass production permits the consumer to discard equipment regularly. Because each change suggested by invention required too great a sacrifice of the existing equipment, it could
enter realiry ouly when the offending conventions, habits, aud
tools dropped Uffiloticed out of regular use. In other words, the
need to retain things may have governed the rate of change in
human manufactures more than the desire for novelty.
In the conception of systematic change, every action belongs
in a series of similar actions. By hypothesis, all such similar actions
are linked as model and copy. The copies vary by minute differences. In every series early actions differ from late ones. The
recognition of these series often is hindered by many kinds of
plSCARD AND RETENTION
79
outside interference, as well as by uncertainties about the morphology of early and late actions. Further uncertainty surrounds
the analysis of any action h1to serial components of which some
are old behavior and others are new. But if our hypothesis holds,
then we must add to any external explanations of any part of
behavior, some account of the behavior in respect to its seriation.
Establishing chronological order is not enough, for absolute
chronology merely arranges the moments of time in their own
sidereal succession. The perpetual pro blcm confronting the historian has always been to find the beghrning and the end of the
threads of happening. Traditionally he has cut the thread wherever the measures of narrative history indicated, but those cuts
have never taken as a possible measure the differences between

different lengths in specific duration. The discovery of these
dumtions is difficult, because only absolute chronology can now
be 1neasured:
all
past events are more remote from our senses
than the stars of the remotest galaxies, whose own light at least
still reaches the telescopes. But the moment just past is extinguished forever, save for the things made during it.
In the subjective order an act of discard relates to the ends of
durations, just as an act of invention initiates them. It differs
from other kinds of rupture (p. 109) as a free decision differs from
an imposed one, or as a slowly accumulated resolve differs from
an 1lllprepared action in au emergency. The act of discard corresponds to a terminal moment in the gradual formation of a
state of mind. This attitude is compoll1lded of fauilliarity and
discontent: the user of au object knows its limitations and its
incompleteness. The thh1g meets ouly a past need without corresponding to new needs. The user becomes aware of possible
improvements in merely noting the lack of correspondence between things and needs.
Discarding useful thmgs differs from the discard of pleasurable
things in that the first operation is more final. The breakup of
old tools has often been so complete that practically nothing
survives of the equipment of many epochs, save a few images
So
THE PROPAGATION OF Tlli!:H~S
l)JSCARD AND RETENTION
8r
striking when eclectic taste was domiuant. 2 They were striking,

pot because architectural change was then more rapid than it is
roday, but because there were only a few historic styles to choose
from, and it seemecl to the men of the period that they were
rapidly exhausting these reserves of the past.
Goller, an architect and professor at the Polytechnikum in
Stuttgart, was an early psychologist of artistic form who belonged
to the tradition of abstract formalism in aesthetics initiated by
flerbart. Since about 1930 Goller's expression, Formermiidung,
has come into wider use, although no one has pursued his ideas
allY farther. To Goller architecture was an art of pure visible
form. For him its symbolic content was negligible, and its beauty
arose from a pleasurable but meaningless play of line, light, and
shade. Goller's principal hope was to explain why the sensations
of optical pleasure are in constant change as manifested in the
sequence of styles. The nucleus of Goller's thought is that our
delight in pleasurable forms arises from the mental effort of
dmping their recollection. Sinoe all mental knowledge consists
of such memory residues, the enjoyment of beautiful forms requires a well-stocked memory. Taste is a function of familiarity.
Our pleasure in pure form din1inishes as we succeed in reconstituting its complete and distinct memory. A total memory prehave a similar purpose, to keep present some record of the power sumably includes the frustrations and dissatisfactions arising with
any recurring unit of experience: ''familiarity breeds contempt.''
and knowledge of vanished peoples.
These
total recollections occasion fatigue, and they lead to the
Aesthetic fatigue. The point of these distinctions is that merely
search
for
new forms. As the mind fastens upon content, however,
useful things disappear more completely than meaningful and
pleasurable things. The latter seem to obey a more lenient rule the rule of Formermudung weakens, and the object holds our atof discard. This rule corresponds to a state of mental fatigue, tention in proportion to the complexity of its meaning.
signifying a jaded familiarity more than muscular or nervous
2. I first became aware of A. GOller in tracing the origin of Formermiidung. His
exhaustion. DijCt vu and trop vu are equivalent expressions in short essay, "Was ist die Ursache der immerw1ihrenden Stilveranderung in der
French criticism for which no current English phrase exists, Architektur?" in Zur Aesthetik der Architektur was amplified one year later as
Entstehung der architektonischen Stilformen (Stuttgart, 1888). Benedetto Croce menunless it be tedium. Its nature and conditions first attracted at- tioned GOller only to reject his empirical and formalistic tendencies (Etil barroca
tention in r887, when Adolf Goller wrote on fatigue (Ermiidsmg) [Bari, 1929], p. 35), hut Wasmuth's Lexikon der B11ukunst, 2 (1930), 653, noted
as it affected the many changes of architectural style that were so GOller's strong influence upon the architectural thought of his own day.
and ceremonial versions of the metal instruments which were

melted down when new forms were needed. A main reason for
this pitiless scrapping of the tools of the past is that a tool usual!y
has but one single functional value.
An obje~t made for emotio~al experience-which is one way
of 1dent1fymg a work of art-dtffers from a tool by this meaning.
ful extension beyond use. Because the symbolic frame of existence
changes much more slowly than its utilitarian require1nents, so
the tools of an era are less durable than its artistic productions. It is
much easier to reconstitute a symbolic facsimile of medieval life
with a small museum of manuscripts, ivories, textiles, and jewel..
ry, than to attempt to describe feudal teclmology. For the tech.
nology, we have only suppositions and reconstructions. But for
the art we have the objects themselves, preserved as symbok
which still are valid in actual experience, unlike the superan.
nuated discards of medieval labor.
The retention of old things has always been a central ritual u1
hmnan societies. Its contemporary expression in the public mu..
seums of the world rises from extremely deep roots, although
the museums themselves are only young institutions going back
to the royal collections and the cathedral treasuries of earlier
ages. In a wider perspective the ancestor cults of primitive tribes
THE PROPAGATION OF Tllil!Qs
The artist himself is most exposed to tedium, overcoming it

by the invention of new formal combinations and by mote
daring advances in previously established directions. These ad.
vances obey a rule of gradual differentiation because they must
;emain as recogniza~le. variations upon the dominant men101~
Image. The differentlatwns are bolder among yotmg designers
and their tempo becomes more rapid as a style approaches it;
end. If a style is interrupted early for any reason, its unused re.
sources become available for adaptation by participants in other
styles.
The work of the htrman mind cannot be accounted for by any
isolated process. Goller underestimated other forces in the com.
plex interplay of actuality and the past, and he lacked an explicit
conception of the tradition-forming power of replication, or of
the human requirement for variety, which is satisfied by inventive behavior. Yet despite his exaggerated insistence upott
tedium or fatigue as the sole motive force driving designers from
fully mapped positions to less well charted ones, Goller clearly
thought of art as consisting of linked series of forms, which
differ gradually from one another until the potentialities of the
class have been used.
4 Some Kinds of Duration
The modern professional humanist is au academic person

who pretends to despise measurement because of its ''scientific''
nature. He regards his mandate as the explanation of human expressions in the language of normal discourse. Yet to explain
something and to measure it are similar operations. Both arc
~anslations. The item being explained is turned into words, and
when it is measured it is turned into numbers. Unfortunately
d1e tissues of history today have only one dimension that is
readily measured: it is calendrical time, which permits us to arrange events one after another. But that is all. The domain of
ilie historical sciences remains impervious to other numbers.
We can nevertheless use the language of measurement without
numbers, as in topology, where relationships rather than magnitudes are the subject of study. Calendrical time indicates nothing
about the changing pace of events. The rate of change in history
is not yet a matter for precise determinations: we will have advanced if only we arrive at a few ideas about the different kinds
of duration.
The history of things is about material presences which are
far more tangible thau the ghostly evocations of civil history.
The figures and shapes described by the history of things arc
moreover so distinctive that one asks whether artifacts do not
possess a specific sort of duration, occupying time differently
from the animal beings of biology aud the natural materials of
physics. Durations, like appearances, vary according to kind:
they consist of characteristic spans and periods, which our
generalizing habit oflanguage makes us overlook, since we can
AND SLOW HAPPENING
SOME KINDS OF DURAl'lON
85
transform them so easily into the common currency o f so1at l..seSSJIUents of cultural effort permeate sociology, anthropology,
economics. One exa1nple is the economic explanation of the
time.
FAST AND SLOW HAPPENING
,;
I
i''
,.!
I
I
'
'
position of the artist and designer in twentieth-century industrial

life. To the economist artifacts change according to the market:
falling or rising demmd for an article affects the volume of
1
Time has categorical varieties: each gravitational field in

the cosmos has a different time varying according to mass. On
earth at the same instant of celestial time, no two spots really
production; opporttmities to change the product vary with

production.
But our earlier distinction between prime objects and replica-
have the sa1ne relation to the sun despite our useful convention
of time-zones regulating the regional concordance of clocks.

When we define duration by span, the lives of men and the lives
of other creatures obey different durations, and the durations of
artifa:ts differ from those of coral reefs or cbalk cliffs, by occupymg different systems of intervals and periods. The conventions
oflanguagc nevertheless give us ouly the solar year and its multiples or divisions to describe all these kinds of duration.
St. Thomas Aquinas speculated in the thirteenth century upon
the nature of the time of angels, and, following a nco-Platonic
tradition, 1 he revived the old notion of the aevum as the duration
of human souls and other divine beings. This duration was intermediate between time and eternity, having a beginning but no
end. The conception is not inappropriate for the duration of
many kinds of artifacts-so durable that they antedate every
hvmg creature on earth, so indestructible that their survival may,
for all we know, ultimately approach infinity.
nasses suggests another line of argument. The existence of masses
of copies testifies to a large public, which may desire or condemn

1chlmg;e. When change is wanted, the public itself requires only
I :~ncwements or extensions npon the actual product. Public derecognizes only what exists, uulikc the inventors and artists
minds turn more upon future possibilities, whose speculaand combinations obey an altogether different rule of order,
IJcscribc;d here as a linked progression of experiments composing
formal sequence.
This separation between designer and consumer of course does
110t hold in many kinds of society, where most objects are made
ill the family circle. But the division by prime objects and replicamasses holds for peasant society as well as for the tribal compound ancl the courts of eighteenth-century Europe. There can
be no replica without an original; and the family of originals
takes us back directly to the genesis of humau society. If we wish
to explore the nature of change, we must examine the sequence
Confining our attention to the history more than the future of of forms.
man-made things, what conditions must we take into account
The replicas may directly reflect such magnitudes as wealth,
to explain the variable rate of change 1 The social scientists de- population, and energy, but those magnitudes do not alone
scribe material culture as an epiphenomenon, that is, as the account for the incidence of the original or prin1e expressions
necessary result of the operation of forces which the social sci- from which the replicas derive. Prime expressions in turl;l occur
entists have already formulated and charted. For instance, small in formal sequences. This conception supposes that inventions
societies dispose ofless energy than large ones, and they arc con- are not isolated events, but linked positions of which we can
sequently less able to initiate costly enterprises. Such quantitative trace out the connections. The idea of seriation also presupposes
I. Duhcm, "Le temps selon les philosophes hellenes," Revue de philosophic
(,91I).
a structural order in the sequence of inventions which exists in-
dependently of other conditions. The order of realization may
86
SOME KINDS OF DURATIQ~
be deformed, and the sequence may be stunted by outside con.

ditions, but the order itself is not the consequence of such con,
~itions. Under certain circumstances an inherent structural ordet
m the sequence of new forms is easily apparent to many ob.

servers: we perceive it most clearly in the early stages of Greek
figural sculpture, Gothic architecture, and Renaissance painting,
where linked runs of related solutions follow one another in
recognizable order as if following out the conditions of a prio:
program common to these various evolutions.
Runs of this type are nevertheless far from common. Because

of the. completeness of the inventory of the European past they
come Into v1ew more frequently there than on other continents.
But their appearances are widely separated, as if events of this
magnitude could not recur frequently. Perhaps they are like the
aurora borealis, best noted at some latitudes rather than others
as field manifestations visible only under special conditions.
'
The typology of artists' lives. More readily available for observation are the lives of famous artists. The pace and tone of an
artist's life can tell us much about his historical situation, although
most artists' lives are uninteresting. They fall usually into routine
divisions: apprenticeship, early commissions, marriage, family,
mature works, pupils, and followers. Smnetimes the artist travels
,,''
and occasionally his path crosses those of more colorful persons:

