Linguistic Theories and Intellectual History in Journal of Art Historiography Number 1 December 2009 Linguistic Theories and Intellectual History in Michael Baxandall’s Giotto and the Orators Allan Langdale A close reading of Michael Baxandall’s Giotto and the Orators1 Michael Ann Holly has noted that, gives insight to one of the most sophisticated accounts of the relationship between words and images in art history. The volume, which appeared in 1971, works an eclectic synthesis of linguistic theories contextualized in remarkable detail and it remains one of the most intellectually rich accounts of Italian Renaissance/Early Modern humanist art criticism and its structural relation to Renaissance painting. This paper, briefly touching on the book’s importance to Italian Renaissance/Early Modern studies, will focus more specifically on the linguistic theories that have relevant interplay with Baxandall’s enterprise. At the time, Baxandall’s insertion of an advanced linguistic methodology into art historical discourse signaled a paradigm shift in art historical studies, representing a crucial marker of the linguistic turn in art history and, moreover, a work which signaled a confrontation of the humanistic confidence of the text-based Warburg Method with a contemporary epistemological anxiety about language and its limitations. I have attempted to map out some of the linguistic theories and demonstrate how they complement certain intellectual strands in anthropology and art history. Baxandall’s scholarly career has been a sustained reflection on the impossibility of closing the gap opened up between words and images in the practice of art history which he inherited, the discipline that supposedly exists in order to bring the two realms of experience into some kind of congruency.2 This ‘sustained reflection’ begins with a group of articles Baxandall published in the Journal of the Warburg and Courtauld Institutes in the 1960s,3 and these formed the core of what would become Giotto and the Orators. To Holly’s comments I would 1 Michael Baxandall, Giotto and the Orators. Humanist observers of painting in Italy and the discovery of pictorial composition, Oxford: The Clarendon Press, 1971. 2 Michael Ann Holly, ‘Patterns in the Shadows: attention in/to the writings of Michael Baxandall’, Art History, 21.4, Dec. 1998, 468. 3 They are as follows: Michael Baxandall and E. H. Gombrich, ‘Beroaldus on Francia’, Journal of the Warburg and Courtauld Institutes, 25, 1962, 113-115; Michael Baxandall, ‘A Dialogue on art from the Court of Leonello D’Este: Angelo Decembrio’s De Politia Litteratia pars LXVIII’, Journal of the Warburg and Courtauld Institutes, 26, 1963, 304-326; Michael Baxandall, ‘Bartholomaeus Facius on Painting’, Journal of the Warburg and Courtauld Institutes, 27, 1964, 90-107; Michael Baxandall, ‘Guarino, Pisanello and Manuel Chrysoloras’, Journal of the Warburg and Courtauld Institutes, 28, 1965, 183-204. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 2 add that Baxandall’s work could be, similarly, positively compared with Foucault’s in the former’s attention to the discursive practices of art history and art historical language. As indicating a moment in the intellectual history of art history, the book is a major milestone in the development of the present state of affairs in the discipline. It is for these reasons as well that it would be rewarding to give close consideration to the book with which Baxandall gained an international reputation.4 This study tries out a variety of comparative samples. For example, one of the ways to reveal significant aspects of Baxandall’s work is to set the objectives of Giotto and the Orators alongside the aims of Panofsky’s Gothic Architecture and Scholasticism, since a comparison of Panofsky’s concept of mental habits or habitus and Baxandall’s notion of habits with language goes far in exposing the methodological parameters of both authors’ works. 5 Of the themes fundamental to Baxandall’s work in Italian Renaissance art at this time, two were paramount. Firstly, there was a pair of related problems closely linked to contemporary art historical and cultural historical debates about Italian Renaissance art: the specific relationships between humanism and the form and content of visual arts, and, more generally, the relationship between written and visual texts. In addressing the first of these problems Baxandall develops a radical contextualization—literary, linguistic, and social—of Alberti’s de pictura, a book which had come to occupy a primary and decidedly authoritative position in articulating Italian humanist art theory, thereby also dominating art historical assessments of Italian Renaissance painting. Baxandall accomplishes this contextualization through a close consideration of a compositional element cherished by Italian humanists, the periodic sentence, which, as Baxandall explains, directed attention to and influenced evaluations of the visual by providing and organizing descriptive criteria for art appreciation and criticism. In addressing the second issue, the relationship between written and visual texts, Baxandall regards the general constitutive role of language in visual culture. To grapple with this problem, he recruits notions from linguistic relativists such as Benjamin Lee Whorf and Melville Herskovits. Another way of illuminating issues in Baxandall’s book is to examine some of the critiques of methods from which Baxandall has borrowed his assumptions. The case of linguistic relativism is a case in point. Noting the manner in which Baxandall selects and adapts such methods to his own requirements does much to reveal the intellectual and scholarly characteristics of his arguments. 6 4 The concerns, though modified, are also present in Baxandall’s Painting and Experience in Fifteenth- Century Italy. A primer in the social history of pictorial style, Oxford and New York: The Clarendon Press, 1972. See as well Michael Baxandall, ‘The Language of Art History’, New Literary History, 10.3, Spring 1979, 453-465. 