09 2953 Loehr chapter 8 with illustrations[1].pdf Max Loehr AND THE Study OF Ch i n e S e BronzeS Style and Classification in the History of Art RobeRt bagley volUME 141 TO PLACE YOUR ORDER NORTH AMERICA & CANADA REST OF WORLD by phone 1-800-666-2211 (USA/Canada) by phone 1-808-956-8255 or fax 1-800-688-2877 (USA/Canada) or fax 1-808-988-6052 or email orderbook@cupserv.org or email uhpbooks@hawaii.edu or mail Cornell University Press Svcs or mail University of Hawaii Press 750 Cascadilla Street 2840 Kolowalu Street Ithaca, New York 14851 Honolulu, Hawaii 96822 or online www.einaudi.cornell.edu/eastasia/publications or online www.uhpress.hawaii.edu 140 Uris Hall, Ithaca, New York 14853, USA tel 607 255 507 ~ fax 607 255 1388 email ceas@cornell.edu www. einaudi.cornell.edu/eastasia/publications Max loehr (1903–1988), the most distinguished historian of Chinese art of his generation, is celebrated above all for a 1953 art historical study of Chinese bronzes that effectively predicted discoveries Chinese archaeologists were about to make. those discoveries overthrew the theories of loehr’s great rival bernhard Karlgren (1889–1978), a Swedish sinologue whose apparently scientific use of classification and statistics had long dominated Western studies of the bronzes. Revisiting a controversy that was ended by archaeology before the issues at stake were fully understood, Robert bagley shows its methodological implications to be profound. Starting with a close reading of the work of Karlgren, he uses an analogy with biological taxonomy to clarify questions of method and to distinguish between science and the appearance of science. then, turning to loehr, he provides the rationale for an art history that is concerned above all with constructing a meaningful history of creative events, one that sees the intentionality of designers and patrons as the driving force behind stylistic change. In a concluding chapter he analyzes the concept of style, arguing that many classic confusions in art historical theorizing arise from a failure to recognize that style is not a property of objects. addressed not just to historians of Chinese art, this book uses loehr’s work on bronzes as a case study for exploring central issues of art history. It will be of interest to anyone concerned with the analysis of visual materials. MaX loeHR aND tHe StUDy oF CHINeSe bRoNZeS Style and Classification in the History of Art RobeRt bagley CeaS Vol 141 2008 200 pp ~ ISbN hardcover 978-1-933947-11-2 hc $85 ~ ISbN paperback 978-1-933947-41-9 pb $55 CoNteNtS 8 Preface 10 1. introduCtion: CLaSSifiCation and StyLe Karlgren and loehr taxonomy Classifying bronzes 17 2. KarLgren’S firSt PaPer: “yin and Chou in ChineSe BronzeS” (1936) 22 3. KarLgren’S SeCond PaPer: “new StudieS on ChineSe BronzeS” (1937) Sorting Vessels by Motif Interpreting the a and b groups Choosing Characters 32 4. KarLgren’S Later writingS “Some early Chinese bronze Masters” (1944) “once again the a and b Styles in yin ornamentation” (1946) “bronzes in the Hellström Collection” (1948) “Notes on the grammar of early bronze Décor” (1951) “Marginalia on Some bronze albums” (1959) “Marginalia on Some bronze albums II” (1960) “Some Characteristics of the yin art” (1962) 49 5. Loehr’S firSt PaPer: “BeitrÄge zur ChronoLogie der ÄLteren ChineSiSChen Bronzen” (1936) 64 6. Loehr'S “the Bronze StyLeS of the anyang Period” (1953) loehr’s Paper archaeology and loehr’s bronze Styles “another way of overcoming certain limitations of Style III” the Invention of animal Images Imaginary animals the Meaning of the bronze ornament 98 7. CLaSSifiCation a Warehouse Full of bronzes Science More Classifiers: Thorp, Zhang, Elisseeff, von Erdberg “Casting bronze the Complicated Way” 121 8. StyLe 130 list of tables and Figures 137 a Note on Dates 138 tables and Figures 185 Works Cited 194 Index Such expressions as that famous one of linnaeus, and which we often meet with in a more or less concealed form, that the characters do not make the genus, but that the genus gives the characters, seem to imply that something more is included in our classification, than mere resemblance. —Charles Darwin, On the Origin of Species (1859), chapter 13 the logicality of innovations . . . can be understood, I believe, as a matter-of-course symptom of the essentially rational natura of art. Innovations do not automatically come into being. they are hard-won achievements in which the individual position of an artist in his rational, critical discourse with tradition or previous achievements asserts itself —Max loehr, “Some Fundamental Issues in the History of Chinese Painting” (1964) For how could one give an idea of the pace of artistic life in periods of great inventiveness, if not by placing the stress firmly on the dynamics of change and on the unpredictability of stylistic invention? —Jean bony, French Gothic Architecture of the 12th and 13th Centuries (1983) [121] 8 style In a critique published in 1981, Audrey Spiro sought to identify some of the assumptions behind Loehr’s 1953 paper. The assumptions she singled out for attention all involve the concept of style—“a concept largely undefined, but usu- ally referring to the changing forms of different types of vessels and to their orna- mentation.”402 In trying to decide what style