Cellini, who was not an interesting artist, led an exciting life,
which kept him from the difficult business of art.
The individual in search of a personal expression, when confronted with the local stock of possibilities available to him upon
his entrance, must select the components he will use. This gradual
accommodation between temperament and formal opportunity
defines the artistic biography. Our evidence is limited to careers
that have withstood the assaults of time: we have only the "successful" outcomes to all these chancy adjustments between the
individual and his moment, and from them all only a few types
emerge, of which our knowledge is confined to Europe and the
r~ST AND SLOW HAPPENING
far East. As a literary genre, artistic biography was not practiced

elsewhere.
By definition a formal sequence exceeds the capacity of any
individual to exhaust its possibilities in one lifetime. He can
nevertheless imagine more than he can execute. What he exe-
cutes obeys a rule of sequence in which the positions but not the
jntervals are determined. Both what he imagines and what he
executes depend upon his position in the sequence, upon his entrance into the form-class (p. 33). An affinity exists between
each of these opportunities and the corresponding human temperament.
There are slow-paced, patient painters, such as Claude Lorrain
and Paul Cezanne, whose lives contain only one real problem.
Both men were alike in their dedication to the portrayal oflandscape and they were alike in fmding outmoded teachers for their
effects. By relying upon his Bolognese predecessors, Domcnichino and the Carracci, Claude renovated the landscape of
Romano-Campanian antiquity. CCzanne turned to Poussin, like
so many French painters with an interest in tectonic order. The
resemblances are not mere biographical coincidences nor are
they temperamental affmitics alone. The anonymous mural
painters of Herculaneum and Boscoreale counect with those of
the seventeenth century and with CCzanne as successive stages
separated by irregular intervals in a millenary study of the luminous structure of landscape, which probably will continue for
many generations more upon equally unpredictable rhythms.
The type flourishes only in those urbane periods when the ascendancy of special vocations allows persons of a ruminative
tendency the leisure to achieve their difficult varieties of exce1lence.
Under these conditions, and for as long as the old pictures or
their derivatives survive, painters of a certain temperatnent
will
feel summoned to meet their challenge with a contemporary

performance. Ingres continued upon the lines marked by Raphael;
88
SOME KINDS OF DUI\ATJQ~
Manet accepted the challenge put before him by Velazquez. 'I'he

modern work takes Its measure from the old: if it succeeds 11
adds previously unknown elements to the topography of ;1
form-class, like a new 1nap reporting unexpected features inle
familiar but incompletely known terrain. Sometimes the
seems fnushed: nothmg more can be added; the class of forms
looks closed until another patient man takes a challenge from th
seemingly cotnplete situation, and succeeds once more in et1~
larging it.
Altogether different fiom these ruminative artists of a single
problem are the versatile tnen. Their entrances m.ay occur at
of the eighteenth century witnessed new and striking fashions,
either of two junctures, of social or technical renovation. The
remarks are in danger of suggesting. But it is also unhistorical to
ma;
'
"
~~ST 1\.ND SLOW HAPPENING
two kinds sometimes coincide, as in the Renaissance. Technical

renovation is like a spring thaw: everything changes at once,
Such m01ncnts in the history of things occur when new tech~
niques suddenly require all experience to assume their mold:
directors of cinema, radio, and television have thus transformed
our world in this century, and the Vasari who should find his entrance ready for him about a generation from now, will then
record and magnify these legendary figures of sound and shadow
whose myths already rejoin those of classical antiquity.
The other moment for the appearance of the versatile men
occurs when a whole society has been resettled along new lines
of force after great upheavals, when for a century or two the
endlessly complicated consequences, in1plications, and derivations of novel existential assumptions must be set in order and
exploited. The greatest concentration of these versatile artists

appears in Renaissance Italy, where they flourished as a recogmzable soCial type w1der the patronage of merchant princes and
small nobility, of popes and condottieri. Alberti, Leonardo, and
Micl1elangelo are its most celebrated Italian representatives;
but there was no fundamental artistic renovation comparable to

that of the fifteenth century in Italy. In general, this renewal has
to occur in works of art and among groups of artists, and it
cannot cmne frmn government decrees. There need be no reno-
vation at times wheu ample future scope still appears in the current traditions. As the versatile man is called into being by the
time of renovation, so the patient student of single problems
flourishes in a time of settled futurities.
It would be unhistorical to suppose that any period of time
ever has a uniformly patterned structure such as the foregoing
represent a given period in the history of architecture, like the
Periclean age, as a tin1e of unpatterned or unlimited possibilities.
Certain aims had already been accomplished, and the outlines of

possibilities were apparent. The men of the generation of
Mnesicles had little choice but to move from the position just
gained to the next position as it appeared to them from their
11ew
vantage as professionals estimating the chances of su.ccess and
failure among the public of the day.

A very few versatile men with favorable entrances also have
in common the ability to assume in rapid succession a large nun1ber of the available positions and thus to anticipate their successors
for several generations, even adumbrating another new series
before the one in which they are engaged has been played out.
is the most notable of these proleptic artists;
Phidias may have been another. Such men prefigure in the work
I of a few years the series that several generations will slowly and
laboriously evolve: they are able by an extraordinary feat of the
imagination to anticipate a future class of forms m relatively
complete projection. The feat is not easily visible to their con-
I Michelangelo
temporaries. It appears more dearly to historians with panoramic
Jefferson was an even rarer type, the artist-statesman.
These epochs of social displacement when new masters take . hindsight, long after the event. It induces the idea that change
control are of course not always periods of artistic renewal. The I can be brusqued into premature readiness by the action of such
revolutionary transformation of French national life at the close I exceptional individuals.
90
,
SOME KINDS OF DURA'flOt-l'
Certain other biographical patterns among artists can be identi.

fied. The number is small, perhaps only because so few biog.
raphies of the world's artists and artisans are preserved, but more
likely because the variation of type is inherently small among
the lives of inventive persons. Thus Hokusai resembles Uccel]0
as an obsessive painter of a type to which Piero di Cosima
Rembrandt, and Van Gogh also belong as lonely and withdraw,;
men to whom painting was a complete life. They are neither
ruminative and patient, nor versatile and proleptic, but solitary
men who totally occupy the positions given them upon their
particular entrances. Even architects, whose work dcn1ands gre~
garious virtues, can be found among their number: the careers
of Francesco Borromini and Guarino Guarini belong to this ob.

sessive group by the strange, intense, and complete reality with
which they clothed their lonely, imaginary worlds.
A contrasting type is the evangelist whose mission is to improve the visible world by the imposition of his own sensibility.
No m'\ior architect of the present century has been able to practice
witl10ut assuming this evangelical garb. The missionary-artist
is often a vigorous teacher and a prolific writer flourishing best
when he is in command of academies of right practice. J.-A.
Gabriel, tl1e leader of French architectural taste, or Frank Lloyd
Wright and Sir Joshua Reynolds are examples. Each of them
exercised an autocratic taste, founded upon selected conventional
traits, which the older men took from an aristocratic tradition,
and which Wright took from H. H. Richardson and Louis

Sullivan.
Two distinct kinds of innovators occur in the history of art.
The rarest of them are the precursors, like Bnmelleschi, Masaccio,
or Donatello, whose powers of invention find a proper entrance
no oftener than once every few centuries, when new domains of
knowledge are opened through their efforts. The other kind is

the rebel who secedes from his tradition the better to have his
own way either by altering its tone like Caravaggio, or by
challenging its entire validity, like Picasso. The precursor may
91
'fAST AND SLOW HAPPENING
Jso be a ruminative or an obsessed artist like Cezann~. Witho~t
Ibeing a rebel the precursor quietly lays ne:" fonndauons w1thm

an old preserve. The precursor can have no mutators: he" always
sffi generis, while the rebel appears in crowds, because the :"~Y of
the rebel is easily imitated. The precursor shapes a new clvilzauon; the rebel defmes the edges of a disintegrating one.
The genuine precnrsor usually appears upon tl1e scene d a
provincial civilization, where people have lo~g been the rcclplnts rather than the originators of new behavwr. The rebel hke
~icasso finds his situation at tl1e heart of an old metropolitan
civilization. The necessary condition for a precursor is that his
activity be new; for a rebel that his be old. Precursors have to

mold their work in the shell of an older guild, like Ghiberti, who
was apprenticed as a goldsmith, or find a place for it at the
bottom of society like the creators of the early cinema. The
rebels, on the other hand, who shape their lives on the fringes of
a society they despise, have to form a new civil condition in the
pursuit of some integrity of life and work. Gauguin's is the mo~t
celebrated example, in the special form of the artlSt as a bourge01s

refugee continuing the romantic convention of Parisian bo-
hemianism among primitive villagers in Tal1iti.

These six types of careers: precursors, hommes
toutfaire, obsessives, evangelists, ruminatives, and rebels all exist at once in

modern occidental society. Of course they cannot all occupy the
same formal sequences. Each seqnence affords the opportnnities
of its particular systematic age to only that group having the
temperamental conditions for a good entrance. Thus telcvtston
calls upon a special temperamental class for its directors today,

but it will require another kind of temperament in a later decade,
while the men then who might have been television directors
today, may turn to another theatrical form which will be more
prepared to give them their entrances.
As we look at other societies and at earlier tin1es, it becomes
impossible to document the existence of any variety of careers.
The bohemian carmot be identifted before the seventeenth cen-
"
'"~ o: ""'""1~" '"" -~"

mo
tury ill Eur~pe or Cluna. Indeed 111 older soc1eues lt lS hkely that
ww
"
such
the borderhnes between careers were much less apparent than
societies the 1nanufactures show change, lt 1s true, but that chat~ge
they are today, that ruminative and obsessed artists merged,
is the change of casual drift, of cumulative habit, of routme
hke precursors and rebels, or versatile and evangelical men, with..
repetition with minor variations, which through the generations
o~t the clear separation. not~d today: fu the Middle Ages the inyield a characteristic pattern.
.
.
The pattem resembles that of the changes ill the thmgs made
div1dual artlst remams illVlSlble behind the corporate fa,ades of
church and guild. Greco-Roman and Chinese histories alone rcunder more complicated social structures. It shows the expected
port in any detail the conditions of individual artists' lives. A few i progression from early to late sy~tem:tic age within th~ different
names and hnes of text are all we have about Egyptian dynastic : classes of pottery, honsmg, and ntualmstruments. Dlstmct formartisans. The records of the other civilizations of antiquity in
classes succeed one another. Within the scope of three or four
generations a clear shape corresponding to the physical identity
America, Africa, and India tell nothing of artists' lives. Yet the
archaeological record repeatedly shows the presence of conof the tribe can be detected by the attentive student. But the pronected series of rapidly changing manufactures in the cities, and
gression, the succession, and the shape all are more muted and
slower ones in the provinces and in the countryside, all maniless distinct than in larger societies, and the pace is slower. Less
f~sting the pres~nce of persons whotn we can call artists. They
happens; fewer inventions occur; and there is litde conscious
did not all flounsh together at the same time as they do today in
self-definition of the tribe by its manufactures.
the great cities of the principal states, where more classes of forms
This contrast selects only extreme cases-the tiny tribe of a
coexist than there are talents to staff them. For instance, profew scores of families strnggling to survive, and the vast metropogressive painting today mainly attracts rebels, willie the prelis with its crannies and ledges sheltering the meditations of many
cursors and the ruminatives either paint as obscure men m1der a ; inventive minds-to exemplify the most sluggish and the most
protective coloration that shields them from success, or they
vertiginous kinds of change. Between them are at least two. inbelong to other guilds like stage design or advertising art, where
termediate positions. It is too simple to suppose that the gradient
,I
gnilds of arusans who are exempted fro':' growmg food.
h1
their peculiar dispositions are more urgently needed than in
is a continuously smooth one. Between London or Paris, and the
painting for fashionable dealers. fu older societies, fewer sequences

were in active development at one time, and the opportunities
for the whole spectrum of temperaments were correspondingly
more modest.
Tribes, courts, and cities. A provisional explanation of the fast
and slow changes in the history of things now emerges. The in-
forest tribes of Amazonia or New Guinea, the gradations of

social organization are far from continuous; they are more l~ke
a motmtainous escarpment with bold high tiers of cliffs separatmg
several terraces.
Absolute demographic size is irrelevant. Small cities have
generated the principal events of history more often than the
tervention of men to whom art is a career, of men who spend all
megalopolis. An urban setting is a necessary if inadequate con-
their time in the production of useless things, occasions the shift

from slow to fast happening. In tribal societies of a few hundred
clition for the occurrence of fast happening. The setting is urban

when the privileged city dwellers are not a self-sufficient tribe
by
but a group of rulers, artisans, merchants, and parasites dependent
endless toil in a harsh environment, there is never enough margin

beyond snbsistence to allow the formation of those specialized
upon the labor of food-growing rural populations scattered

throughout the countryside.
families, where everyone Inust raise food most of the time
94
SOME KINDS OF DURATIO~
Urban life alone is not enough. The provinces all have cities
but the tedium of provincial city life is proverbial. It is tediot~
because the provincial city is like an organ that usually cau only
receive and relay 1nessages frmn the higher nervous centers: it
fhST AND SLOW HAPPENING
95
generating centers for important commissions; the surr~~nding
countryside was the receiving province, and the larger c1t1es depended upon the intermittent presence. of the royal court for
their access to favor and power. Followmg the Renaissance, the
cannot issue many messages of its own, other than of pain or
capital cities rose to greater importance, but until 1800 tl~e ~any
discomfort; and its active elements perennially emigrate to the

true centers of happening, where the central decisions of the
whole group are made, and where the concentration of power
draws together a class of patrons for the inventions and designs
of the artist. These are true metropolitan conditions. They are
the only necessary and adequate conditions for the appearance of
the rapid historical pace that has always marked life in the chief
snrall princely courts of Europe were true centers of artistiC excellence, taking only rarely a provincial relationship to the great
cities, which often were themselves tnore provincial in certain
respects than, for instance, the duodecimo duchy of Weimar

which supported Goethe. Today the leveling of the world by
mass entertainment and industrial monotony is so complete that
only the richest cities and a few university towns remain as
cities of man.
citadels of learning and discrimination.