5 Panofsky’s concept of the habitus was adopted and much expanded by Pierre Bourdieu, who also championed Baxandall’s work. See Allan Langdale, ‘Aspects of the Critical Reception and Intellectual History of Baxandall’s Concept of the Period Eye’, Art History, 21.4, Dec. 1998, 479-497. Henceforth ‘Aspects of the Critical Reception’. Reprinted in About Michael Baxandall, Adrian Rifkin, ed. Oxford: Blackwell, 1999, 17-35. 6 The book is another example of the art historical uses of complementary theories from anthropology (Whorf) and experimental psychology (Herskovits). One is tempted to consider Baxandall’s Warburg Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 3 It is significant that Baxandall’s exposition does not originate with art objects but with literary remains that contain, at some level, art criticism, theory, or descriptions of objects of art. By approaching the art of the Quattrocento from written to visual text—against the grain of what had been the usual art historical procedure where the art object was selected, then textual and other ‘con/textual’ information was appended to it to support a reading—Baxandall discovers discrepancies in the accepted views of the relationships between humanism and the arts. Because Baxandall’s notion of context in Giotto and the Orators is decidedly literary and linguistic, the manner in which he selects and examines documentary evidence is crucial to evaluating his epistemology. Embracing the basic logic of contextuality, Baxandall assumes that texts are best interpreted with reference to external factors such as the aspects of production, consumption, and reproduction in their specific milieu. But, more than this, he also articulates the apparatus of mediation by which humanist literary documents might have affected visual art within that dynamic and reciprocal context. Giotto and the Orators focuses on a body of literary remnants that address visual art and, through a close examination of their style and content, brings attention to their generic qualities and compositional dependencies upon grammatical and structural paradigms. Where cultural and art historians of the Italian Renaissance had often given major humanist texts such as de pictura an emphatic centrality and ignored the rather more bulky remainders of humanist textual production as simply poor literature, fawning encomia, or empty posturing, Baxandall re-inserts the major text—here Alberti’s de pictura and its concept of compositio—within the wider context of less vaunted, pragmatic humanist literary concerns. Given these aims it is crucial to situate Giotto and the Orators within the scholarship on Italian humanism that, among other things, attempted to foreground the rhetorical standards by which humanism constituted itself and its objects of attention.7 Baxandall’s revisionary position towards the humanists and the products of humanism is derived in part from P. O. Kristeller, 8 connections and Warburg’s anthropological interests and Gombrich’s earlier uses of experimental psychology in works like Art and Illusion. whose work aimed at articulating the pragmatic applications and professional status of humanism as 7 One early example is John R. Spencer, ‘Ut rhetorica pictura: A Study in Quattrocento Theory of Painting’, Journal of the Warburg and Courtauld Institutes, 20, 1957, 26-44. In this formulation rhetoric ceases to be merely cosmetic and becomes constitutive. 8 Baxandall notes Kristeller’s help in the preparation of Giotto and the Orators in his preface, p. viii. See also the interview with Michael Baxandall first published in Allan Langdale, Art History and Intellectual History: Michael Baxandall’s Work between 1963 and 1985, PhD Dissertation, University of California at Santa Barbara, 1995. These interviews are reprinted in this volume of the Journal of Art Historiography, 1.1, 2010. For Kristeller see p. 4. Henceforth, Baxandall Interview, 4. Kristeller’s ideas are presented in an accessible form in the section entitled ‘Humanism’, in The Cambridge History of Renaissance Philosophy, Charles B. Schmitt, Q. Skinner and E. Kessler, eds, Cambridge: Cambridge University Press, 1988, 113-137. This volume offers a comprehensive bibliography which may be supplemented by the bibliography in the volume by Brian P. Copenhaver and Charles B. Schmitt, Renaissance Philosophy, Oxford: Oxford University Press, 1992. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 4 opposed to propagating their self-constructed mythologies.9 Those scholars attracted to Kristeller’s point of view, and similarly oriented Warburg scholars like Ernst Gombrich, often focused their attentions on less heroic humanist documents so as to discover the various ordinary uses to which they were put. Historically significant material was revealed not only in the content of the documents but also in the forms that played a role in constituting that content. This type of critical examination was accompanied by four supplemental strategies to complicate and broaden Renaissance humanism’s identity: the focusing on minor humanists and their work (or minor works by not so minor humanists),10 the examination of humanist activity in locations other than Florence,11 attending to the more vernacular applications of humanist texts12 and, finally, examining the qualitative effects of literary and rhetorical standards placed on humanist thought and textual production.13 By examining marginalized humanist texts produced and consumed in social settings other than Florentine, and by tracing some of the rhetorical and compositional templates (ekphrasis, periodic sentence) which helped Alberti formulate and express his ideas, Baxandall attempts to draw attention to the fact that Alberti did not, as he wrote in an earlier article on the subject, create ‘modern art criticism out of a void’.14 9 Warburgian scholarship, it is important to note, was amenable to Kristeller’s position which, indeed, may have been formed by it to some extent. An early article which characterized a similar direction was Augusto Campana, ‘The Origin of the Word Humanist’, Journal of the Warburg and Courtauld Institutes, 9, 1946, 60-73. It is the first reference Baxandall gives in Giotto and the Orators along with, significantly, P. O. Kristeller, Renaissance Thought: The Classic, Scholastic and Humanist Strains, New York: Harper and Row, 1961. See also P. O. Kristeller, Studies in Renaissance Thought and Letters, Rome: Edizioni di storia e letteratura, 1956. It is not that Baxandall tries to unseat Alberti, since he recognizes de pictura’s unique and exemplary contribution. It is simply that he is 10 For a similar orientation towards humanist attitudes see E. H. Gombrich, ‘Apollonio di Giovanni: A Florentine cassone workshop seen through the eyes of a humanist poet’, in Norm and Form: Studies in the Art of the Renaissance, London and New York: Phaidon, 1966, 11-28. Baxandall edited this volume for Gombrich, and thus knew it well. Gombrich was Baxandall’s dissertation advisor at the Warburg in those years. See Baxandall Interview, 8. 