meant to Loehr she examined not only his work but also, expecting them to share the same assumptions,403 that of his teacher Ludwig Bachhofer and his student Virginia Kane: The first [assumption] is that at any one time and place, objects of any one class have in common certain properties that, in sum, may be termed, their style. None of the three art historians, in fact, specifies these prop- erties entirely, although we may make some reasonable inferences on the basis of their articles adumbrated in this paper. For both Loehr and Bachhofer, although they write of style as a separate property, analyses of this property for any one vessel include the shape of the vessel, its qualities of production—e.g., thin-walled, thick-walled; the decor; the technique for producing the decor—incised, modeled, etc.; and the types of forms used in that decor—geometric, “naturalistic,” etc. There are others, of course, and these are but some of the obvious ones. For Kane [who studied the calligraphy of bronze inscriptions], the prop- erty of greatest importance appears to be the shape of the character, whether lines are straight or curved, whether they bisect or touch, etc. Style appears, in fact, to be an ensemble of characteristics which are described, as above, and about which judgments are made.404 It does not speak well for the state of our discipline that Spiro had to guess what style meant to her three authors (nor does the fact that all three of them thought differently about it). Art history has still reached no consensus about the meaning of its most basic concept, in part at least because the concept is deeply entangled with unresolved problems in historiography.405 I would like to sug- gest that Loehr’s idea of style (but not Bachhofer’s) was a simple one, and that style meant to him what it means in practice, if not in theory, to most art histo- rians today. In the passage just quoted Spiro suggests that when Loehr and the others speak of an object’s style they are referring to a certain ensemble of its physical proper- ties. What puzzles her is how they selected those properties. Why, for instance, is incised/modelled décor taken to be a part of the object’s style while square /round leiwen is not? It is important to her to know how they arrived at their selection because of two assumptions she believes them to make about style. The first is that “styles evolve from previous styles, and that this evolution takes place almost inevitably regardless of conditions external to the manufacture of the vessels.” The second is that “at any one time and place, the entire population of vessels of any one type has one style only.”406 The second assumption is certainly false, and Spiro is right to reject it, but nei- ther Loehr nor Bachhofer actually made it, though Bachhofer came close.407 As for the first, in the absence of a definition of style Spiro’s sentence has no very 402. Spiro 1981, p. 110. 403. Spiro 1981, pp. 108–9. 404. Spiro 1981, pp. 123–4. 405. See e.g. Elkins 1996. 406. Spiro 1981, pp. 124–5. 407. Bachhofer (1944, p. 111) did declare that “one style reigned supreme at one time,” but under pressure he was willing to allow that “reigned supreme” does not mean “reigned alone” (Bachhofer 1945, pp. 243–4). As for Loehr, he never took such a position. He believed in a main line of development; he did not believe that all the bronzes made at a given moment belonged to the main line. When Spiro takes his 1953 paper to be proposing a sequence of five period styles, each the only style of its time, her conflation of the views of Bachhofer and Loehr has misled her. (At the one point where she sees disagreement between the two—she suggests on pp. 117 and 127 that Bachhofer believes the coherent taotie to precede the dissolved taotie while Loehr believes the dissolved to come first —they are speaking of different periods. The dissolved taotie Bachhofer is referring to is the same late Western Zhou motif whose dissolution Loehr traced in his 1936 paper. Loehr and Bach- hofer are describing particular historical events. Spiro perhaps believes them to be championing different laws of artistic evolution, Loehr favor- ing “dissolved to coherent,” Bachhofer “coher- ent to dissolved.”) [122] max loehr and the study of chinese bronzes clear meaning, but such meaning as it does have is dependent on the belief that style is a property of objects. This is a belief shared by most art historians. Most of us think and speak of style as though it belonged to the objects we study and as though it were an entity capable of evolving. But we cannot define an entity that behaves in this way, and what emerges from our daily practice is something quite different. Operationally style is a simple and unproblematic concept. It becomes problematic only when it is detached from practice and pressed into the service of problematic historical conceptions. To explain the point, let me take a celebrated work of European architecture, Claude Perrault’s east front of the Louvre (Fig. 81), and ask: What is its style? The question has no answer. Considered in isolation, the Louvre has no style, it only has physical properties. But of course no architectural historian ever dis- cusses the Louvre in isolation. Explicitly or implicitly it is always being compared