11ms we have four societal phases to consider in this discussion
The different historical climates of patronage afforded difof the velocity of artistic events: (r) tribal life, face-to-face with fcrent settings for the six kinds of career we discussed earlier.
nature and unable to afford artisans; (2) provincial towns and The precursor and the rebel could have no place either in a tribal
cities with their derivative arts, including those capital cities ! society or in provincial life where non-conformity leads to grave
which specialize only in government; (3) tribal societies that in- i punishments. Only the richest centers of power can support the
clude professional craftsmen with originating powers; (4) cities : homme tout faire or his academic counterpart, the evangehcal
or courts which issue the invisible yet ultimate orders. The tabu- : chef d'ecole. The obsessed men and the rumiuatives can work
1
lation pertains to Greco-Roman civilization, to Chinese dy- anywhere, but their training and formation require that they
nastic society, and to the modern world since r8oo with its politi- spend some time at court or near the sources of patronage to
cal division by ruling states and colonial empires, and by capital become part of the stream of fast happening.
cities drawing to themselves the flower of provincial talent. It In other words, there are ouly two significant velocities in the
is valid also for the urban civilizations of ancient America. The history of things. One is the glacier-like pace of cumulative drift
ranking of provincial cities as less favorable environments than , in small and isolated societies when little conscious intervention
tribal societies which have their own craft traditions may seem occurs to alter the rate of change. The other, swift mode rearbitrary, but it is justifwd in regard to the conditions of original sembles a forest 'fire in its leaping action across great distances,
artistic activity: an Ashanti bronzeworker in mid-nineteenthM , when unconnected centers blaze into the same activity. The hiscentury Africa was perhaps more favorably situated as an artist ; tory of a recent invention has many instances of this apparent
than his contemporary colleagues in Chicago or Mosul, who
action at a distance, as when two or more professional men withwere limited to the making of provincial replicas and useful , out knowledge of one another's work, com_e to the same solustereotypes.
: tions independently yet simultaneously, from common premises
In medieval Europe before 1400 a different scheme is needed.
and similar method.
The feudal courts, the abbeys, and the cathedrals were the
An important variant of fast happening can be described as
SOME KINDS OF DUnAt[Q~
!!IE SHAPES OF TIME
97
classes we are considering contain events related as progressive

the pace of intermittent duration: a problem receives earl
.
~ H
,
y~
olutions to problems of which the requirements are modified
tentton,
e cron s stea1n engine, or Greco-Roman still-]
I
t
h
b
f
1fe
~Y each successive solution. A rapid succession of events is a
pamung. 1 t e a sence o supporting conditions and reinfo
techniques, the invention then languishes in obscurity for rclng dense array; a slow succession with many interruptions is sparse.
tmtil tl1ose cond"1tions are produced that allow themany
ln the history of art it occasionally happens that one generation,
centuries
fo
I
t
th"
rru
and
even one iudividual achieves mmy new positions not only
c ass o reenter e mventlve conscience of another civiliu
.
.
.
d
'
m%
ThIs mtermlttent mo eo =ppeniug has rapid instants ru1d ." in one sequence but in a whole set of sequences. At tl1e otl1er
" d to t )1e pnnc1pal
centers of civilization but it is ' irre It.1 li 1 extreme a given need will subsist for generations or even cencmnme
.
d"
,
~~
m pacde, m 1ts full consequences are extremely slow to be ex. : turies without fresh solutions. We have already examined these
tractc .
. occurrences under the heading of fast md slow happeniug. They
The full range of artistic careers, from precursors to rebels have been explained as contingent upon position iu the series
th~s can unfold only tmder metropolitan conditions, when ; and upon the varyiug pace of invention in different centers of
population. Let us now look at further varieties in the array of
w1de selectmn of active sequences is avail.ablc. Fast happening
serial positions.
depends upon fav~rablc conditions of patronage md career,
Positional values. An Apostolado by Zurbar:ln is a unified at1d
wlule slow happenmg charactenzes provincial or tribal settin
where neither patronage nor possibilities of career exist to stim~: coherent work of art.3 It consists of twelve or thirteen paintings
that portray the apostles. Each picture can be seen alone, but the
late a more rapid exploration of the various classes of forms.
painter's intention and the patron's wish were to have the entire
group seen together as a corporate work of art in a prescribed
THE SHAPES OF TIME
sequence and occupyiug a specified space. Many things have
simil.ar group properties which require them to be perceived in
The nu~ber of ways for thiugs to occupy time is probably
a predetermiued order. Buildings iu their settiugs are a sequence
no more unhm1ted tl1an the number of ways in which matter
~ccupies space. The difficulty with delimitiug the categories of of spaces best seen iu an order iutended by the architect; the
sculptured faces and separate parts of a public fountaiu or monutlm.e has always been to find a suitable description of duration,
ment also should be approached as planned; md mmy paintiugs
which would vary accordiug to events whil.e measuriug them
were originally meant each to have a ftxed position in a sequence,
agaiust a fixed scale. History has no periodic table of elements
from which a total narrative effect might arise.
and no classification of types or species; it has only solar tim~
In such corporate works of art, each separable part has a posiand a few old ways of grouping eveuts,Z but no theory of temtional value in addition to its own value as an object. Usually our
poral structure.
comprehension of a thiug is iucomplete until its positional value
If my principle of classiug events be preferred to the impossible
can be reconstructed or recovered. Hence the same thing can be
conception that every event is unique and unclassable, then it
quite differently valued as m object separated from context, md
mu~t follow that classed events will cluster during a given portion
as
a corporate work in its intended setting. Greco-Roman art
of tlme m an order varyiug between dense and sparse array. The
2. J.H.J. van der Pot, De periodisering der geschiedenis, Ben overzicht der theorieC'n
(The Hague, 1951).
J. M.S. Soria, The Paintings ofZurbardtz (New York, 1953), e.g. Nos. 78, 144,
145
'"
m~
"""'
0>
"'"""<>1M
l "'""
depends greatly upon positional values: Philostratus' lmaai

. pe d'1menta! narratives at Aegina or onones
andtl1e contrastmg
th
Parthenon _are examples. Often the positional value adds to the
mterpretat!On, as when Old and New Testament stories are paire~
by the parallels, antetypes, and prefigurations that have bee
parts of Christian teaching since before Prudentius' Dittochaeuni:
To these obvious values accruing from spatial position we q,
add another category accruing from temporal position. Becaus:
no work of art exists outside the linked sequences that coimeq
every man-made object since the remotest antiquity, ever
thing has a unique position in that system. This position is marke~
by coordinates of place, age, and sequence. The age of an object
~s not only the customary absolute value in years elapsed since
It was made: age also has a systematic value in tern1s of the positlon of a thmg in the pertinent sequence.
Systematic age enlarges our conception of historical position,
The 1dea requires that we relate each tiring to the several changing
systems of forms in which its occurrence belongs. Hence the
us~al names of things are inadequate because they are too general.
It 1S too broadly inclusive to speak of the systematic age of an
Eughsh country house like Sevenoaks, built during many centunes; we can consider only its parts and the ideas for their mrification when we establish systematic ages. A great stair, for example, if built in I 560 would be a very new form in its class
for .great stairs began to be built in Spain and later in Italy n~
earher. m:u: the opening years of the sixteenth century.' A remo~elmg m 1760 that imposed Ullsymmetrical, gothicizing
v~nety upon an older house by John Webb (r6n-74) was likeWISe very new in the form-class of picturesque arclritectural
4 K. Lehmann-Hartleben, "The Imagines of the Elder Philostratus " Art
Bulletin, 23 (1941), 16-44. A. Baumstark, "Frtihchristliche-Palastinische' Bildkompositionen in abendliindischer Spiegelung, '' Byzatttinische Zeitschrlft 20 ( 191 1)
~~
'
5 N. Pevsner, Att Outline of European Architecture (Baltimore, 6th ed., 1960),

pp. 47-48 on square open-newel staircases in Spain.
0>
"
effects, while the Webh nucleus was late in its class of Italianate
fOrms.
I Periods and their lengths, Thus every thing is a complex having
not only traits, eacl1 wim a different sy~te":atic age, but having
! also clusters of tralls, or aspects, each with 1ts own age, hke any
I other organization of matter, such as a mammal, of which me
i blood and me nerves are of different biological antiquity, and
the eye and the skin are of different systematic ages.
Because duration can be measured by the two standards of
. absolute age and systematic age, historic time seems to be com posed of many envelopes, in addition to being mere flow from
future to past through the present. These envelopes, which all
have different contours in tl1e sense that they are durations defined by their contents, can be grouped easily by large and small
families of shapes, We are not concerned now with the diminutive shapes of personal time, almough each of us can observe in
, his own existence the presence of snell patterns, composed of
early and late versions of the same action, They extend through
all individual experience, from the structure of a few seconds'
duration, to the span of the entire life. Our main interest here
is in the shapes and forms of those durations which either are
longer than single human lives, or which require the time of
more than one person as collective durations. The smallest
family of such shapes is the annual crop of costume fashions care..
fully nurtured by the garment industries in modern commercial
life, and by court protocol in pre-industrial regimes where fashion
was the surest outward mark of high social class. The largest
shapes, like metagalaxies, are very few: they dimly suggest
their presence as the giant forms ofhuman time: Western civilization; Asiatic culture; prehistoric, barbarian, and primitive society. In between are the conventional periods based upon me
solar year and the decimal system, Perhaps the real advantage
of me century is that it corresponds to no natural or determinable rhythm of happening whatsoever, unless it be the
eschatological mood overtaking people at the approach of a
roo
SOME KINDS OF
DURA.Tio~ ~TilE
millenary number, 6 or the fin de siecle languors induced after

1890 by mere numerical parallelism with the 1790's of the
French Terror.
There is certainly nothing in the history of art correspondin
. tent11 part. Yet when we consider theg
e1'ther to a century or to Its
conventionally recognized span of Greco-Roman art, we are
confronted with ten centuries-one millennimn, fron1 c. 6oo n.c,
to c. A.D. 400, as a possible duration. But no other millenary
durations cmne to mind, and the Greco-Roman one depends
upon arbitrary cuts at both beginning and end.
Rather than take space to review the cyclical conceptions of
"necessary" historical durations which belong to another kind of
speculation, we might consider more closely the periods that are
said to correspond to known "working" duration~ and intervals
in the history of things. A year is surely valid-it contains the
round of the seasons. Many sorts of work fit into its span. The
human frame ages perceptibly in a year, and forward plans in
any detail are put forth year by year.
The Roman lustrum of five years has returned to favor in the
plam1ing of the socialist states. It is certainly more practicable
a span for human affairs than the decade, which is too long for
practical plans, and too short for many records. The decade is
only a decimal position in the century. Both the decade and the
century are arbitrary intervals rather than working durations.
Other civilizations preferred a shorter duration, like the 52-year
period of the Mexican peoples, composed of four indictions of
13 years each. It corresponds roughly to the length of an adult
existence. This correspondence may have been coincidental, as
the 52-year cycle was constructed for astrological use by an
agricultural people, upon the combination of the solar year with
a ritual farmer's year 260 days long.
More to the point than our century is the length of a human
generation. It is differently calculated according to different pur6. H. Focillon, L'An mil (Pads, 1952).
SHAPES OF TIME
IOI
1poses and in different periods: in demogr~phic studies it is esti