11 Several studies from the late 1960s and 1970s record these expansions. Examples include David Herlihy, Medieval and Renaissance Pistoia: The Social History of an Italian Town 1200-1430, New Haven: Yale University Press, 1967; William M. Bowsky, The Finances of the Commune of Siena, 1287-1355, Oxford: The Clarendon Press, 1970; Werner Gundersheimer, Ferrara: The Style of a Renaissance Despotism, Princeton: Princeton University Press, 1973; and Lauro Martines, Power and Imagination: City States in Renaissance Italy, New York: Knopf, 1979. 12 Along with the work of Kristeller and his followers, see: Charles Trinkaus, ‘A Humanist's Image of Humanism: The Inaugural Orations of Bartolommeo della Fonte’, Studies in the Renaissance, 7, 1960, 90- 147; and his In Our Image and Likeness: Humanity and Divinity in Italian Humanist Thought, Chicago: University of Chicago Press, 1970 See also John F. D’Amico, Renaissance Humanism in Papal Rome: Humanists and Churchmen on the Eve of the Reformation, Baltimore: Johns Hopkins University Press, 1983. Important to consider among these works are those studies on education in the Renaissance, particularly the work of William Harrison Woodward which includes Studies in Education during the Age of the Renaissance 1400-1600, Cambridge: Cambridge University Press, 1924; and Vittorino da Feltre and other Humanist Educators, Cambridge: Cambridge University Press, 1921. 13 See in particular Jerrold E. Seigel, Rhetoric and Philosophy in Renaissance Humanism: The Union of Eloquence and Wisdom, Petrarch to Valla, Princeton: Princeton University Press, 1968; and Nancy S. Struever, The Language of History in the Renaissance: rhetoric and historical consciousness in Florentine humanism, Princeton: Princeton University Press, 1970. 14 Michael Baxandall, ‘Guarino, Pisanello and Manuel Chrysoloras’, Journal of the Warburg and Courtauld Institutes, 28, 1965, 200. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 5 challenging, first, what seemed at the time to be the book’s autocratic hold on the world of Renaissance art history, and, second, how de pictura was used by modern scholars to posit canted notions about the origins and developments of Italian Renaissance painting. Baxandall’s intent in Giotto and the Orators is to assign de pictura a place within its literary family group, thereby exposing its generic dependencies and most constituent formal attributes. It is this critical position towards the documentary that makes Giotto and the Orators most successful in terms of its revisionary project. Alberti’s de pictura is the result of relations, of a system, of a sophisticated and interdependent language and social game that drew on diverse traditions and innovations—consciously and unconsciously—during its production. High theory does not tower above more vernacular expressions or ‘low’ theory, rather, they are interconnected by a web of various more vernacular activities in a hierarchy of social practices involving artists, humanists, patrons, and viewers. Giotto and the Orators is arranged into three chapters with a number of subsections. The first chapter outlines the literary and linguistic concerns of ‘typical’ humanists, their uses of antique sources, and their understanding of key words. The discussion moves towards a distinction between the linguistic possibilities inherent in Latin vocabulary, grammar, and literary convention, and what could be accomplished in vernacular Italian. Thus prepared, the second chapter concentrates more specifically on humanists’ comments on painting and painters in their conventional forms: the commonplace of painting as a model art, the convention of artistic progress as a progression of great men, and the rhetorical exercise of ekphrasis. These designate the limits of humanist art criticism. The third chapter finds a common denominator in the humanists’ use of the periodic sentence as a defining pattern of their practice with language and further identifies two stylistic strains, composita and dissoluta, extending into the world of painterly style. The book works towards this conclusion: In the terms of 1435—in terms, that is, of compositio— we can make out a polarization of styles common to both painting and writing. On the one had there is the painting of artists like Pisanello as Alberti saw it, and the writing of Guarino as George of Trebizond saw it; on the other, there is painting of the more or less Neo-Giottesque kind recommended by Alberti, and the more elaborately periodic writing recommended by George. Composita are opposed to dissoluta... in 1435 these things did appear as similar, in humanist terms.15 What Baxandall traces are differences in predominant styles of humanist literary composition, and pairs these with similar variations observed by humanists in the realm of visual art, thus establishing a structural homology between the two.16 15 Baxandall, Giotto and the Orators, 139. 16 Baxandall had had some exposure, though only informal, to the work of Lévi-Strauss. See Baxandall Interview, 26-7. For Structuralist aspects of Baxandall’s concept of the period eye see Langdale, Art History and Intellectual History, 151-156; and Langdale, ‘Aspects of the Critical Reception’, 479-497. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 6 Also of interest are the broad assumptions about the functions of language that are at work in Giotto and the Orators, something Baxandall himself admits to not confronting directly. This passage is found early in the book: The preliminary question becomes: in what way was the exercise of Latin words and grammar on the subject of painting likely to affect people’s attitudes and notions about painting? Obviously this raises old problems about language and cognition not solved and only intermittently recognized in this book; it is clear too that humanists brought other things to painting than just Latin words and syntax. But this preliminary question will set the character of the first chapter, both because it seems historically the proper thing, and because humanist criticism of painting is specially interesting as a linguistic case: here highly formalized verbal behaviour bears, with little inference, on the most sensitive kind of visual experience.17 Indeed, it is by examining some of these ‘old problems about language and cognition’ that we can discover some of the foundations underpinning the book’s arguments. By postulating that the lexical and semantic elements of languages are determinant factors in the cognitive patterns Baxandall places himself in line with a tradition of thinkers including Wilhelm von Humboltd,18 Ernst Cassirer,19 and the American linguists Edward Sapir and Benjamin Lee Whorf.20 The relativist/determinist thesis (frequently referred to as the ‘Sapir-Whorf Hypothesis’) states that the structural properties and lexical resources of particular languages make certain kinds of distinctions, descriptions, and discriminations more easy or difficult. This is one of the first points Baxandall wishes to establish, and he does so by citing the example of translatability, a problem which preoccupied both von Humboldt and Whorf: 17 Baxandall, Giotto and the Orators, 7. 18 See Wilhelm von Humboldt, Linguistic Variability and Intellectual Development, George C. Buck and Frithjof A. Raven, trans., Coral Gables, Florida: University of Miami Press, 1971. The best recent discussion of von Humboldt's theory of language is in Hans Aarsleff’s introduction to Wilhelm von Humboldt, On Language. The Diversity of Human Language Structure and its Influence on the Mental Development of Mankind, Peter Heath, trans., Cambridge: Cambridge University Press, 1988. von Humboldt’s On Language was first published in 1836 as the introduction to Über die Kavi-Sprache auf der Insel Java. 19 Cassirer’s notions about the relationship between culture and language are found in Language, volume one of Ernst Cassirer, The Philosophy of Symbolic Forms, Ralph Mannheim, trans., New Haven and London: Yale University Press, 1953. The work was originally published in 1923. 20 A general introduction to Whorf’s thought is Benjamin Lee Whorf, Language, Thought, and Reality. Selected Writings of Benjamin Lee Whorf, John B. Carroll, ed., Cambridge, Mass.: MIT Press, 1956. The name of the linguist Edward Sapir (1884-1939) is often mentioned in conjunction with Whorf, since it was actually Sapir who first suggested Whorf explore what was essentially the relativist thesis. A number of books treat the Sapir-Whorf hypothesis. Recent discussions include Emily Ann Schultz, Dialogue at the Margins: Whorf, Bakhtin, and Linguistic Relativity, Madison: University of Wisconsin Press, 1990; and Julia Penn, Linguistic Relativity vs. Innate Ideas. The Origins of the Sapir-Whorf Hypothesis in German Thought, The Hague: Mouton, 1972. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 7 All languages are...systems for classifying experience: their words divide up our experience into categories. Each language makes the division in a different way, and the categories embodied in the vocabulary of one language cannot always be transferred simply into the vocabulary of another language.21 And further, on linguistic enforcement, clearly suggesting affinity with linguistic relativism: But there are areas in which one language differentiates more than another, or in a different way, and this put identifiable pressures on what humanists said; observation was linguistically enforced.22 It is these ‘identifiable pressures’ that, indeed, need to be distinguished, and their particular forces measured. Consideration of the basic assumptions of the linguistic relativist/determinist thesis can help problematize as well as clarify Baxandall’s position. Linguistic relativists generally saw language as a manifestation of the Weltanschauung or world-view of speakers. Some translation between languages might be possible, but perfect translatability was not feasible because the variant lexical and semantic elements of the different languages create divergent, or dissimilar, thoughts and concepts. The terms ‘relativism’ and ‘determinism’ are closely related in the Sapir-Whorf Hypothesis; variant worldviews are relative to differences in linguistic components (words or lexemes) and grammar (sentences). Similarly, cognition is determined by these same linguistic resources, which precede the subject born into the culture. The Weltanschauung is manifested in a culture’s language and the parameters of thought are predicated upon linguistic units and structures. Thus the subject is born into a pre-formed, linguistically enforcing and culturally specific worldview where language determines the cognitive patterns of the subject. Two fundamental objections can be made of the Sapir-Whorf Hypothesis, against elements that Baxandall might at first seem to take as axiomatic. First of all, the hypothesis suffers from a circular or, at least, potentially tautological structure. If all differences in worldview are linguistic in origin, and all the evidence for these differences are linguistic, then we find the thesis both unprovable and unassailable. Secondly, the assumption of untranslatability is questionable. It is quite possible to make highly accurate translations between languages, even those with very different lexical resources and grammars. The determining factor is often less the differences in the languages than the skill and sensitivity, literary and cultural, of the translators.23 21 Baxandall, Giotto and the Orators, 8-9. One of the things Whorf liked to present as evidence in support of 22 Baxandall, Giotto and the Orators, 9. Each issue is discussed in Roger Brown’s Words and Things, New York: The Free Press, 1958. Baxandall's linguistic enforcement seems parallel to Brown's ‘forced observation in grammar’, discussed on p. 253 of Words and Things. 