with other buildings, and it then has any number of styles, according to the com- parison being made or assumed. In other words, the comparison the historian has in mind determines which of the building’s physical properties are relevant to a characterization of its style. This will be clear if we consider a few of the answers that might in practice be given to the question, “What is the style of the east front of the Louvre?” It belongs to the classical style. If we call the Louvre a classical building, we are contrasting it with buildings we consider non-classical, such as Gothic cathedrals, and we are grouping it together with such build- ings as the Parthenon and the Colosseum and the U.S. Capitol. In this context the features that constitute its style are features that it shares with the Parthenon and the Capitol but that we would not find in a Gothic building or a Chinese building. If we wanted to list them we would begin with the classical orders, that is, with columns and capitals and entablatures designed and used in accordance with cer- tain rules. Its style is French. An architectural historian who calls the east front of the Louvre a French building is likely to be contrasting it with con- temporary Italian buildings, and probably also with the designs by Bernini (Fig. 82) that Louis XIV rejected in favor of Perrault’s. To this characterization the classical orders are irrelevant; there is nothing dis- tinctively French about Corinthian columns (Bernini used them too). Summerson tells us, “The division of the composition by centre and end blocks is specifically French, and so is the sculptural decoration.”408 These are, in other words, features that we could find in other French buildings but not in buildings in Italy or elsewhere. A national or period style is made up of features and qualities that we encounter often in the nation or period in question but seldom elsewhere. If we find that a certain kind of sculptural decoration appears regularly on French build- ings but seldom in other countries, then we regard it as characteristi- cally French. It is something that a French architect invented or adopted and that French architects of later generations retained because they and their patrons liked it. 408. Summerson 1963, caption to plate 47. style [123] The Louvre is a Baroque building. The terms Renaissance and Baroque are in the first instance names for period styles in Italian art. The historian who applies the label Baroque to the east front of the Louvre is saying that it has borrowed from seventeenth-century Italian buildings some of the qualities that distinguish them from sixteenth-century Italian build- ings. To call the building French is to direct attention to qualities that distinguish it from Bernini’s designs; to call it Baroque is to say that Bernini and Perrault have something in common. Since seventeenth- century France did not adopt Italian styles wholesale (witness the rejec- tion of Bernini’s designs), finding anything Baroque about Perrault’s design may be difficult. But the features that attach it to the Baroque style will, if they exist, be no different in principle from the features that attach it to any other style: they will be features shared with other build- ings we call Baroque and absent from buildings we call Renaissance. Anthony Blunt points to the great size of the columns, their pairing, and the recession of the wall behind them; Summerson detects an element of Baroque rhetoric.409 The Louvre belongs to the style of Perrault. When we speak of an artist’s per- sonal style, we are speaking of traits that recur in his work but that are not seen in the work of other artists—the distinctive features, or a dis- tinctive constellation of features, that enable us to recognize his hand in an unsigned work. A historian who wished to find elements in the Louvre’s design personal to Perrault would compare it with the work of other architects in the service of Louis XIV. He would be looking for traits special to Perrault, features not found in buildings designed by Le Vau or Le Brun. Those features, whatever they might happen to be, would have no significance apart from the comparisons that isolated them. They would be entirely different from the features that make the build- ing classical and the ones that make it French, since those are shared by Le Vau and Le Brun; and they would differ also from the features that make it Baroque, since those are shared by Italian buildings.410 It is early Perrault. If Perrault had had a long and productive career (he did not), with the Louvre’s design coming at an early stage, we might discover systematic differences between it and his later works. Per- haps we would find the Louvre more sumptuous, the later buildings more restrained, chaste, bare of ornament. “Early Perrault” would then mean sumptuous and richly ornamented. But “sumptuous and richly ornamented” is not an objective fact about the Louvre. It is not a physi- cal property of the building. By comparison with the Bernini drawing shown in Figure 82, Perrault’s design is not sumptuous, it is downright austere. In sum, we cannot produce one definitive characterization of the style of the east front of the Louvre. Not even the most exhaustive inventory of its physical prop- erties would meet that description. A complete inventory of physical properties would not be the building’s style, it would be the building itself. 