1rnated as 25 years, but m general h1Story It ruus to 32-33 years.
I fhe longer duration probably corresponds better to the happening of general events, while the lesser figure matches the simple
~ happening of biological replacement. Three generations approach
! our century, and it might be thought that such a cycle would be
' useful in our studies, corresponding to the revolutions of fashion,

whereby the taste of the grandparents in clothes and furniture,
after having been rejected by the children, returns to favor in
the generation of the grandchildren. In practice, however, these
cyclical returns of fashion require less than half a century, and
they are subject to defonning interferences from other sectors of
happening, as Kroeber and Richardson showed in their remarkable study based on ladies' fashions since about 1650. 7
The indiction as module. We have no customary calendrical
duration corresponding either to single life expectancies (the
Biblical "threescore and ten" is the exception) or to the statistical
estimate of the length of a generation (25-33 years). Both of
course have rapidly changed in the past century. Before r8oo
life expectancy differed little from the expectancies recently estimated for paleolithic man, who at birth like Romans and ancien
regime Frenchmen, would be expected by a modern actuary to
enjoy less than 25 years. In their own time, of course, these per7 Jane Richardson and A. L. Kroeber, "Three Centuries of Women's Dress
Fashions. A Quantitative Analysis," Anthropological Records, S (1940), No. 2
(Univ. of California): "the how of the unsettlement of dress style is not dictated
by the sociopolitical conditions; that must be due to something in the set of the
fashions themselves...-.-something within the structure of fashion ... " Elsewhere
and later Professor Kroeber, in Style and Civilizations (Ithaca, 1957), returned
to the theme of immanence. On the independent appearance of similarities in
corresponding phases of different cultures he wrote (p. 14.3) that the phenomenon
(e.g. Japanese "impressionism" in the painting of Sesshu, A.D. 1419-1506) is
"immanent in a scheme of development or growth ... as a toddling gait is a
function of infancy, tumultuousness of adolescence, a stooping body of age."
He concluded, however, that "assumption of immanent forces is best left as a
last recourse. , , , More likely are secondary or pseudo immanences: internal
cultural sets of varied strength which have been gradually developed as a result
of external forces" (p. 159).
I02
SOME KINDS
OF DURATIQ~
sons were afflicted by no such idea: some people lived to he old

and others died young, but no one kept very close records. 'I'h;
working divisions of a human life were therefore much the satne
as they are today. Infants, children, and adolescents; young
mature, and old adults were the six usual age classes. Of the,;
only the last four concern us, embracing the productive life-span
of artists and craftsmen, from about twelve or fifteen years of
age upwards. Hence the working lifetime of the man of art may
be put at about 6o years, of which probably only so are spent at
full power. We may take 50-60 years as the usual duration of an
artist's life, rather than any figure over 6o. Its four periodspreparation, followed by early, middle, and late maturity-each
lasting about IS years, resemble the indictions of the Roman
calendar as well as the climacteric periods of developmental
psychology. In any case, the term "indiction" is better than the
"decade" of conventional usage. The decade is so short that it
often fails to match the significant shifts in an artist's life-work.
Periods longer than the indiction miss them again: something
more than ten and less than twenty years best corresponds both
to the vital periods of biography as well as to the critical stages
in the history of forms.
Let us turn from artistic biography to the duration of the
linked series of events which principally engage us. Certain classes
of teclmical developments in the history of art require about 6o
years for their formulation and 6o years for the first systematic
applications. The early history of large-scale rib-vaulted construction in Gothic architecture began in the Ile de France about
a. D. I 140, and all the components of the Gothic spatial assembly
were in existence
'I
by
1200.
Some students of the question see the
first formulation in Anglo-Norman territory, and its critical initial period as enduring from Io8o to Il40. Two distinct stages
of inventive elaboration are present; the point here is that each
lasts about 6o years. A similar phenomenon is the appearance of
Greek vase painting in two stages of about 6o years each, hinged
c. 5ro B.c. Other examples are the development of the pictorial
taB
SIIAPBS
OF
I03
TIME
system of the Renaissance in central Italy during about sixty

ears in the fifteenth century, or the appearance of tall steel-frame
Yrchitecture after I850 in the United States and Enrope. Each of
jd1ese series had its prodrome of scattered experiments preceding
'
.
! the rnam campaign.
i This suggestion of doubled 6o-year durations for certain im-
! ortant sequences in
i
~rior
the history of art is an empirical one. No

idea of "nece.ssary"
go:erned. the
evolutio~ ~1as
observati~n
0fiengt:h. Small differences of opnuon mtght anse about hegm: 1ung and ending dates, but no one would contradict the magnitude itself in any one of these instances, especially when it is
dear that we are not talking about "styles of art," but only about
01e history of special forms among related examples occurring
in limited regions.
The duration embraces only the generations of the invention

and mise au point, m1til the time when the system attained general
.currency in a much larger region, as a completed entity suitable
for indefinite repetition. This time corresponds to what has often

been called the "classic" stage: the age of Phidias, the great
. cathedrals c. I200 in northern France, or the High Renaissance
c. I500 in Italy.
Non-European examples with similar double-stage durations
of 6o years each are known, such as Maya sculpture in the fourth
and fifth centuries a.D., or Japanese wood-block prints after
1650. The examples all seem commensurate, involving novel
teclmical, thematic, and expressive resources and providing fresh
means of achieving a wide range of structural or descriptive

aims. The general validity of the doubled 6o-year duration of
about 120 years, in two stages of tentative fonnulation and rapid
exploration requires much more verification and many more
examples before we can know whether or not it depends upon

a specific kind of cultural organization. The pre-Columbian
American evidence already strongly suggests such a minimal
duration for urban civilization.
A possible test case is offered by paleolithic painting, with
104
I'
!:
'
I
.1:
'
"
~ 0~.:1 '"' "

O>
two main regional variants in the Dordogne and the Cantabrian

caves which vary in types and expression about as much as
French and Spanish painting in the seventeenth century. Our
question concerns duration. Although some scholars ascribe
thirty or forty centuries of ice-age time to the artistic traditi011
of these paintings, it is on small evidence, and the possibility
still exists that each group of paintings, exemplified at Lascaux
and Altamira, was produced by a few generations of painters
reprieved from the routine of nomadic existence for a few ce11..
turies by some favoring accident of time and place.' If that were
the case, then our typical minimal duration for the ages of invention might be shown to occur as a ten1poral category in
human effort, independently of cultnral mode.
The separation into paired uo-year periods-formulation followed by systematic extension-is drawn mainly from the historical accumulations the1nselves. A comtnon span connecting
history and biography is snggested by the uo-year period, which
of course is only an approximation for convenient reference,
Sixty years is also the length of the single complete productive
life. Very few if any artists, however, remain npon the crest of
the wave for so long. Staying power probably depends on "entrance." Otherwise, the individual's powers of invention usually
are limited to youthful years. If later in life he achieves new
forms, they are likely to be mature realizations of early intimations. The significant phasing of most lives is by portions like
the indiction of fifteen years. The full, active life-span embraces
about four such indictions. As we study the pulses of those linked
series of events-Anglo-French Gothic architecture, Greek pre-
o TIW
!05
Idassic sculpture, or Central Italian painting-all are similar or

IiuJked series of related solutions spreading over about 120 years
[n tWO Stages of 60 years, each divided by artistic generations or

of r 5 years.9 The indiction measures many distances
:ill historical duration. It is a measure drawn from experience,
'like paces, feet, and ells, providing us at least provisionally with
a module to connect things with individual lives and the genera, tions of 1nan.
Even longer durations are now the next units upon which to
base observations of shape. One likely span is near three centuries,
and it is the approximate duration of each principal stage of
several civilizations whose durable 1nanufactures have been recovered in some detaiJ.Io An example is the length of the major
l,Ijp.dictions
9. This indiction is an empirical observation. It does not derive from any

of generation by birth-year, such as W. Pinder's in Das Problem der
Gcnerationen (1928). Its resemblance to the table of generations averaging fifteen
years each from 1490 to 1940 in European letters, as presented by H. Peyrc: Les
Glnirations fittt!raires (Paris, 1948), is fortuitous. M. Peyre observed that occastonal
~ 10 rt generations of IO years occurred when the work demanded: in other words,
that the independent variable is not the generation but the work it has to do,
as in the sense of my argument. M. Peyre, whose remarks here are brief, wrote
as follows (p. 176), "11 y a des epoques maigres ou mal dCfinis ... il yen a au contraire aU l'on vit plus intensCment, et aU l'on brUle les emotions, vide les idees,
Cpuise les hommes avec fievre." Other French periodizations by fifteen-year in~
tervals occur in the works by J.L.G. Soulavie, PMces itufdites sur les rCgnes de Lnu{s
XIV1 Louis XV et Louis XVI (Paris, 1809) and L. Benloew, Les Lois de l'histoire
{Paris, I881), scanning the period rsoo-r8oo by "evolutions" averaging 15 years
each.
10. In advancing this division, I bear in mind the astringent observations by
R. M. Meyer, "Principien der wissenschaftlichen Periodenbildung," Euphorion,
8 (1901) 1 1-42. He shows that periods in history are neither necessary nor selfevident; that all development is continuous; that periodology is entirely a matter
of convenience dependent mainly upon aesthetic considerations, especially in re8. The most recent radiocarbon dates from Altamira in Spain and Lascaux in
gard to the proportion and number of periods.
France suggest that the two systems of cave paintings are just possibly contempoJ.H.J. van der Pot, De pen'od{sering der geschiedenis (The Hague, 1951), concurs
rary within the span of one quarter-century (Altamira charcoal, Magdalenian
in stating that "it is impossible to formulate directives for the correct proportion
III midden deposits, 13,540 B.c. 700; Lascaux charcoal from lower fissure of
of the duration of the periods in relation to each other, and for their number."
Grotto, 13,566 B.c. 900). H. Movius, "Radiocarbon Dates and Upper PalaeoHe urges that periodization, which "forms the quintessence of history," should
lithic Archaeology in Central and Western Europe," Current Anthropology, 1
be idiographic (i.e. not based on "historical law") and endocultural, arising
(1960), 37o-72. Of course at the other extreme the dated materials may be 1,600
from subject-matter rather than from exocultural "necessities of geography,
years apart.
Mology, or the psychical qualities of man."
1nystique
ro6
SOME KINDS OF DURAT!Q~
l'
tHE SHAPES OF TIME
107
divisions of the pre-Columbian American civilizations for some revival in this centnry after long disuse, first with Gaudi and in
two thousand years prior to the Spanish Conquest. This frgure ltater ferro-concrete studies of ribbed structure.
Such intermittent classes are easily recognized as being comwas at first-early in this century-only a guess based upon
several converging lines of evidence, but it recently has been ' posed of impulses so separate that distinct groups of inventions
confirmed by radiocarbon measurements showing the intervals : are really present. Yet the new group would be impossible withbetween the crises in the archaeological record to be of this order out the tradition and the accomplishments of the earlier group
buried deep in its past. The old class conditions its new conof magnitude. It measures the duration of artifact stages, such as
the early, middle, and late pottery crafts of the villages of Mexico tinuation more pervasively than the living generation usually
or the central Andes tmder tlte rule of tlteocratic states during cares to retnember.
the millennitrm before A.D. 1000.
The history of transcultural diffusion in turn contains several
Closed series were mentioned earlier (p. 35) as an illusory and kinds of motion. Under pre-industrial conditions of travel, great
artificial conception, since no class ever finally closes, being distances as between Imperial Rome and Han-dynasty China
always subject to renewed activity when novel conditions re- were traversed at first only by the most useful inventions. Sysquire it. We nevertheless might distinguish now between con- tematic missionary efforts to transform the entire symbolic
tinuous and intermittent classes. Continuous classes concern only structure of Chinese civilization, by Buddhists from India after
the largest groups of things, such as the whole history of art, or the sixth century, and by Christians in the sixteenth and seventhe most common classes, such as household pottery, of which teenth centuries were temporarily successful, bnt they could
the 1nanufacture never ceases.
never have been begnn without the ample prior tradition of
Intermittent classes. Two kinds of intermittent classes immedi- useful learning carried to China by commerce. Occasionally, as
ately appear: those which lapse inside the same cultural grouping, in the sixteenth-centnry Spanish conquest of Mexico and Peru,
and those which span different cultures. Intermittent inside the abrnpt military action replaced these motions of commercial and
same culture are such arts as enameled jewelry, which lapsed missionary penetration. Conquest was followed at once by masafter the Renaissance, excepting for infrequent resumptions such sive European substitutions of useful and symbolic behavior for
as the jewelry of the Faberge family in nineteenth-century Russi~ native traditions. Only tbe useful items new and necessary to
or the work of John Paul Miller in Cleveland, who has resumed Europeans survived the wholesale destruction of the native
the gold granulation technique commonly used by Etruscan Atnerican civilizations (potatoes, tomatoes, chocolate, etc.).
Very few native art forms have so far survived this wreck.u
goldsmiths. Tempera painting was long disused because of the
ascendancy of oil painting in the fifteenth century, tmtil a variety The village art of Mexico has a few muted or commercial recalls
of nineteenth- and twentieth-century conditions led to its re- of Iudian antiquity. The principal frgures of twentieth-century
vival, as in the academy of tempera painting that flourished at
I I. The eschatology of civilizations is a subject still undisturbed by deep thought.
Yale nntil 1947, based npon the fourteenth-century text by See W.H.R. Rivers, "The disappearance of useful arts," Festskrift tillegnad Edvard
Cellllino Cennini as edited by D. V. Thompson, and taught by Westermarck (Helsingfors, 1912), pp. 109-30; or my essay entitled "On the Colonial
of the Motifs of Pre-Columbian Art," Essays in Pre-Columbian Art
Lewis York, in order to prepare stndents for the mural-painting Extinction
and Archaeology for S. K. Lothrop (Cambridge, Mass., 1961); and A, L. Kroeber's
commissions that the public works program of the 1930's had remarks on "The Question of Cultural Death" in Cotifiguratlons of Cultural Growth
made possible. True rib-vaulted construction had an astonishing (Berkeley, 1944), pp. 818-25.
~.~
-_-,{
!
108
SOME KINDS OF DUR.A.'riOt-.~
iit.fiB SHAPES
Oll TIME
109
'
Mexican painting, Orozco, Rivera, and Siqueiros, all amplified
the Indian past, and certain foreigners enlarged many native
themes in their own terms. Frank Lloyd Wright renewed , 11
experimentation with Maya corbel-vaulted compositions that
had lapsed since the fifteenth century in Yucatan, and he resumed
it with the technical resources of his time at the point where the
Toltec-Maya builders of Chichen ltza desisted (Barnsdall House,
Los Angeles, 19zo). Henry Moore, the modern British sculptor,
!be ancient messages thus continue indefinitely to echo down the

i10rridors of time is nnclear when the record still isso short. There
: evertheless has been only one brief moment when a people
1
1avr consciously sought total independence from past formulas
Of expression. This moment lasted for a generation after I9ZO
~Europe with the label of fnnctionalism. Historical parallels are
erhaps to be sought in various iconoclastic movements of re1
ugious reform, as in Constantinople, in fifteenth-century FlorJ
likewise returned to variations upon the thetne of angular re~ mce, in Islam, in Jewry, and in Puritan Protestantism. Under the
cumbent figures, based upon a Toltec-Maya tradition of about

the twelfth century. John Flanagan, the American, produced
compact animal studies in an idiom with fifteenth-century Aztec
antecedents. These twentieth-century continuations of the u11
finished classes of fifteenth-century American Indian art can be
interpreted as an inverted colonial action by stone-age people
upon modern industrial nations at a great chronological distance.
Through its formal vocabulary alone, the sensibility of an extinct
civilization survives in works of art to shape the work of living
artists it1 a totally nnrelated civilization half a millennium later.
The phenomenon is of course possible at every level of historical relationship. It transformed Western civilization most profoundly in the Renaissance, when the unfinished work of GrecoRoman antiquity took possession of the entire collective mind of
Europe to dominate it nntil deep it1to the twentieth century,
even as late as Picasso's illustrations of Ovid's Metamorphoses.
Today classical antiquity has been displaced by even more remote models, from the prehistoric and primitive art of all parts
of the world, in a similar mortmain action as if to round out the
principal contours oflong-unrealized possibilities. In other words,
when people create new forms, they commit posterity at some
remote interval to continue in the track by an involuntary act
of command, mediated by works of art and only by them.
Here is without doubt one of the most significant of all the
mechanisms of cultural continuity, when the visible work of an
extinct generation still can issue such powerful stimuli. Whether
fanctional program all possible products were designed anew it1