23 The most important difference is in understanding the social function of a term or measuring the cultural ‘weight’ of a concept. Diderot noted that poetry was the most difficult to translate, while translation became easier as language moved towards prose. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 8 his hypothesis was that Inuit have several words for different types of snow, while ‘Standard Average European’ languages24 such as English have only one. But is this really true? Even though we usually might not use the terms ‘slush’, ‘powder’, ‘deep-powder’, ‘icy’, ‘well-packed’, ‘dry snow’, and so on, we—and especially sub- groups like skiers or snowboarders—are quite attentive to variations in snow, and, even if there are not specific designations for each type, there is a recognition and conception of the variant types and how the diverse conditions might affect speed, maneuverability, avalanche conditions and so on. Therefore, one cannot necessarily posit cognitive discrepancies between two languages on lexical grounds, or at least it is often highly problematic to do so.25 The Sapir-Whorf Hypothesis end up being one of those unprovable yet sort of undeniable kinds of theories; it is simply a question of how far one wants to take it in any given situation. 26 Roger Brown’s Words and Things, one of only two references Baxandall gives which supply a firm link to the literature on linguistic relativism,27 criticizes much about the hypothesis but is, nonetheless, resigned to recognizing some validity in it. The critical boundaries of linguistic relativism are evident in Baxandall’s procedure, which demonstrates that he sees efficacy in the hypothesis but also some important limitations.28 One of the ways to examine how Baxandall attempts to make more efficient use of relativism is to pay attention to how he attends to documents in a specific cultural situation involving distinct cultural sub-groups. One of the complaints sounded by Roger Brown in Words and Things is that linguistic relativists failed to articulate the particularities of what he calls ‘subcommunities’ (such as the case of the skiers mentioned above): For Baxandall, linguistic resources may determine expression and understanding to a degree, but what he seems most interested in is discovering the measure of this degree and how the operations of its trajectory might be discernible in specific practices. 24 The designation is Whorf’s. 25 This objection to the Whorf-Sapir hypothesis, among others, is made in Roger Brown’s Words and Things, 234-5 and 255-6. Positing difference simply in terms of the lexicon has obvious limitations, since translation of single words is not as problematic as when dealing with more complex linguistic structures. Setting aside the several problems in the definition of the sentence, I understand the term sentence to refer to the smallest syntactic unit which reveals significant grammatical structure. The sentence was the focus of attention of many relativists including von Humboldt and Whorf, both of whom felt grammar was the most important factor in determining the cognitive patterns of peoples. For a discussion of some of the problems involved in defining such linguistic units as the sentence, see the chapter ‘Defining the Statement’ in Michel Foucault, The Archaeology of Knowledge, A. M. Sheridan Smith, trans., New York: Pantheon Books, 1972, 79-87. 26 The comment made by John B. Carroll in his introduction to Benjamin Lee Whorf. Language, Thought, and Reality. Selected Writings of Benjamin Lee Whorf, is still applicable: ‘In truth, the validity of the linguistic relativity principle has thus far not been sufficiently demonstrated; neither has it been flatly refuted’. See pp. 27-28. 27 The other is E. H. Lennenberg and J. M. Roberts, The Language of Experience: a case study, Indiana University Publications in Anthropology and Linguistics Memoir 13, 1956. Baxandall, Giotto and the Orators, 47, note 70. 28 He rejects a vulgar Whorfian formula. See Baxandall Interview, 3. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 9 We have subcommunities for whom particular regions of experience are more differentiated than they are for the majority and, within these subcommunities, there is a special lexicon to meet special cognitive needs.29 Baxandall seems to share this position and has set himself the task of articulating the linguistic features of the early Quattrocento humanist subcommunity. Brown makes another request of the linguist relativist that Baxandall is evidently sympathetic to: A more exciting form of the relativity thesis would define ‘language’ so as to include semantics and define ‘thought’ in terms of some non-linguistic behavior. In general, the thesis here would be that some non-linguistic evidence [painting, in Baxandall] of thought covaries with some linguistic evidence....I don’t know of any attempts as yet to show that an independently defined linguistic pattern has either historical or biographical priority over the thought pattern it is supposed to determine.30 Another objection, this one directed specifically at Cassirer and Whorf, is made by one of Brown’s Harvard students, the psycholinguist Eric Lennenberg: Neither Cassirer or Whorf was sufficiently explicit in stating the nature of the relationship which they purported to describe. They failed to state in general, yet concrete, terms which types of behavior were supposed to be related.31 Giotto and the Orators seems corrective of each of the shortcomings articulated by Brown and Lennenberg, leading one to consider how Baxandall proceeds, and thereby consider how far he is willing to test the axioms of linguistic relativism. One way to tease out the nuances of Baxandall’s enterprise is to compare Giotto and the Orators with Panofsky’s Gothic Architecture and Scholasticism. What interests me in this comparison is the exploration of the parameters of Panofsky’s and Baxandall's relativism and the strategies through which they apply it. Such comparison helps clarify Baxandall’s attitude about how language systems might affect the structures of the visual system. Baxandall begins his argument in Giotto and the Orators by positing the word as the primary linguistic unit which determines the expressive and denotative potentialities of the Latin language; he moves on to the sentence but will give little attention to general Latin grammar since he discovers a unifying structural element in the literary orthodoxy of the periodic sentence in which both grammatical and stylistic forms are embodied. Like the periodic sentence for Baxandall’s humanists, it was just such a larger literary element, the Scholastic Summa, which Panofsky saw as a formative factor constitutive of the aesthetic of Gothic architecture. Like 29 Brown, Words and Things, 256. 