409. Blunt (1973, p. 332): “In certain respects it is Baroque: the scale of the Order, the depth given by the free-standing colonnade, the variety of rhythm due to the coupling of the columns.” Summerson (1963, p. 33): it uses “the classical language of architecture with force and drama in order to overcome our resistance and persuade us into the truth [of what it has to tell us about] the paramount magnificence of Louis XIV.” 410. In fact Perrault, Le Vau, and Le Brun were all members of the committee Louis XIV appointed to design the east front of the Louvre, and how much of the design should be credited to Perrault remains in dispute. If we had earlier buildings by him that we could use to define a “Perrault style,” we might be in a better position to detect his hand in this building. [124] max loehr and the study of chinese bronzes Style is not a property of single objects considered in isolation. It is a way of talking about one object’s relationships with other objects. It comes to be thought of as a property of the object only because the comparisons we make are so often unconscious and automatic. Recall for a moment the Pillsbury yu described in Loehr’s catalogue entry (Fig. 77). By explicitly comparing the Pillsbury bronze with another Style V vessel of the same shape (Fig. 76), Loehr was able to give a precise and subtle characterization of its qualities. Suppose we were to settle for a much looser characterization and just labeled the vessel “Style V.” We would still be making comparisons—we would be announcing the result of comparing the Pillsbury bronze with five large subgroups of the bronze corpus—but we would be leaving our comparisons unspoken. To ascribe a style to an object is to state the result of a comparison. To suppose the style to be a property of the object is only to be unaware of one’s comparisons.411 Loehr’s use of the concept of style throughout his writing can, I believe, be understood in these very concrete and straightforward terms, as nothing more than a way of talking about comparisons. Comparisons were essential to every aspect of his enterprise, the only possible approach to the questions that inter- ested him. They made it possible to talk about qualities. They were necessary for talking about change, hence about history. And they were necessary to the study of artistic creativity because they are intrinsic to the artist’s creative process: what the artist creates is his reaction to what he knows. Style is inseparable from the practice of art history because without comparison there can be no connected history of artistic invention.412 The problem Loehr set himself was to construct a history of the bronzes. For simplicity let us visualize the history he arrived at as a single typological sequence. This he created by arranging a selection of vessels into a linear ordering in which each was separated from its predecessor by small differences. For the arrange- ment to constitute a history the differences had to be of a particular kind. He could have arranged the vessels according to size; or he could have arranged them by color, putting them into a sequence in which the shade of green varied smoothly from one to the next; but historical information would not have emerged from such orderings. His purposes required him to choose differences likely to have chronological significance. The features he settled on as likely to be chronologi- cally sensitive included one or two that are technical, such as wall thickness (thin walls probably reflect the early bronze industry’s parsimonious use of metal). But the most revealing differences by far, the ones richest in chronological informa- tion, were the ones that resulted from artistic decisions made by designers. Loehr sought to arrange vessels in a sequence in which each vessel differed from its immediate predecessor by conscious and purposive changes, changes made by a designer for reasons of design. To do this, of course, he had to assume that the changes were rational decisions that he could understand and intuit: he had to believe that he could enter into the thinking of the bronze caster who was looking at one vessel and designing the next. But notice also what he did not assume about the changes he charted. He did not suppose that they were anything but free inventions of designers. Though they might seem in hindsight to fall into a logical pattern, they were not foreordained. They did not follow an inexorable sequence that recurs in the arts of other times 411. In an essay on style by Gombrich (1968) the repeated occurrence of the word “distinctive”— the essay begins “Style is any distinctive, and therefore recognizable, way in which . . . an artifact [is] made”—might be taken to betray an awareness that comparison is intrinsic to the notion of style, but Gombrich never actually says so. The point that styles are entirely a function of the comparisons historians choose to make applies even to the largest and most familiar of the categories we operate with. If we do not nor- mally think of the classical style as a national or period style, while we do think of, for example, Gothic as the