1arch of forms corresponding solely to uses, under the doctrinal
Oe]ief that the necessary alone is beautiful.
Arrested classes. As we approach incomplete classes, we should
~riefly explore the related theme of the abandoned, arrested, or
"starved" classes, which have been left unexploited. Examples
lfC fairly common in the lives of unrecognized inventors whose
Jiscoveries remained obscure for many years until a lucky chance
~rought their work to the notice of persons contpetent to conrinue it. Notorious examples occur in the history of science: the
fundamental genetic studies of Gregor Mendel, which went unattended for nearly forty years, are a celebrated it1stance. The
>istory of art has many parallels, such as Claude Ledoux, whose
oeo-classic use of the forms of pure geometry under Napoleon
~refigured the hard abstractions of the "International Style" in
1he twentieth century; or Sir Joseph Paxton, whose prefabrica;on of metal parts allowed him to design the prophetic spaces of
lass and steel in the Crystal Palace of London in 1850-51. Pre0
:ursors like these have the power to produce the solutions to a
!'neral need long before the majority experiences the need itdf. Indeed, the way in which the need is framed often owes its
:jnal form to these premature talents.
After neglect, conquest is the other great occasion for incomJiete classes, when the victor overthrows native institutions and
'places them with extensions of his own. If the victor has
tiluring benefits to offer, like Alexander or Cortes, he makes the
IIO
SOME KINDS OF DURATXoN
ltHE SHAPES OF TIME
III
continuation of many traditions both unnecessary and impossible. learned from the Italian theorists of the preceding century. What
The locus classicus for incompletion is the case of sixteenth- came to Mexico was therefore systematically old, whether it was
century America, when native initiative quickly ceased under the :a
. retarded late medieval ornament, an out-of-date Italian idiom '
blows of the Conquest and the attraction of superior European :or an up-to-date Plateresque decoration.
On the Indian side, Aztec sculpture displays an exceptional
knowledge.
command
of symbolic indications of death and vitality, but it
At the same time the creation of a colonial Spanish civilization
was
a
new
art, fanned from the resources of 1nany subject
in America can be taken as the classic case of extended classes.
These occur when inventions and discoveries 1nade in the parent peoples, perhaps utilizing a tribal tradition of vigorous expressivesociety are passed on to the colony together with the persons-- ness, and probably not antedating the reign of Ahuitzol late in
mechanics and artisans-needed to get up the corresponding the f1fteenth century, less than a generation before the arrival
crafts. Latin America before r8oo is an impressive example of of the Spaniards. The identity of these gifted sculptors will never
Hispanic extension, although innmnerable other cases of smaller be knoW!l. Yet we can be certain that it was a systematically new
territorial and demographic size illustrate the point quite as well, art which yielded to an older art brought from Spain. The diflike the imposition of Islam upon Christian Visigothic Spain, or ference of systematic age was not great, but the differences of
the Hellenization of India by Alexander's armies.

It is in the nature of events that incomplete classes are likely
to be less well documented than extended classes: conquerors and
colonial agents usually keep poor records of the ways of the
people whose traditions they seek to extinguish. Christian ethical
behavior in the Spanish colonization of America nevertheless produced encyclopedists of native culture like Bishop Diego de
Landa for Yucatan, Friar Bernardino de Sahagun for Mexico,
and Father Bernabe Coho for the Central Andes, to whom we
owe unusually full records of native behavior prior to the sixteenth century. Above all in the things made by Mexican Indian
craftsmen, the abrupt replacement of one visual language by
another is clearly apparent within the span of a single generation
pivoting upon the mid-sixteenth century.
In highland Mexico the peninsular Plateresque art of I525-50
displaced Aztec art. The systematic ages of these two visual languages particularly interest us here. Spanish Plateresque forms at
that time were in the later half of their history. The early expressive character of abrupt, uncontrolled dissonant energy that
marks early Plateresque work had yielded in Spain by I540 to
the later more modulated manner of proportioned harmonies
technical equipment and antecedent tradition were enormous.
The value of the situation to us is that we know beyond doubt

of the incompleteness of the native series: it was cut off before
its time. It never had any later occasion to find its natural con-
clusion. It gives us the clinical example of the incomplete cultural series cut off by an extended one.
The displacement of new classes by older ones shows the discontinuity, as when native laborers had to regress in European
terms, in order to master the new techniques. Thus they learned

vaulted construction by building the simpler types like early
twelfth-century vaults in Anglo-Norman territory, or they
carved simple chip-cut ornament with steel instead of stone tools,
before trying more intricate modeling. The process is one of
recapitulation. New participants review the whole class in a contracted form in order to learn the present position. It recurs in
every school and academy, when the previous habits of the pupil
are displaced by new routine, taught in sequence from fundamental motions to final operations. Every mon1ent of routine
learning contains this discontinuity between the two types of

past knowledge, that of the learner who does it the first time and
that of the teacher, who does it for the t1th time.
'"[
!
Il2
SOMB KINDS OF DtlUA'l'IOl-f
~~.r:f!E SHAPES OF TIME
II3
'
In this sense of routine learning, all pre-professional education
and all colonial situations belong to the replica-mass (p. 39)
rather than to the form-class where innovations are discovered
and explored. Hence colonial societies more often than not resemble learners with inadequate prior training to whom the new
experiences are insurmountably difficult, and who fall back upon
a convenient minimum of working knowledge. In this way a
characteristic arrested development of active form-classes tnay
IJllen those European traditions of design and representation from

'which the colmues never lapsed even when revolting from
,Spanish political rule.
The consequences of the colonial extensions from the mother
com1try are easily charted in Latin America. Equipping a contiuent with cities, churches, houses, furnishings, and tools required
.a gigantic outlay of energy at mhUmum standards of performance. The native labor learned a behavior at the outset which has
occur in provincial and rustic environments. Rustic arts are the been perpetuated ever since by small human numbers, by the
principal elements of colonial artistic life. Usually one stage of unfavorable dispersal of habitable zones, by the immense disdevelopment is separated from a metropolitan series in a remote tances between towns, and by the intperfect cmnmunications
and isolated setting, where the original impulse is repeated agah1 ;UllOng colonies as well as between the colonies and the Peninsula.
and again with dimhushed content but enriched accessories. ExThe sluggish and careless pace of colonial events was overcome
amples are the peasant costumes of nineteenth-century Europe, only thrice in three centuries, and ouly in architecture: the buildwhere arrested moments of ancien rfgime court fashions, some of ings of Cuzco and Lima from r6so to 17xo; Mexican viceregal
them several hundred years old, flourished by repetition in the architecture from 1730 to 1790; and the Brazilian Third Order
chapels of Minas Gerais from 1760 to 1820. Of course Latin
rural countryside.
It is not easy to define a colonial society to everyone's satis- America has towns and villages of extraordinary beauty, like
faction. In tills context, however, it can be regarded as a society Antigua in Guatemala, Taxco in Mexico, and Arequipa in Peru,
in which no major discoveries or inventions occur, where the but their charm, favored by climate and setting, rests upon the
principal initiative comes from outside rather than from within relaxation of more rigorous standards of invention rather than
the society, until it either secedes from the parent-state or revolts. upon the eager quest for excellent newness that made Florence
Many politically independent self-govenring societies neverthe- or Paris for so long the centers of many epochal changes in the
less remain colonial in our sense for long periods after inde- history of things. In Antigua or Arequipa or Ouro Preto, as at
pendence because of continuing economic limitations that re- picturesque towns in every province of Europe, beauty was at-
strict inventive freedom. Thus colonial states created by conquest

all display incomplete series in differing grades of dilapidation.
Extended series. Colonial states also display many kinds of extended series which manifest the dependence of the colony upon
the mother state. These extensions are staffed by persons schooled
in the parent country: the classic example as usual is colonial
Latin America, where the higher officials of the regional govern-
tained by concord in simple old themes, much reduced from the

difficult pieces de maftrise we see in great cities and at court. It is
the beauty of oft-repeated traditional forms favored by nature,

rather than the beauty of things separated from the immediate
past in their makers' intense quest for new forms.
The effects of such provincial extensions upon the arts of the

mother country are on the whole beneficial. The opportmuties
ments all were of Peninsular origin, with native-born officials available to builders, painters, and sculptors iu the Spanish
(Creoles) admitted only to the lower ranks. Peninsular architects, colonies swelled their numbers at home. Thus Seville, the chief
sculptors, and painters very early implanted among native crafts- port for transatlantic shipping, was the focus of the golden age
II4
SOME KINDS OF DURAT(Q)l \rUE SHAPES OP 'fiMB
115
of Spanish painting in the seventeenth century rather than the

new court at Madrid, which often drew its great talents from the
Sevillian school. The florescence of Greek architecture in the
fifth century likewise depended in part upon favoring conditions
Wandering series. Not to be confused with such replicating

\ettensions to province and colony are certain classes whose coni1;nued development appears to require periodic changes of set
:1ing. The examples of this phenomenon of wandering series are
in the prior expansion of the Greek cities to colonies in the 'best taken among extremely large classes, such as Romanesquewestern Mediterranean. Roman imperial architecture beneftted Gothic-late medieval architecture, or Rcnaissance-Matmeristfrom colonial extensions in much the same way, with a gigantic Daroque painting in Europe, where remarkably uniform dis-
increase in tl1e overall building requirements of the state, which

stimulated a swift increase both in nnmbers and in professional
quality among homeland architectural designers. These correlations cannot possibly be proven directly, and they are suggested
only by parallels with more recent situations. If tenable, they
placements of the focus of invention may be noted. They occur

at intervals of about ninety years when the entire geographical
grouping of the centers of innovation shifts to different bases.
Symptomatic of the principal changes are the well-known displacements from abbey to cathedral to city in medieval archibelong among the few occasions when the economic situation lecture. Another is the migration of the centers for important
and the artistic activities of a people appear to interlock closely. painters from the little Central Italian city-states to the sixteenthEconomic historians have discussed the idea that artistic flo- century courts and the prosperous seventeenth-century comrcscence and economic disorders are correlated.1 2 Certainly the ltlCtcial centers.
Spanish example of a relationship between artistic excellence and One explanation, which reduces art to a phase of economic
the abundance of artistic opportunities also fits such a correlation history, is that artists follow the true centers of power and wealth.
between artistic excellence and material crisis, for the seventeenth It is an incomplete explanation, for there are many centers of
century in Spain was an epoch of staggering economic difficulties wealth and power, but there are few centers of major artistic inabove which painting, poetry, and the theater flowered imperish- novation. Artists often gravitate to the lesser centers of wealth
ably. But to keep aesthetic events together in the same per- and power, like Toledo, Bologna, and Niirnberg.
spective, let us say that colonial or provincial stagnation is the Despite the inventor's solitary appearance he needs company;
reciprocal of metropolitan vivacity, that one is secured at the he requires the stimulus of other minds engaged upon the same
cost of the other in the same regional entity. Then every focus questions. Certain cities early accepted the presence of artists'
or center of invention requires a broad provincial base both to guilds, thereby establishing precedent and ambience for their
support and to consume the productions of the center. Hence

for every extended class, like thirteenth-century Gothic art in
the Mediterranean basin, we find a world of replicas in Naples
or on Cyprus, differing mainly in regional accent, but all point-
continuing presence. In other cities a puritanical or iconoclastic
tradition long proscribed the arts of tl1e age as useless or frivolous.