30 Brown, Words and Things, 262. 31 E. H. Lennenberg and J. M. Roberts, The Language of Experience: a case study, 2. Other criticisms by Lennenberg had appeared in Eric Lennenberg, ‘Cognition in Ethnolinguistics’, Language, 29, 1953, 463- 71. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 10 Panofsky, Baxandall sees the linguistic and literary as primary and active, exerting a patterning force on the cognitive structures through which societies organize their experience and expression. Both authors presume language to be originary and formative. ‘Notions about painting’, Baxandall writes of the humanists, ‘becomes very much a matter of their habits with language’.32 The term ‘habit’ appears often in Giotto and the Orators, and this term has certain resonances. In Gothic Architecture and Scholasticism Panofsky desired to articulate a: ….connection between Gothic architecture and Scholasticism which is more concrete than a mere ‘parallelism’ and yet more general than those individual...’influences’ which are inevitably exerted on painters, sculptors etc....What I have in mind is a genuine cause and effect relation....It comes about by the spreading of what may be called, for want of a better term, a mental habit.... Such mental habits are at work in all and every civilization.33 Some parallels can be drawn between Baxandall’s humanists’ habits with language and Panofsky’s habitus or mental habits.34 32 Baxandall, Giotto and the Orators, 7. Both authors’ attention to the literary represents an important modification of the usual linguistic relativist strategy for two important reasons. First, a literary convention is much more concrete, identifiable, and analyzable because of its high degree of codedness and its dependence on tropes. Von Humboldt and Whorf dealt with grammar and vocabulary on enormous scales, considering vast amounts of lexical and grammatical elements from which they gleaned generalities about the communicative and cognitive (and especially creative, in the case of von Humboldt) languages like Kawi, Sanscrit, or Hopi. But in Whorf’s or von Humboldt’s impressive empiricism the use of words and sentences, the ergon or work of utterances, was lost amid a sea of linguistic abstractions divorced from the 33 Erwin Panofsky, Gothic Architecture and Scholasticism: An Inquiry into the Analogy of the Arts, Philosophy, and Religion in the Middle Ages, New York: Meridian Books, 1967, 20-21. Originally published in 1951. It is revealing to compare this with a passage from Lucien Fevre’s 1942 book on Rabelais: ‘Each civilization has its mental tools [my emphasis]...each epoch...requires a renewed set of tools, more highly developed for certain needs, less for others’. See Lucien Febvre, Le problem de l’incroyance au XVIe siècle. La religion de Rabelais, Paris, 1968, 141-142. This translation, by Roger Chartier, appears in his ‘Intellectual History or Sociocultural History? The French Trajectories’, in Dominique LaCapra and Steven L. Kaplan eds, Modern European Intellectual History, Ithaca and London: Cornell University Press, 1982, 19. La Capra comments on this aspect of Febvre: ‘What defines mental equipment in these pages is the state of language, its lexicon, its syntax, the scientific language and instruments, and also the “sensitive support of thought” represented by the system of perception, whose variable economy determines the structure of affectivity’. 34 The feature of habit was also important for Whorf, though it was a feature of the ‘world view’, rather than a focus of analysis in and of itself. Note that language is the origin of all the patterns implicit in this ‘world view’. Whorf writes: ‘By “habitual thought” and “thought world” I mean more than simply language, i.e. than the linguistic patterns themselves. I include all the analogical and suggestive value of the patterns (e.g. our “imaginary space” and its distant implications), and all the give-and-take between language and the culture as a whole, wherein is a vast amount that is not linguistic but yet shows the shaping influence of language. In brief, this “thought world” is the microcosm that each man carries about within himself, by which he measures and understands what he can of the macrocosm’. From Whorf, ‘The Relation of Habitual Thought and Behavior to Language’, in Language, Thought, and Reality, 134-159. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 11 complexities of both everyday, ordinary uses of language and highly specialized practices.35 This leads directly to the second revision of a more unconstrained model of relativism: by limiting the range of study to a literary model, the analysis is similarly narrowed to more definable social sub-groups whose expressive workings of the paradigm are more accurately plotted. The patterning forces of the convention are marked very clearly, thus making their structural resonances more traceable. The context is delimited so that the linguistic relativism may be deployed so as to produce more sophisticated and enlightening results. The comprehensive Weltanschauung of a culture was sought at the expense of real diversity within cultures and their languages. Whatever strength an analysis of language potentially offered was thus dissipated by a vulgar relativism. By focusing on literary conventions, Baxandall and Panofsky narrow the field of inquiry to explore a well-defined linguistic sub-unit and the purpose it served; that is, the work that it accomplished in any linguistic or literary act. Locating the origins of their literary models in an intellectual elite—the medieval Scholastic theologians for Panofsky and the fifteenth-century humanists for Baxandall—is only the first step in the process, that is, the identification of a structural