style of a particular place and time—western Europe in the late Middle Ages— this is only a consequence of the way in which we have chosen to define and use the two. We define “the classical style” in terms of the clas- sical orders and certain ideas about proportion because we want to focus attention on features that Greek temples share with Roman amphi- theaters, French palaces, and American govern- ment buildings. We do not think of it as the style of a particular place and time because the build- ings we wish to compare are scattered over the world and over a period of 2,500 years. When we wish to pursue different questions we group the same buildings differently and define other styles (e.g. “Greek Revival”). That we do often think of Gothic as a period style despite the fact that buildings we describe using the word “Gothic” have been constructed continuously over the last 800 years is entirely a matter of choice (or habit). In a sense, of course, the classical style is a historical reality codified in the textbooks of architecture schools. But rules and schooling are merely mechanisms by which the architects of a given moment derive ideas from the buildings of the past. Buildings and decisions about design- ing them are the reality that we (and architects) invent styles to help us talk about. If we lose sight of this and suppose that classical or Gothic or any other style has an existence independent of our definitions—if for example we forget that Romanesque, Gothic, and the boundary between them are our own inventions—we risk entangling ourselves in imaginary problems. 412. Because he wanted to talk about style, Hein- rich Wölfflin (1864–1945) introduced a second lantern-slide projector to the art history lecture room, a position from which it has never been dislodged. There is nothing accidental about the fact that we cannot do without the second projector. Only style “makes it possible to get beyond a history of artists or a descriptive cata- logue of works toward a meaningfully ordered whole” (Loehr 1964, p. 186). style [125] and places, and they could not have been reliably predicted by an ancient observer. The bronze caster looking at a vessel with heavy flanges might design a new one with still heavier flanges; or he might design one with no flanges at all (and, if his patrons approved, might thereby trigger a reversal of taste that radically shifted the course of the bronze industry). In retrospect we can see either decision as a rational reaction to an existing state of affairs, but we could not have predicted the decision. As in other branches of history, the historian of art can find a retrospec- tive logic in what did happen without in any way implying that it had to happen. It is here that Loehr parts company with many other art historians, including his teacher, and it is here that the concept of “style” becomes problematic. Impressed by the retrospective logic of their own stretches of art history (usually the Italian Renaissance), many art historians would like to believe that long-term patterns of stylistic change are governed and hence explained by laws.413 For writers of this turn of mind, style becomes a metaphysical entity with a life of its own, a life that unfolds independent of individual caprice. It is not a way of talking about comparisons between objects, it is a phenomenon that precedes and shapes the objects. Artists are the instruments through which it acts. Period style, instead of being the shared inheritance of the artists of a given time, is a mysterious force that determines what they do. For art historians who think in this way, the history of art is explained by whatever power is imagined to direct the movement of style. For some it is the spirit of the age, or a world view, or a national character or racial essence. For others it is a law internal to art itself: a style, like a plant, is imagined to have a life history, and its growth to unfold according to an immanent law— hence talk of the birth, maturity, and death of styles.414 But whatever is imagined to rule the movement of style, style rules artists.415 This is emphatically not a part of Loehr’s understanding of style or of history. Style as something that is born, matures, and dies is conspicuously absent from his work. He never suggests that Style III is in any way incomplete, or that it is an immature stage on the road to Style V perfection. He would instead agree with Jean Bony that “each work of art (or group of closely related works) is meant as a final statement for its moment in time and each moment has its right to be considered ultimate.”416 In his 1964 article “Fundamental Issues in the History of Chinese Painting” he firmly distanced himself from all conceptions of history that regard style as something that moves independently of the will of artists. Observing that “a new style is a new idea, a conscious change and creative event, something that marks off one period from all other periods,” he asks: If the sequence of the styles that were new in their time is properly rec- ognized as the first concern and true substance of the history of painting, a question is likely sooner or later to trouble those who study it: Was that sequence inevitable, as if determined by immanent forces? He answers that it was