Certain cities show the touch of important artists at every turn:
Toledo and Amsterdam still bear signs of the presence of their
ing to a common center of origin as copies of inventions made great seventeenth-century painters; Bruges molded and was
among the new cities of northwestern Europe, in southern molded by many generations of painters; and the greatest archi-
England, or northern France less than a century earlier.
lects have shaped the urban presence of Florence and Rome. The
u. Robert Lopez, "Hard Times and the Investment in Culture," The Renais- artist requires more than patronage; he also needs association
sance: A Symposium (New York, 1952).
with the work of others both dead and alive engaged on the same
r
II6
SOME KINDS OF DURA'l'IO:N
I
ftlJB
!
SHAPES OF TIME
!17
problems. Guilds, c6teries, bottcghe, and ateliers are an essential i~ewer linked forms to occupy the same positions. Arotmd every
social dimension of the endless phenomenon of artistic renewal lsuccessful form, furthennore, there arises a protective system of
and they cluster by preference in permissive environment~ !sorts for its maintenance and perpetuation, so that the opporhaving both craft traditions and proximity to power or wealth. i cunities for replacement by new design are further reduced in
Hence the slow movement of the centers of ilmovation from one :places where older things ftll the same need. A living artist often
region to another cannot adequately be explained by ecmtotnic may encounter harder competition from the work of artists dead
attraction alone, and it is justified to search for other motives.
'for ftfty years than from his own contemporaries.
Possibly more important than wealth in accotmting for wanA region with many unfulfliled needs, having the wealth to
dering series is the question of saturation. An old solution often satisfy the1n, will under certain conditions attract itmovations.
satisfies its need better than a more recent one. As noted earlier, Chicago after r876 was doubly attractive to architects, both as
each class of form~ both shapes and satisfies a need which con- the established n1etropolitan center of a new economic region,
tinues throughout several stages of cltange in the forms. The need and as a city which the great fire had left in ashes. The florescence
changes less than the different solutions devised for it. The history of the "Chicago School," with men like Burnham, Sullivan, and
of furniture has many examples of this relationship between fixed Wright later on, was the consequence. But the rebuilding of
need and varying solution. Today many furniture forms of Chicago after 1876 would have been a provincial extension rather
eighteenth- and nineteenth-century date still fulftll perfectly the than an epochal renewal of American architecture, without a
need for which they were designed, and often far better than the favorable juncture in the history of forms. Such a favorable
machine-made chairs and tables of modern design. When the jtmcture consists in general of unused technical and expressive
industrial designer discovers a new shape to satisfy an old need, opportunities allowing the institution of new form-classes across
his difficulty is to find enough buyers for the new shape among a broad band of needs. With the ageing of the entire spectrum
people who already own satisfactory old forms. Thus every suc- of form-classes, as in the later centuries of each great epoch of
cessful manufacture tends to saturate the region in which it is
'I
I
made, by using all the occasions that might require the thing.
To take another example: after II40 for about a century the
use of columnar statues of Biblical figures in the embrasures
flanking church doorways became common, as the royal portal
formula which eventually radiated from the Ue de France around
Paris throughout Europe. In France north of the Loire, the principal stages of its elaboration can still be retraced in the great cathedral doorways. In that region, however, the success of the
royal portal formula prevented the successful emergence of any
other solution. Instead, the theme of grouped statues in the embrasures became more and more stereotyped with the spread of
French Gothic art. In other words, every durable and successful
form saturates the region of its origin, making it impossible for
civilization, most urban environments have been saturated 1nany

times over by the work of earlier stages. Thus one dominant
characteristic of such late periods near the end of each major
division ofhistory, like the eighteenth century in western Europe,
is the ascendancy of fashionable manners of decoration such as
Rococo, which are adapted to the superficial cosmetic remodel-
ing of still serviceable older structures inside and out.

Simultalleous series. The wide range of systematic ages an1ong
different classes at the san1e n1oment always makes our own
present seem like a complicated and confusing n1osaic, which
resolves into clear, simple shapes only long after it has receded
into the historical past. Our ideas about Middle Minoan time are
clearer than our ideas about Europe between the World Wars,
partly because less is known, partly because the ancient world
II8
:;
.I
SOME KINDS OF DlJRA'riQ~
~ !llB SHAPES Of TIME
II9
was less complex, and partly because old history comes into long ; figure prevented him from describing gestures or expressions,
perspective more easily than the close view of recent happening. : and it drew the eye to and beyond the f1gural contour rather than
The older the events are, the more are we likely to disregard . into the content of any linear enclosure.
About 520-500 n.c., black-figure style had reached that stage
differences of systematic age. The Parthenon is a retarded example
of the peripteral temple. This class was already very old when in the exploration of these graphic possibilities which we have
lctinos was born. The fact of systenutic age, however, is rarely here designated as "late." At the same time a radical teclmical
if ever mentioned in classical studies. Classical scholars have to change was introduced that allowed a new form-class to take
rely upon approximate dates for large groups of things, and with- shape. The relation of figure and gronnd was inverted by the
in series of things they rarely can fix dates exact to the year. The simple device of letting the linear enclosures have the color of
idea is mor< developed in studies of Gothic medieval sculpture, the pottery gronnd, while the surronnding areas were painted
as when E. Panofsky sought to distinguish the hands of old and black. This new red-figure style allowed painters to describe
yotmg masters at Reims Cathedral in the same decade of the gesture and expression by more copious linear means than before,
thirteenth century. In the cmmoisseurship of Renaissance paint.. bnt it destroyed the old harmonious relation of figure and
ing, apparent inconsistencies of dating and authorship have often gronnd, conferring upon the ftgure a prepotence that robbed the
been resolved with an implicit invocation of systematic age, by gronnd of its former decorative significance. The hmovation persaying that the master persisted in the use of an old-fashioned mitted the opening of a new series: early and late examples of
idiom long after his contemporaries had abandoned it. In studies red-figure style follow in order after the disappearance of late
of contemporary art, finally, there are no problems of dating, black-figure technique.
About eighty or ninety Athenian vessels are preserved on
but the need to sort the schools, traditions, and innovations im~
plies the idea of systematic age.
which scenes are painted in both styles, and some of these repreDifferent configurations vary this fundamental structure of the sent the same scene, such as Heracles and the bull, by the Andopresent without ever obscuring it completely. One of the uses kides painter, on opposite sides of the same vessel, ill black-figure
of history is that the past contains much clearer lessons than the and red-figure techniques, as if to contrast the possibilities of the
old and the new styles. These dimorphic vases (or "bilingual"
present. Often the present situation is merely a con1plicated instance of conditions for which an ideally clear example can be vases as Beazley called them) are unique a11cicnt documents of
fow1d in the retnote past.
the coexistence of different formal systems for an individual
Athenian vase painting in the closing decades of the sixth cen- painter. They show with great clarity the nature of the artistic
decision at any and every moment of history, in the perpetual
tury n.c. offers a lucid instance of simultaneous form-classes at
a small scale and under completely intelligible conditions. The crisis between custom and innovation, between exhausted formublack-figured style of bodies silhouetted like black paper cutouts la and fresh novelty, between two overlapping classes of forms.
A gronp of students upon my suggestion once arranged ranupon light grounds had prevailed for some generations, and it
dom selections of black-figure and red-6gure vase paintings in
a11owed an entire series of advances in the technique of repreorder by early and late formal traits. They worked separately
sentation, favoring always the decorative il1tegration of ftgure
and ground by harmonious and interesting void shapes. But this under instruction to disregard all other traits snch as technique
manner limited the painter's expressive resources. The solid dark and vessel-shape. The outcome of each list was a pairh1g of early
~
'I
120
SOME KINDS OF DUR.A'riO:tf
black-figure with early red-figure, and late with late, associatin

things of similar systematic age rather than by absolute chrono:
logical date.
Another kind of simultaneity comes from a tomb of the third
century A.D. at Kaminaljuyu in the Guatemalan highland. In this
tomh A. V. Kidder found many stuccoed tripod vessels, all of the
sante shape, and of a variant most fatniliar fr01n Teotihuadn in
the Valley of Mexico, a thousand miles away. These vessels were
painted in clear unfired earth colors of two styles: early classic
Maya and Teotihuadn IT, which differ from one another about
as widely as Byzantine and Irish manuscript illumination in the
11inth century. Kaminaljuyu was a colonial outpost of Teotihuacin on the edge of!owland Maya civilization. Maya pottery
traditions were less advanced than in Mexico at that tnoment,
although the Maya were in other respects the possessors of a

learning much more complicated than that of their Mexican contemporaries, especially in written astrological and astronomical
lore. The potters of Kaminaljuyu, who probably furnished
painted vessels in commercial quantities to Maya buyers, simultaneously worked in both styles, tbat of their homeland and that
of their clients.
In brief, the more completely we know the chronology of
events, the more obvious it becomes that simultaneous events
have different systematic ages. A corollary is that the present

always contains several tendencies competing everywhere for
each valuable objective. The present was never uniformly textured, however much its archaeological record may appear to
have been homogeneous. This sense of the present which we
live each day, as a conflict between the representatives of ideas
having different systematic ages and all competing for possession
of the future, can be grafted upou the most inexpressive archaeological record. Every sherd mutely testifies to the presence of the
same conflicts. Each material remnant is like a reminder of the
lost causes whose only record is the successful outcome among
simultaneous sequences.
! tHE
SHAPES OF TIME
121
The topography of simultaneous classes can be presented in

; tWO groups, guided a11d self-determining. The guided sequences
depend explicitly upon models taken from the past: thus re: vivals, renaissances, and all other model- or traditio11-bound
forms of behavior are guided sequences.
Self-determining sequences are much rarer, and they are harder
ro detect. Early Christian art was a deliberate rejection of pagan
traditions. The survivals of pagan antiquity were either strategic
or unconscious in early Christiauity. The Christian sequence,
however, rapidly became model-bound, as when the close array
of these revivals of Early Christian architectural types fmally
constituted the Early Christian tradition."
These terms-guided and self-determining sequences-are
more than mere synonyms of tradition and revolt. Traditioll and
revolt suggest cyclical sequence: revolt is linked to tradition in
a circular motion, with revolt becoming tradition, which breaks
again into revolting fractions and so on da capo. Our terms were
chosen to avoid this suggestion of necessary cyclical recurrence.
It follows that self-determining motions are necessarily brief,
and that guided motions usually become the substance of history. In general, self-determining classes cease either when they
turn into classes guided by their own past victories, like Early
Christian art or when they lose the fteld of actuality to some
other series. Any present moment therefore consists principally
of guided series, each contested by self-determining motions of
dissidence. These gradually subside as their substance is kneaded
into previous tradition, or as they themselves become new
traditional guides to behavior.
Lenses vs. fibers of duration. Here is a possible resolution to
the question asked by the proponents of Struktuiforschung (p. 27).
We do not need to assume with them tbat the parts of culture
13. R. Krautheimer, "The Carolingian Revival of Early Christian Architecture,''
Art Bulletin, 24 (1942), 1-38. Erwin Panofsky, Renaissance and Renascences (Stockholm, 1960) has charted the later moments in the guided revival sequence which
gradually came to displace the self-determining sequence of Early Christian art.
-,..,.
I
122
SOME KINDS OF DURAT!Ol'{
are all either central or radial. They seem to think of a culture as

if it were a circular lens, varying in thickness according to the
antiquity of the pattern. Instead, we can imagine the flow of time
as assuming the shapes of fibrous bundles (p. 37), with each fiber
corresponding to a need upon a particular theater of action, 'llld
the lengths of the fibers varying as to the duration of each need
and the solution to its problems. The cultural bundles therefore
consist of variegated fibrous lengths of happening, mostly long,
and many brief. They are juxtaposed largely by chance, and rarely
by conscious forethought or rigorous plarming.
Conclusion
The historical study of art on systematic principles is about

two tl10usand years old if we include Vitruvius and Pliny. This
accumulated knowledge now far surpasses the ability of any individual to encompass its detail. It is w1likely that many major
artists remain to be discovered. Each generation of course continues to revaluate those portions of the past which bear npon
present concerns, but the process does not uncover towering new
figures in the familiar categories so much as it reveals twfamiliar
types of artistic effort, each with its own new biographical roster.
The discovery of hitherto unknown painters of the stature of
Rembrandt or Goya is far less likely tl1an our suddenly becoming
aware of the excellence of many craftsmen whose work has only
lately come to be accepted as art. For example, the recent advent
of Western action painting has prompted the revaluation of a
similar tradition in Chinese painting since the ninth century-a
tradition to which the West was insensitive Wltil recent years.
FINITE INVENTION
Radical artistic hmovations may perhaps not continue to
appear with the frequency we have come to expect in the past

century. It is possibly true that the potentialities of form and
meaning in human society have all been sketched out at one
,i
time and place or another, in more or less complete projections.

We and our descendants may choose to resume such ancient in-
complete kinds of form whenever we need them.