parallel. Each author must also demonstrate a point of dissemination or contact with producers of art or architecture. In other words, there must be some articulation of the path or process, a mediating practice, by which characteristics of the literary system transfer to or imprint the visual system. With Panofsky it is the architects themselves who ‘had gone to school...listened to sermons...attended the public disputationes de quolebet’;36 with Baxandall it is a complex network of subtle intellectual relationships between individuals such as Pisanello, Alberti and Guarino and others, mainly in the context of the Ferrarese court. Baxandall also suggests, as did Panofsky to a lesser degree, that formal education formed many of the patterns of expression and this is what formed habitual actions. In each case, the literary model is broken down into constituent structural elements. For the Scholastics the modus operandi of the Summa involves manifestatio, clarification, and conciliation of opposites. For the humanists, the compositio characteristic of the periodic sentence is organized along a parallelizing pattern of clauses involving comparison of similar things or procedures. Baxandall outlines the general structure: Compositio was a technical concept every schoolboy in a humanist school had been taught to apply to language. It did not mean what we mean by literary composition, but rather the putting together of the single evolved sentence or period, this being done within the framework of a four-level hierarchy of elements: words go to make up phrases, phrases to make clauses, clauses to make sentences: fit autem ex coniunctione verborum comma, ex commate colon, ex colo periodos.37 35 Von Humboldt especially repressed the ergon or language in favour of what he called its energeia. See the discussion of these terms in Hans Aarsleff’s introduction to von Humboldt, On Language, xix-xxi. 36 Panofsky, Gothic Architecture and Scholasticism, 24. 37 Baxandall, Giotto and the Orators, 131. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 12 In both cases, the literary forms are articulated, even if in Panofsky it is a discursive form that is foregrounded while with Baxandall it is a stylistic form. Even though there are several similarities, there are significant differences between Panofsky’s ‘mental habits’ and Baxandall's humanists’ habits with language. Baxandall makes clear the mediating route or practice by which words relate to visual art. The key is the act of description: I have tried to identify a linguistic component in visual taste; that is to show that the grammar and rhetoric of a language may substantially affect our manner of describing and, then, of attending to pictures and some other visual experiences.38 and, further, in a succinct demonstration of relativity: It is difficult to judge quite how much difference this made to the humanist’s description of visual things, but it made some difference. Here are two short exclamations about the painting of Pisanello. The first in neo-classical Latin, is by the humanist Guarino of Verona and was written around 1427: What understanding of light and shade! What diversity!/ What symmetry of things! What harmony of parts! The second was written in 1442 in Italian by Angelo Galli, who was secretary to Federigo da Montefeltro, Count of Urbino: Arte mesura aere et disegno/ Manera prospectiva et naturale/Gli ha data el celo per mirabil dono. No doubt Guarino and Galli did indeed observe different things in Pisanello...But it is certain that Galli is directing attention to qualities in Pisanello which Guarino could not, even if he had so wished, have verbalized in Latin.39 Panofsky does not articulate a mediating practice to the degree that Baxandall stresses the centrality of description and the role of the directing of attention that such descriptive exercises play, and so in Panofsky the relation between the mental habit generated by the Summa’s procedures seems to affect visual material in an indefinite, almost mysterious, way. It is one of the limitations of Panofsky’s procedure that he does not make the point of contact between language and art as explicit as it could be. Conversely, this is where Baxandall’s strength lies. The most significant contrast between the two books, then, is simply in the detail with which Baxandall elaborates the reciprocal social intricacies of the language game of art criticism; a game played by humanists, artists, and patrons on the social field of fifteenth-century Italy.40 I have used terms like use, work and ergon. Baxandall’s attentiveness to the use of language may be a result of his (unconscious?) sympathies with ordinary language philosophy. Baxandall’s ordinary language interests are hinted at when he 38 Baxandall, Giotto and the Orators, vii. 39 Baxandall, Giotto and the Orators, 11. 40 I use the term ‘field’ (champ) to invoke Bourdieu since Baxandall’s contextualization of practices was very amenable to Bourdieu, and Bourdieu began to champion Baxandall’s work, particularly after the publication of Baxandall’s Painting and Experience in Fifteenth-Century Italy in 1972. See Langdale, ‘Aspects of the Critical Reception’, 488-9 and Langdale, Art History and Intellectual History, 162-174. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 13 repeats, in Giotto and the Orators, the dictum of Wittgenstein's Philosophical Investigations, ‘Meaning is use’.41 Baxandall continues, ‘...and in classical Latin much of the meaning of words lay in an institution of relationships with other words, a system of cross-reference, distinctions, contraries, and metaphorical habits’.42 Compare Wittgenstein’s characterization of the language game: ‘And the result of this examination is: we see a complicated network of similarities overlapping and criss-crossing: sometimes overall similarities, sometimes similarities in detail’.43 For Baxandall language is a system44 and his aim is to articulate all the relevant components of this system and what the function of the system was for the group(s)—humanists, courtiers, painters—that utilized it.45 In this sense we might see Baxandall’s concerns in Giotto and the Orators as representing an intersection of linguistic and literary relativist interests