not: At a given time, the artists share in the same tradition and stand at the same frontier between the realities of the past and the unknown of a present. If there is a measure of inevitability in the sequence of styles, it must be due to that ineluctable frontier condition. . . . Yet, while they 413. Art history began as the study of classical Greek and Italian Renaissance art, and the grand theoretical systems that still linger in its collec- tive unconscious mostly originated as attempts to fit the art of other times and places to pat- terns abstracted from Greek and Italian devel- opments (see e.g. Schapiro 1961, pp. 92–6). The urge felt by architectural historians to find something Baroque about the east front of the Louvre may owe less to a wish to trace influences from seventeenth-century Italy than to a vague sense that Renaissance and Baroque are not mere period styles but stages in the life cycle of all styles. 414. The teleology implicit in this biological meta- phor, firmly embedded in art history by Vasari, is given short shrift in some eloquent remarks by Jean Bony (1983, pp. 1–3). 415. Ideas of the kind just sketched are sum- marized and apparently taken seriously in an oft-cited essay by Schapiro (1961), the flavor of which can be suggested by a few sentences (pp. 81–2): “The historical study of individual and group styles . . . discloses typical stages and processes in the development of forms.” “For the synthesizing historian of culture or the philoso- pher of history, the style is a manifestation of the culture as a whole, the visible sign of its unity. The style reflects or projects the ‘inner form’ of collective thinking and feeling. What is impor- tant here is not the style of an individual or of a single art, but forms and qualities shared by all the arts of a culture during a significant span of time. In this sense one speaks of Classical or Medieval or Renaissance Man.” “Common to all these approaches are the assumptions that every style is peculiar to a period of a culture and that, in a given culture or epoch of culture, there is only one style or a limited range of styles.” Hauser (1959, pp. 119–39), in a discussion of Wölfflin, gives a decisive critique of such ideas. Some remarks by Brody (2004, pp. xix–xx) are also pertinent. 416. Bony 1983, p. 1. The biological metaphor was by contrast very active in the thinking of Loehr’s student Virginia Kane. When she writes (Kane 1974, pp. 79–83) that a vessel type “attained its climactic development . . . and thereafter became extinct” or speaks of its “definitive form,” “the correct combination of all of its elements,” “this rapid progress toward perfection,” “mature perfection,” “the final stage of unimprovable typological perfection,” she is assuming—to quote Bony—that “a long sustained movement of civilization could be condensed into one short phase of supposed perfection and the whole movement then defined in terms of that moment alone.” But Loehr does not think in this way. The biological metaphor is sometimes casually invoked in his 1936 paper, e.g. on p. 33, but not in his later writings. [126] max loehr and the study of chinese bronzes are thus tied to a given point of departure, innovations do not thereby seem either predetermined or rendered predictable. In hindsight they may, however, appear to be logical—in the sense of a logical “next step” suggested by some property of the preceding style. . . . The logicality of innovations . . . can be understood, I believe, as a matter-of-course symptom of the essentially rational nature of art. Innovations do not automatically come into being. They are hard-won achievements in which the individual position of an artist in his rational, critical dis- course with tradition or previous achievements asserts itself. . . . Logi- cality stems from the rational and conscious act of innovation achieved by an individual artist; and . . . without the creative individual’s mind there would be no change, no sequence, no logicality, and no inevitabil- ity to speculate upon.417 Loehr continues by rejecting “the arcane concept of an inner unity of all creative thought in a given period, something that would somehow determine or explain a new style . . . and thus bring its autonomy to grief.” “It is a widely held belief that art expresses its time, and that meaningful parallels between the art and the cul- ture of the same period can be found.” In fact, he says, the style of a work “corre- sponds to no given objective reality supplied by the time or its character or spirit.” On the contrary, the work “is a constitutive feature in a pattern which comes to be accepted as representative of a time. The Zeitgeist is nothing but a hypostasis of the pattern engendered by a group of works. The latter are the only reality.”418 There is nothing mystical or deterministic about this conception of history. When in his 1936 paper Loehr spoke of sensing “the inner logic of development,” he used a phrase that is regularly employed by art historians who do see style as a mysterious force transcending artists and evolving according to immanent laws. Bachhofer would have understood the phrase in some such way. But there is little in the 1936 article to suggest that Loehr shared his teacher’s belief in immanent laws, and in