As it is, our perception of things is a circuit unable to admit a
.. !
'
;
r24
great ~ariety of new sensations
'
..,
. :r
'
CONCL USIO!;
all at once. Human perception is
best smted to slow modiftcations of routine behavior. Hence invention has always had to halt at the gate of perception where
the narrowing of the way allows much less to pass than the im.
portance of the messages or the need of the recipients would
justify. How can we increase the inbound traffic at the gate I
The purist reduction of knowledge. One old answer is to reduce
the magnitude of the inbound messages by amplifying what we
are willing to discard. This answer was again attempted in Europe
and America by the generation between the wars from 1920 to
1940: it required the rejection of history. The hope was to diminish the traffic by restricting it only to pure and simple forms
of experience.
Purists exist by rejecting history and by retuming to the
imagined primary forms of matter, feeling, and thought. They
belong to a 6mily spread throughout history. The Cistercian
architects of the high middle ages are among its members, as
well as the New England craftsmen of the seventeenth century
and the pioneers of functionalism in our century. Among these
last, men like Walter Gropius took on the old burden of their
purist predecessors. They sought to invent everything they
touched all over again in austere forms which seem to owe
nothing to past traditions. This task is always an insurmountable
one, and its realization for all society is frustrated by the nature
of duration in the operation of the rule of series. By rejecting
history, the purist denies the fullness of things. While restricting
the traffic at the gates of perception, he denies the reality of
duration.
Widening the gate. Another strategy has been more normal.
It requires the widening of the gate until more messages can enter.
The gate is bounded by our means of perception, and these, as
we have seen throughout the history of art, can be enlarged repeatedly by the successive modes of sensing which artists devise
for us. Still another strategy is to codify incoming messages so
that redundancies are eliminated and a larger useful volume of
fliNITE INVENTION
125
traffic flows. Whenever we group things by their style or class,

we reduce redundancy, but it is at the cost of expression.
In these ways the history of art is like a vast mining enterprise,
with innumerable shafts, most of them closed down long ago.
Each artist works on in the dark, guided only by the tunnels and
shafts of earlier work, following the vein and hoping for a
bonanza, and fearing that the lode may play out tomorrow. The
scene also is heaped with the tailings of exhausted mines: other
prospectors are sorting thetn to salvage the traces of rare elements
.once thrown away but valued today more than gold. Here and
there new enterprises begin, but the terrain is so varied that old
knowledge has been of litrle use in the extraction of these altogether new earths which may prove worthless.
Upon this scene investigators have proceeded as if the life histories of all the principal workers were not only indispensable
but adequate and sufficient. Biographical summation does not
yield an accurate description of the mother-lode, nor will it
accotmt for the origins and the distribution of this immense re-
source. The biographies of artists tell us only how and why the
vein was exploited in such a peculiar way, but not what the vein
is or how it came to he there.
Perhaps all the fundamental teclmical, formal, and expressive
combinations have already been marked out at one time or
another, permitting a total diagram of the natural resources of
art, like the models called color-solids which show all possible
colors. Some portions of the diagram are more completely
known than others, and some places in it still are sketchy, or they
are known only by deduction. Examples are Rene Huyghe' s
Dialogue avec le visible (1955), an effort to chart the theoretical
limits of painting, and Paul Frankl's System der Kunstwissenschaji
(1938), which seeks to define the boundaries of all art.
The finite world. Were this hypothesis to be verified, it would
radically affect our conception of the history of art. Instead of our
occupying an expanding universe of forms, which is the contemporary artist's happy but premature assumption, we would
rz6
;
CONCLUSION
be seen to inhabit a fmite wodd oflimited possibilities, still largely

unexplored, yet still open to adventure and discovery, like the
polar wastes long before their human settlement.
Should t!Ie ratio between discovered positions a11d undiscovered ones in human alfairs greatly favor the former, tl1en the
relation of the future to the past would alter radically. Instead of
regarding t!Ie past as a microscopic annex to a future of astronomical magnitudes, we would have to envisage a future with
limited room for changes, and these of types to which the past
already yields the key. The history of things would then assume
an importance now assigned only to the strategy of profitable
inventions.
'sQUIVALENCB OF FORM AND EXPRESSION
127
'remarkable. But the grand catalogue of persons and works is

' ear that state of completion that is heralded by diminishing
11
returns. The techniques of exact dating and attribution, which
.,1e methods of measuring the extent and authenticity of bodies
of work, change little from generation to generation. Only the
historical study of meaning (iconology) and the conceptions of
lllorphology are new in our century.
Iconological diminutions. The learned and subtle iconologists
who trace out the convolutions of a humanistic theme through
thousands of years have charmed all wit!I the discovery that each
period endowed the theme with its characteristic enrichments,
reductions, or transfonnations. As such studies accumulate they
are like the chapters of a book by many authors, each treating

THE EQUIVALENCE OF FoRM AND EXPRESSION
Wheu we scan things for traces of the shape of the past,

everything about them deserves our attention. Yet this conclusion, which is self-evident once we recognize tl10t things alone
allow us to !mow the past, is generally ignored under the demands
of specialized stndy. Archaeological studies and the history of
science are concerned with things only as technical products,
while art history has been reduced to a discussion of t!Ie meanings

of things without much attention to their teclmical and formal
,,
011 element of humanistic tradition, all treating the survival of

antiquity. Continuity rather than rupture is the criterion of value
among students of meaning.
In iconology the word takes precedence over the image. The
image nllexplained by any text is more intractable to the iconologist than texts are without images. Iconology today resembles an
index of literary themes arranged by titles of pictures. It is not
usually realized t!Iat iconological analysis depends upon discovering and expanding chosen texts with visual aids. The
method is most successful when orphaned images can be reunited
organization. The task of the present generation is to construct a
with t!Ie texts to which they were originally more or less direct
history of things that will do justice both to meaning and being,

both to the plan and to the fullness of existence, both to t!Ie
scheme and to the thing. This purpose raises the familiar existential dilemma between meaning and being. We are discovering
little by little all over again that what a thing means is not more
important than what it is; that expression and form are equivalent
challenges to the historian; and t!Iat to neglect either meaning or
being, either essence or existence, deforms our con1prehension
of both.
As we examine t!Ie procedures used in the study of meaning
illustrations. In the absence oflitcrature, as an1ong the pre-literate

Mochica or Nazca peoples of ancient America, there is nothing
and in the study of form, their increasing precision and scope are
in writing with which to enlarge our knowledge of the images.

We are obliged to overlook conventional meaning. There is no
coeval verbal plane to which the image can be reduced or transformed.
When he has texts, however, the icouologist strips the fullness
of tl1ings down to those schemes that the textual apparatus will
allow. Frequent repetition ancl variation measure the importance
of a theme, especially if it has survived the barrier of the Middle

Ages. The examples and illustrations spill forth into a few verbal
I
I28
CONCLUSION i .eQUIVALENCE OF FORM AND EXPRESSION
129
molds suggested by the texts, and their substance is correspond- crises of accent and stress. Thus the seventeenth-century archiingly diminished lllltil only meanings survive from the plenitude ; tects align either with a planiforn1 tradition or a curviplanar one,
of things themselves.
and it is confusing to call them both Baroque.
On the other hand, studies of morphology, based upon the
Indeed names of styles entered general use only after being intypes of formal organization and their perception, now are u11.. flated by misuse and incomprehension. In 1908 Otto Schubert
fashionable and they are dismissed as mere formalism by the had already extended the Italian term to Spanish forms, when he
diligent detectives of texts and meanings. Yet the same sorts of wrote Geschichte des Barock in Spanien. Here the "historic styles"
schematic deformation limit both iconographers and morpholo- now compel more belief than the things themselves. "Spanish
gists. If the former reduces things to skeletal meanings, the latter Baroque" has becmne a ntore compelling verbalistn than the
submerges them in strea1ns of abstract tenns and conceptions easily verified reality that Italian and Spanish architecture I6oowhich mean less and less the more they are used.
I700 had very few persons or traits in cotrunon. The direct
The deficiencies of style. To choose one example among many, study of Spanish art impedes this sort of generalization: for inthe phrase "Baroque style," arising from the study of seventeenth- stance, it is difficult to sec a close relation between the archicentury R01nan works of art, now extends in general use to all tectnres of Valencia and Santiago de Compostela at any time
European art~ literature, and music between r6oo and r8oo. The
term itself is in no way descriptive of form or period: it was

originally baroco, a thirteenth-century nrnemonic term describing the fourth mode of the second syllogism coined for the use
of students of logic by Petrus Hispanus, who later became Pope
John XXI.! In effect, to speak of Baroque art keeps us from
noting either the divergent examples or the rival systems of
fonnal order in the seventeenth century. We have become re~
luctant to consider the alternatives to Baroque art in most
regions, or to treat the many gradations between metropolitan
and provincial expressions of the same forms. Nor do we like

to think of several coeval styles at the same place. Actually, the
Rhone Valley, Galicia was related to Portugal and the Low

Countries.
The plural present. Everything varies both with time and by

place, and we cannot fix anywhere upon an invariant quality
such as the idea of style supposes, even when we separate things

from their settings. But when duration and setting are retained
in view, we have shifting relations, passing moments, and changing places in historic life. Any imaginary dimensions or continui-
throughout Europe and America is an architecture of curved

planes approaching a system of lllldulant membranes. These
ties like style fade from view as we look for them.

Style is like a rainbow. It is a phenomenon of perception
governed by the coincidence of certain physical conditions. W c
can see it only briefly while we pause between the snn and the
rain, and it vanishes when we go to the place where we thought
mark the changing pressures of inner and outer environments,
we saw it. Whenever we think we can grasp it, as in the work
But elsewhere in Europe, especially in Spain, France, and the
of an individual painter, it dissolves into the farther perspectives

of the work of that painter's predecessors or his followers, and
it multiplies even in the painter's single works, so that any one
picture becomes a profusion oflatent and fossil matter when we
see the work of his youth and his old age, of his teachers and his
Baroque architecture of Rome and its scattered derivatives
northern countries, another mode of composition prevails. It may
be called planiform, or non-lllldulant, having only ascending

I.
between r6oo and r8oo for the two cities were not in touch, and
while Valencia's c01mections were with Naples, Liguria, and the
Petrus Hispanus, Summulae log{cales, ed. I. M. Bochenski (Rome, 1947).
I
IJO
']'::
'
,-.I
CONCLUS!Ot;
pupils. Which is now valid: the isolated work in its total physical
presence, or the chain of works marking the known range of its
positiom Style pertains to the consideration of static groups of
entities. It vanishes once these entities are restored to the flow
of time.
Not biography nor the idea of style nor again the analysis of
meaning confronts the whole issue now raised by the historical
study of things. Our principal objective has been to suggest other
ways of aligning the main events. In place of the idea of style,
which e1nbra.ces too many associations, these pages have outlined
the idea of a linked succession of prime works with replications,
all being distributed in time as recognizably early and late versions of the same kind of action:
Index
Abstract expressionism, zs, 66

Action painting, 123
Actuality, 16-19, 77
Aegina, 98
Aesthetic fatigue, So
Aevum, 84
African tribal sculpture, 35
Age, :5:5; absolute, 57: systematic, 55,
j6, 98, III, 120
Ahuitzol,
I II
Ajanta, 40
Artists, as entertainers, 53; representations of, 49-50; solitary and gregarious, sz; twentieth-century position, 85; typology of lives, 86, 91
Arts, rustic, 112
Ashanti, 94
Astronomers, 19
Athena Parthenos, 42, 43
Atonal music, 66
Australia, 3 5
Avantgardisme, 62
Albetti, Leon Dattista, 88

~
Altamira, 104, 104 n.
Amsterdam, IIS
Andokides painter, II9

Anthropocentdsm, 19
Antigua, IIJ
I;'
Antiquarian, 12
Antiquity, 29, roB
Archaic, 32
Archaic art, I I
Architecture, viii, II, 36-37, 41, 48,
s6-sB, Bo-BI, 90, 102, IoB, u4, n6,
II?, 129-130; steel-frame, 103
Arequipa, II3
Array, dense and sparse, 97
Art, r; and astronomy, 19-20; as a system of formal relations, vii; as symbolic language, vii; history of, r, 61;
identification of, So; meaning of, vii
Art nouveau, n
Articulations of historical substance, 4
Artifacts, 9, 20, 83
Artisans, 10
Artistic invention, 15, 66, 69, 70
Baalbek, 56
Bark-painting, 35
Baroco, 128
Baroque, II, 32, 128-129
Baroque art, 128
Being, 126
Bergson, Henri, 67
Bernini, Giovanni Lorenzo, 52
Binomial nomenclature, 3
Biography, 5, 32, 36, 86, 88-90, 125,
130, 131
Biological time, 13, See also Time
Biology, 8
Bohemianism, 91
Bologna, I I 5
Bonampak, 40
"Bordeaux pilgrim," 42
Borromini, l::rancesco, sz, 90
Boscoreale, 87
Bruges, II5
Drunelleschi, Filippo, 90
Burnham, D. H., II7
Bushman rock. paintings, 47
IJ2
Caravaggio, M. A., 90
Careers, 91
Carolingian illuminations, so
Carracci, the 87
Cassircr, Ernst, vii
Catalogues, 6
Cathedrals, 37
Causality, 20
Cellini, Benvenuto, 86
Cennini, Cennino, 106
Centers of innovation, I 15
Central Italian painting, 105
Century, 99
cezanne, Paul, 51, 87, 91
Change, 6o, 75, 77, Bs; historical, 59;
linguistic, 6o
Chiaroscuro, 49
Chicago, 94, II?
Chronology, 79
Cistercian, 124
Classes, 31-33, 56, 67, 73, 82, 87, 93,
II7, ug; and fashion, 39; arrested,
IOS)-II2; by sequence, 36; continuous, 106, 108; diagnostic difficulties,
41-45; extended, no, II4; inactive,
45; intermittent, ro6; simultaneous,
121. See also Sequences, Series
Classic, 32
Claude Lorrain, 87
Cabo, Bernabe, no
Cohesion, 72
i.'
'I
,,
::::
Colonial society, uz
Colony, 112
Commercialism, 76
Connoisseurship, 44
Contemporary art, II, II8
Continuities, 27, 108
Convention, 67-68
Convergence, 47
Copying, 62, 71, 73-74, 7!1