tempered by exposure to ordinary language philosophy and the German Kunstwissenschaft tradition exemplified by Warburg, Panofsky, and Baxandall’s mentor, E. H. Gombrich. Yet, as one reads Giotto and the Orators, one wonders about sympathies other than those with the world of modern linguistic relativism or ordinary language philosophy. Baxandall notes that the humanists themselves ‘were sufficiently linguistic determinists’, 46 and it is possible that he derived some of his relativist attitude from a feeling of intellectual kinship with these Renaissance humanists. One has the feeling that Baxandall’s title at the Warburg: ‘Professor of the History of the Classical Tradition’—a title previously held by Gombrich and Gertrud Bing (to whom Giotto and the Orators is dedicated47)—in some way connects him dialogically with the Italian humanists who are his precursors and whose fundamental interests he obviously shares.48 41 Baxandall, Giotto and the Orators, 14. And see Ludwig Wittgenstein, Philosophical Investigations, G.E.M. Anscombe, trans., New York, 1953. The phrase is adapted from section 43 of the first part of the Philosophical Investigations: ‘The meaning of a word is its use in the language’. From page 20 in the above edition. Baxandall had not read Wittgenstein while preparing Giotto and the Orators, but he had studied at Cambridge where Wittgenstein’s influence was strong. Baxandall, for example, had studied literature with one of Wittgenstein’s friends, F. R. Leavis. See Baxandall Interview, 2-3 and 10. One might consider a figure, clearly admired by Baxandall, one who is often credited with the invention of philological 42 Baxandall, Giotto and the Orators, 14. 43 Ludwig Wittgenstein, Philosophical Investigations, 32. See also Garth Hallett, Wittgenstein’s Definition of Meaning as Use, New York: Fordham University Press, 1967. 44 ‘System’ is a word repeated often throughout Baxandall’s text, as it is in the works of most linguists; and we could include Saussure here as well. See Ferdinand de Saussure, Course in General Linguistics, Charles Bally and Albert Sechehaye, eds, and Wade Baskin, trans., London: P. Owen, 1959, 117. 45 Martin Jay writes of the ordinary language philosophers: ‘...these philosophers understood language less as a neutral medium of expression or representation than as a complex human activity. Language was first of all speech, which was a central component of what Wittgenstein called a form of life. Accordingly, the philosopher’s task was not to construct an ideal metalanguage neutralizing the concrete mediation of the speaker, but rather to examine and clarify ordinary language within specific social contexts’. Martin Jay, ‘Should Intellectual History Take a Linguistic Turn? Reflections on the Habermas-Gadamer Debate’, in Modern European Intellectual History: Reappraisals and New Perspectives, Dominick LaCapra and Steven L. Kaplan eds, Ithaca and London: Cornell University Press, 1982, 87. 46 Baxandall, Giotto and the Orators, 46. 47 See Baxandall Interview, 5-6. 48 At times, for example, Baxandall seems involved in a dialogic relationship with, as he puts it, a ‘number of outstanding men—the generation of Alberti and Lorenzo Valla’, who Baxandall lists on almost a page of exemplae. Baxandall, Giotto and the Orators, 7-8. Allan Langdale Linguistic Theories … Baxandall’s Giotto and the Orators 14 textual criticism and even linguistic relativism itself: Lorenzo Valla.49 Indeed, it has even been suggested by more than one scholar that Valla was the first ‘ordinary language philosopher’.50 Giotto and the Orators is a crucial text signaling the linguistic turn in art history, a turn that was, indeed, to be made even more emphatic when Baxandall published, the year after Giotto and the Orators, his Painting and Experience in Fifteenth-Century Italy. Given the confluence of cultural studies, visual culture and art history on our academic landscape, it is useful to consider the intellectual history of these tendencies. Allan Langdale is currently a lecturer at the University of California at Santa Cruz. His publications include ‘Aspects of the Critical Reception and Intellectual History of Baxandall’s Concept of the Period Eye’, Art History, 21.4, December, 1998; Hugo Münsterberg’s Photoplay and Other Writings on Film, Allan Langdale, introduction and ed., Routledge, 2002; and ‘At the Edge of Empire: Venetian Architecture in Famagusta, Cyprus’, Viator, 41.1, Spring, 2010. Allan Langdale Department of Art History and Visual Culture, Porter Faculty Services, UCSC, 1156 High Street, Santa Cruz, CA, 95064 langdale@filmandmedia.ucsb.edu 49 Baxandall Interview, 9. 50 Valla’s sensitivity to the historical relativity of language and quasi-ordinary language concerns are argued for most directly in Richard Waswo, ‘The Ordinary Language Philosophy of Lorenzo Valla’, Bibliothèque d’Humanisme et Renaissance, 41, 1979, 255-271; and, by the same author, Language and Meaning in the Renaissance, Princeton: Princeton University Press, 1987. Many of Waswo’s ideas along these lines are shared—to varying degrees—by Sarah Steuver Gravelle. See her comments in ‘Lorenzo Valla’s Comparison of Latin and Greek and the Humanist Background’, Bibliothèque d'Humanisme et Renaissance, 44.2, 1982, 269-289; and ‘The Latin-Vernacular Question and Humanist Theory of Language and Culture’, Journal of the History of Ideas, 49.3, July-Sept., 1988, 367-386. Observations in Victoria Kahn’s ‘The Rhetoric of Faith and the Use of Usage in Lorenzo Valla's De libero arbitrio’, The Journal of Medieval and Renaissance Studie, 13.1, Spring 1983, 91-109; and Lisa Jardine, ‘Lorenzo Valla and the Origins of Humanist Dialectic’, Journal of the History of Philosophy, 15, 1977, 143-164, also seem to support some degree of linguistic relativism in Valla’s thought. Waswo’s ordinary language thesis was somewhat indebted to observations by Hanna-Barbara Gerl in her Rhetorik als Philosophie: Lorenzo Valla, Munich: Fink, 1974. Waswo’s article and book drew invective from John Monfasani in his ‘Was Lorenzo Valla an Ordinary Language Philosopher?’ Journal of the History of Ideas, 50.2, Apr.-June, 1989, 309-323. Both Waswo’s and Gravelle’s replies to the attack appear immediately after Monfasani’s denunciation.