his later writings, as in the passage just quoted, it is clear that “logic” had no such meaning for him. On the contrary, it refers only to that retrospective pattern that the historian finds in events that were not predictable at the time of the actors. Thomas Kuhn, in an essay critical of efforts to find laws of history, describes historical explanation in terms that Loehr would surely have approved: The final product of most historical research is a narrative, a story, about particulars of the past. In part it is a description of what occurred (philosophers and scientists often say, a mere description). Its success, however, depends not only on accuracy but also on structure. The his- torical narrative must render plausible and comprehensible the events it describes. In a sense to which I shall later return, history is an explana- tory enterprise; yet its explanatory functions are achieved with almost no recourse to explicit generalizations.419 No one, I think, still believes that history is mere chronicle, a collection of facts arranged in the order of their occurrence. It is, most would con- cede, an explanatory enterprise, one that induces understanding, and it must thus display not only facts but also connections between them.420 417. Loehr 1964, pp. 188–9; cf. Loehr 1967 (unnumbered page, paragraph at n. 23). In the same vein Bony (1983, pp. 1–2) writes: “The driving force of human inventiveness [is] a criti- cal dissatisfaction with the immediate past. . . . The past must be relived as what it was when it was happening: as a sequence of distinct and unforeseeable presents.” Compare also Gom- brich (1968, p. 357): “By placing an oeuvre into a continuous chain of developments, we become alerted to what its creator had learned from pre- decessors, what he transformed, and how he was used, in turn, by later generations. We must only guard against the temptation of hindsight to regard this outcome as inevitable. For every one of the masters concerned, the future was open.” 418. Loehr 1964, p. 190; cf. Loehr 1961, p. 149. Compare also Loehr 1962, p. 231: “The term, spirit of an epoch, connotes an inner unity which in historical retrospect is recognizable, a deter- mining force which sets that epoch apart from others. It can be argued, however, that the only kind of unity of a given period is the past, the sum of tradition and memories, while there is no unity in its most distinctive aspect, namely, its new creations; and that there can be no epochs except in the course of each particular intellec- tual discipline because the creations—which are personal, unpredictable, and causeless—merely happen to be synchronous with other events. The reviewer’s contention is that there are no mean- ingful ‘epochs’ comprising the totality of life and thought.” 419. Kuhn 1977, p. 5. 420. Kuhn 1977, p. 15. style [127] [The explanatory force of a historical narrative] is carried, in the first instance, by the facts the historian presents and the manner in which he juxtaposes them.421 An ability to predict the future is no part of the historian’s arsenal. He is neither a social scientist nor a seer. It is no mere accident that he knows the end of his narrative as well as the start before he begins to write. History cannot be written without that information.422 The basic criterion for having done the job right is the primitive recogni- tion that the pieces fit to form a familiar, if previously unseen, [pattern of behavior]. . . . I am suggesting that in history that obscure global relationship carries virtually the entire burden of connecting fact. If history is explanatory, that is not because its narratives are covered by general laws. Rather it is because the reader who says, “Now I know what happened,” is simultaneously saying, “Now it makes sense; now I understand; what was for me previously a mere list of facts has fallen into a recognizable pattern.”423 None of this requires any adjustment to fit the historical territory in which Loehr worked, a territory concerned with works of art. What is distinctive to Loehr’s area of study is only that Kuhn’s “facts” are material objects, and what brings them into a “recognizable pattern” is a guess at the artistic decisions that led from one to the next. When Jean Bony asked “For how could one give an idea of the pace of artistic life in periods of great inventiveness, if not by placing the stress firmly on the dynamics of change and on the unpredictability of stylistic invention?”424 he was speaking of French Gothic architecture, but Loehr obviously felt that in study- ing Shang bronzes he too was concerned with artistic life in a period of great inventiveness. His conviction on this point has not always been shared, however, and I should like to close with a comment on the status of the bronzes as art and on the possibility of understanding art that is not explained to us by documents contemporary with it. It is sometimes asserted that art is a modern concept, one invented by the Renaissance, or by Kant, and that nothing we can properly call art existed in ear- lier periods. In effect this is an attempt to withdraw the word from common use. The attempt is unlikely to succeed, nor should we want it to. We cannot doubt that objects were consciously designed with their visual effect in