Craft education, I 5
Craft techniques, 47
Creoles, I 12
Critic, 26
Croce, Benedetto, 31, 67
Crystal Palace (London), 109
133
Curviplana.r,
Cuzco, II3
129
Darwin, Charles, 63
Decade, 100
Dedalus, 53
Demography, 93
Dimorphic, 119
Dipylon vases, 27
Directed graphs, 34 n.
Discard, 62, 77-82
Discontinuities, 27. See also Classes,
Sequences, Series
Dittochaeum, 97
Divisions of the arts, 14-16
Domenichino, 87
Donatello, 90
Drift, 6o, 71, 95; historical, 75
Durations, 54, 75, 79,- 83, 96, 99-106,
us; fibers of, 121
Earth-clocks, 14
Economic, 85
Eight-part vaults, 49
"Eliot effect," 35 n.
Eliot, T. S., 35, 52
Enlightenment, 10
Entrance, 6, 41, !16, 88-90, 104
Envelopes, 99
Eschatology, 107 n.
Escorial, 58
Etruscan tombs, 40
Euhemerist, 24
Faberg6, Peter Carl, 106
Factuality, 46
Faraday, Michael, 9
Fashion, 38-39, 59, 74, IOI, 101 n. See
also Styles
Ferro-concrete, 107
Fifty-two-year period, 100
''Pine art," 14, r6
Flanagan, John, 108
Florence, r I 5
Focillon, Henri, 3, 16, 32, 48, 62
Folk arts, 14, IS
;formalism, viii, 128

'pormalist, 32
:form-class, 33, 88-89,93, 112
: FcmnermUdung, 8r, 81 n.
frankl, Paul, 125
functionalism, 109
functionalists, 16
Gabriel, J.-A., 90
Galicia, 129
Gaudi, Antonio, 107
Gauguin, Paul, 91
Gcmmeau glass, 44
Generation, 100
Genius, 7, 8
Genotypes, 41
Gesta deorum 1 29
Gestalt, 28
Ghiberti, Lorenzo, 91
Glottochronology, 24
Goethe,J. W. von, 95
GOller, Adolf, Bo-81
Gothic art, II
Goya, Francisco, 123
Graph theory, 33 n.
Gropius, Walter, 124
Guarini, Guarino, 90
Guercino (G. F. Barbieri), 52
Guilds, 93, ns
Hagiography, 29
Happening, fast, 95, 96; slow, 96
Herculaneum, 87
Hercules, 29
Heron, 48, 96
Historians, 12, 19
Historical knowledge, 21, 30
Historical time, 13, 19
History, 77; segmentation, 17
Hokousai, 90
Homeric poetry, 27
Homo faber, 10
Humanist, 83
Hutchinson, G. E., viii, 2 n.
Huyghe, Rene, 125
Iconography, 25, 26-27, 28

Iconology, 4 1 26, 127, 128
lctinos, n8
Image, 127
]m(lgines, 98
Immanence, 101 n.
Indiction, 102, 104, 105 n.
Individuality, 36
Ingres, J. A. D., 87
Initial events, 21
Instinct, I 8
Instrumental usc of art, 16
Instruments, tools, 26
"Intentional fallacy," 45
"International style," 109
Invention, 62, 63, 67-68, 71-72, 85;
finite, 123; useful, 65-67, 69, 70
Ion, 12
Jefferson, Thomas, 88
Joyce, james, 52
Kaminaljuyu, 120
Kant, Immanuel, 16
Kaschnitz-Weinberg, Guido von, 27
Kidder, A. V., 120
Klee, Paul, 52
Kroeber, A. L., 2 n., 101, IOI n.
Landa, Diego de, uo
Laon Cathedral, 51
Lascaux, 104, 104 n.
Latin America, I 12
Ledoux, Claude, 109
Leonardo da Vinci, 7, 88
Lexicostatistics, 24, 60 n.
"Life-cycle," 8, 55
Life expectancy, IOI
Liguria, 129
Lima, II3
LinBuistics, vii, 6o-61
Linked solutions, 33, 82, 131
Linnaeus, 9
Lodoli, 16
Low Countries, 129
Luini, Bernardino, 7
Lustrum, 100
135
134
Madrid, II4
Mahon, Denis, 52
Malraux, Andre, 62
Manet, Edoua:rd, 88
Mannerism,
n, 32
Mantes Cathedral, 37
Masaccio, 90
Material culture, 1 , 9
Matz, Fricdtich, 27
Maya sculpture, 56
Meaning, 126, 130; adherent, 28; conventional, 26; intrinsic, 26; natural,
26
Meissen porcelain, 28
Mendel, Gregor, 109
Mendicant friars, 57
"Mental culture," 9
Metals,64
Metamorphoses, 108
biological and physical, 5
Met~phors,
MeXIco, 57
Michelangelo, 88, 89
Millennium, 100
Miller, John Paul, Io6
.:
Minas Gerais, I 13
Minoan, I 17
"Minor arts," 15
Mnesicles, 89
Modern art, 70
Moore, Henry, 108
Morphemes, vii
Morphology, 8, 127
Mosul, 94
Motion, 75
Museums, So
Mutants, 40
Myths, 22
Naples, 129
Nauation,
Nativity, 25
Need, 55, 62, 77
Neo-classic, 32
Neomorphs, 56
Networks, 34 n.
''Noise," 6x
Non..artistic, x6
NUtnberg, II5
Obsolescence, 77
Oil painting, 64
Orbit, 19
Ore, Oystein, 33 n.
Orozco, Jose Clemente, w8
Ottro Preto, Ill
Ovid, 108
Palenqnc, 51
Panofsky, Erwin, 118
Parthenon, 41, 43, 98
lla:x:ton, Sir Joseph, 109
Perception, 124
Pcriodology, xo6
Periods, 99
Petrus Hispanus, 128
Phenotypes, 41
Phidias, 42, 89
Philostratus, 98
Phonemes, vii
Picasso, Pablo, 52, 56, 90, 91
Picturesque, .98
Fiero di Cosima, 90
Planiform, 129
Plateresque, 58, no
Plato, u
Pliny, 123
Pol de Limbourg, 66
Portugal, 129
Positional value, 9?
Poussin, Nicolas, sr, 52 , 8?
Prehistoric tradition, 47
Prime numbers, 39
Prime objects, 39--45, 70, ss, 131
"Primitive," 35
Primitivism, II
Problem, 55
Promorphs, 56
Provenr;:al poetry, S?
Provincialism, ?6
Prudentius, 97
Purist reduc~ion, l2S
Purpose, 8
:p,adical invention, 69, 70

jl.aphacl, 7, 87
Reformation, I I
jl.eims Cathedral, II8
Relays, 21-24
Rembrandt van Rijn, 90, 123
Renaissance, xo, 74
Renaissance painting, uS
Renoir, Auguste, 56
Renovation, 89
Repetition, 62, 63
Replica~mass, 39, 42, II2
Replicas, 2, 39, 42-44, 71-72, 76, IJI
Replication, 39, 71-76, 82
Retention, 77-82
Revival, 74
Reynolds, Sir Joshua, 90
Rh6ne Valley, 129
Rib vaults, 57
Richardson, H. H., 90
Riegl, A., 31
Ritual, 77
Rivera, Diego, roB
Rococo, u, 32, II?
Roger van der. Weydcn, 50
Romano-Campanian painting, 51
Romano, Giulio, 7
Rome, IIS
Routine, 72-75
Routine learning, III
Royal portal, rx6
Rubens, Peter Paul, 52
Rtu1s, 86
Sahagtln, Bernardino de, r 10
St. Luke, 25
St. Thomas Aquinas, 84
Santi~go de Compostela, 129
Saturation, 116
Schapiro, Meyer, 4 n.
Schlosser, Julius von, II n.
Schubert, Otto, 129
Science, 61; history of, 4
Segovia, 58
Sequences, viii, 2, 64; and fashion, 39;

and technical innovation, 48; arrested, 35: as open classes, 53; formal,
33, 37-38, 42, 44, 50, 55, 85, 87;
guided, I2Ij open, 45; open and
closed, 35-38; open-ended, 34; retarded, 48; self-determining, 121;
simultaneous, 120; stunted, 49. See
also Classes, Series
Serial position, 54-61, 64
Seriation, 79, 85
Series, viii, 2, 6, 33, 45, 54, 64, II9i
appreciation of, 45-47; as closed
classes, 53; closed, 106; extended,
II2-II4i open, 45; rule of, 54-57,
75, 78, 125; simultaneous, II?-I2Ij
wandering, IIS-117. See also Classes,
Sequences
Sevilla, II3
Signals, 17, 20-21, 62; adherent, 25;
primary, 23; redundancy of, 23;
secondary, 23; self-, 24, 25
Simultaneity, 56, 120
Siqueiros, David Alfaro, roB
Socrates, r 2
Speciation, 34
Stagnation, II4
Stairs, 98
Stars, 19
Strangford shield, 43
Stmkturanalyse, 27
Struktuiforschung, 27 n., 121
Style, 3, 32, 129; "Franr;:ois 1er," 4;
"Gothic Mannerist," 4; "Hellenistic
Baroque," 4; "Louis XVI," 4;
Middle Minoan, 3
Styles, 3-5; deficiencies, 129-131; historic, 3
Sullivan, Louis, 90, I I7
Symbol, vii, 74
Talent, 7
Taste, Sx
Ta:x:co, IIJ
Tedium, So, 94
Temperament, 7, 92
Teotihuadn, 120
Text editing, 46
Time, 19-20, 32, 84, 89, 99-101, I I7II8
Toledo, IIS
Toltec-Maya, ro8
Tomb furnishings, '77
Topology, 34
Trait, 36
Tree-rings, 14
Triviality, 46
Twain, Mark, 64
Typology, 8. See also Artists
Uccello, Paolo, 51, 90
Uxmal, sr
Valencia, 129
Van Gogh, Vincent, 90
Variation, 63
Vasari, Giorgio, sr
Vase-painting, Greek, 35, 102, u8;
black-figured, 48; red-figured, 48
.. i'
\.-.
Velazquez, Diego, 88
Vignola, 58
Villani, Filippo, 5
Villard, de Honnecom;t, 5 I
Vis a tetgo,
so
Visigothic sculpture, 56
Vitruvius, 123
Vivacity, II4
Wallace, Alfi:ed, 63
Webb, John, 98
West, Rebecca, 45
Wickhoff, F., 31
Winckelmann, J. J., r r
WOlftlin, Heinrich, 31
Words, 59
Work of art, Bo
Wright, Frank Lloyd, 90, ro8, II7
Writing, 2.4
Zevi, Bruno, 5, 5 n,
Zurbal"<ln, Francisco, 97

George Kubler - The Shape of Time

Uploaded by

Copyright:

Available Formats

George Kubler - The Shape of Time

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

George Kubler - The Shape of Time

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What are some of the main topics discussed in the book?

What are some of the main topics discussed in the book?

F

THE SHAPE OF TIME

NEW HAVEN AND LONDON: YALE UNIVERSITY PRESS

The title-page and chapter-head illustrations

vicinity. (Courtesy American Museum of Natural

Originally published with assistance from the foundation

Copyright 1962 by Yale University

printed in the United States of America by

(except by reviewers for the public press),

''"''J ~~ULU:GE LIBHAR't

Cassircr' s partial defmition of art as symbolic language has

the rest of history.

ments undergo more or less regular evolutions in time without

The main framework of these ideas was set down at Gaylord

SYMBOL, FORM, AND DURATION

and to Professor James Ackerman of Harvard, whom I taught

Symbol, Fvrm, and Duration

The Limitations of Biography: Individual emrances. Talent

The Historian's Commitment: The divisions of the arts

Formal Sequences: Linked solutions. Open and dosed sequences. Fashions

ties. Serial appreciation. Technical renewals. The invisi-

ble chain. Solitary and gregarious artists

Invention and Variation: Artistic invention. Convention and

Replication: Permanence and change. The anatomy of routine. Historical drift

Discard and Retention: Obsolescence and ritual. Aesthetic

Fast and Slow Happening: The typology of artists' lives.

The Shapes of Time: Positional values. Periods and their

The History of Things

minutions. The deficiencies of style. The plural present

THE HISTORY OF THINGS

industry. The family of things begins to look like a smaller family

THE HISTORY OF THINGS

ologists and anthropologists classify things by their nses, having

THE HISTORY OF THINGS

But the whole arrangement is unstable: the key word has

as soon as we admit the idea that this substance possesses a

The lives of the artists have been a genre in the literature of

THE HISTORY OF THINGS i

"Good" or "bad" entrances arc more than matters of position

traditions cannot be treated properly in biographical segments.

in the sequence. They also depend upon the union of tempera-

country in terms of the experiences of a single traveler on several

mental endowments with specific positions. Every position is

person at a different time. Thus every birth can be imagined as

elements of continuity, and not the travelers or the functionaries

artists are not so n1uch those of talent as of entrance and position

The analogy of the track yields a useful formulation in the dis-

in sequence. Talent is a predisposition: a talented pupil begins

cussion of artists. Each man's lifework is also a work in a series

younger; he masters the tradition more quick1y; his inventions

extending beyond him in eitl1er or both directions, depending

come more fluently than those of his untalrnted fellows. But

to fmd a better entrance. Without a good entrance, he is in danger

less than its presence.

of wasting his time as a copyist regardless of temperament and