mind long before Kant and the Renaissance. The existence of such objects in ancient times is no more uncertain than the existence of ancient music. Neither Shang bronzes nor Egyptian wall paintings nor Assyrian reliefs acquired their appearance by accident. The bronzes were designed, and to the kings and princes who commanded their manufacture it mattered that this should be so. Ancient patrons were less con- fused than we are about the uses of art. A slightly different position holds that, though the objects may be art, we can- not know how ancient viewers reacted to them. We have no access to Shang minds, we are told; unable ever to know the intentions of casters or the standards by which ancient viewers judged, we frivolously impose our own taste on works that 421. Kuhn 1977, p. 16. Compare Loehr (1971, p. 294): “If there is any arbitrariness in the selec- tion of what is regarded as historically signifi- cant, we have to admit that there is no other way. Every historian is arbitrary to some extent. . . . The decisive criterion is the meaning that springs from the act of selecting.” 422. Kuhn 1977, p. 16. 423. Kuhn 1977, pp. 17–18. 424. Bony 1983, p. 3. [128] max loehr and the study of chinese bronzes we have no way of understanding as they were originally understood. To this a variety of answers can be given. Few students of ancient literature would maintain that we have no access to the emotions of ancient readers. We are confident that those readers, like us, were touched by Euryclea’s recognition of Odysseus. Why then should we believe that we are cut off from the emotions of ancient viewers of Assyrian lion-hunt reliefs? Obviously we cannot assume that all our responses to the material culture of the past are authentically ancient (we do not assume this about our responses to ancient literature). But to assume that all our responses are wrong and historically irrelevant would be equally unsafe. That is a logic that leads to solipsism. The foregoing discussion of Loehr’s work has already given another, more spe- cific answer to the claim of inaccessibility. In order to construct the typological sequences from which he deduced his history of the bronzes, Loehr had to believe, at each step in each sequence, that he could see the artist’s reason for taking the step and that he could experience the reaction that persuaded the patron to ratify the step. If he had been wrong to think that he could understand the reasoning of successive artists; if the reactions of Shang patrons had been totally different from his own; if his immersion in his material had not given him some feeling for the taste of Shang viewers; how could he have arrived at a sequence that effectively predicted the sort of bronze that archaeologists would find at Zhengzhou? If his confidence had been misplaced, none of his typological sequences would be any- thing but a random ordering of objects. Art history today has many interests besides those that engaged Loehr and many methods besides those he employed, but most of its favorite lines of investigation depend on written evidence and thrive only in periods abundantly furnished with documents. When our subject is the art of periods or cultures that lacked writing or that have left us no documents bearing on their art, art history as practiced by Loehr is our only means of access. We will never know anything about Neolithic or Palaeolithic art beyond what we can read from objects, but surely we would not want to deny prehistoric people an aesthetic sense merely because they did not tell us in writing that they possessed one. Nor does the arrival of documents on the scene render irrelevant a concern with the visual. To the extent that an artist’s thinking is nonverbal, it is unlikely to be recorded in texts. Beethoven’s thinking is embodied in his music. It never existed in words because he did not compose in words. The recovery of visual thinking will always depend in large part on the kind of immersion and sym- pathetic engagement that Loehr practiced so well. If the Shang period were as profusely documented as the Italian Renaissance, our understanding of Shang bronzes would no doubt be immeasurably enriched, but our understanding of why they look the way they do might not be too dramatically changed. No matter what Shang authors might happen to tell us, our questions about artistic inven- tion, about the invention of imaginary animals for example, would by and large remain visual problems. Such problems are not necessarily of much interest to scholars in neighbor- ing fields (though archaeologists and anthropologists do at times make state- ments about imaginary animals). In art history, however, they are central concerns. Unless we take them seriously, our effort to understand Shang bronzes, Assyrian style [129] reliefs, and countless other things of high importance to the ancient societies that made and used them cannot succeed. The approaches criticized in this book have in common a refusal to treat the bronzes as objects consciously designed and redesigned, over generations, to serve functions that were in large part visual. It is a failure to grasp that visual power can be a functional attribute, perhaps even a refusal to believe that ancient people cared about or were capable of design.