The Past Is Evolutionary, the Future Is Byzantine: Kurt Weitzmann’s Contribution to the Research on Pictorial Narration

Abstract

In the Illustrations in Roll and Codex (1947), Kurt Weitzmann developed a methodological apparatus for studying Byzantine and medieval narrative book illumination. His approach had two important features: an evolutionary narrative typology that paid attention to the narrative strategies the painter chose for presenting a story and a comparative narrative analysis that observed stories in illustrations in relation to their textual source. The focus of this paper is the personal and institutional background of this method, its context, dissemination, and legacy. Weitzmann advanced the study of pictorial storytelling through his pedagogical work and introduced it into the academic curriculum. Alongside stylistic analysis and iconography, it soon became an essential methodological tool in art history that constituted a link between the art of the Byzantine East and the Latin West. This approach also had a key influence on the style of his autobiographical writings. Weitzmann propagated the study of visual narratives through his extremely productive oeuvre and effective personal influence. In the dissemination of Weitzmann’s ideas, three institutions played a key role: Princeton University, Dumbarton Oaks, and the University of Chicago. Weitzmann’s circle made Byzantine studies the leading field for research into visual narratives over the period from around 1940 to 1980.

1. Introduction

Kurt Weitzmann (1904–1993) was a Byzantine art historian with an impressive oeuvre, a huge influence, and a controversial legacy; he also developed new methodologies that this paper addresses. Weitzmann had graduated in Berlin and left Germany for Princeton in 1935 when the Nazi regime came to power, finding a permanent home there. First as a fellow of the Institute for Advanced Study (IAS), and later as a professor of Medieval and Byzantine art at the Department of Art and Archaeology at Princeton University, he devoted himself to the study of Early Christian and Byzantine manuscript illuminations. He was a prolific author of Eastern and Western medieval art and published on various types of Old and New Testament books, often providing the first scholarly treatment of these precious objects. Through his institutional affiliation with Princeton and close collaborations with other centers of Byzantine scholarship, first of all Dumbarton Oaks and the University of Chicago, he was in a position to leave a deep impact on his students and colleagues and determine the further directions of Byzantine studies.

As a connoisseur with a nineteenth-century German scholarly heritage, Weitzmann applied object-oriented art historical methods, primarily stylistic analysis and iconography. He was aware of the two traditions of modern iconography, the German, launched by Aby Warburg and subsequently reworked by Erwin Panofsky, focusing on symbolic meaning, and the other, established in Princeton by Charles Rufus Morey and utilized mainly for provenance and dating.1 More important is that, in his case, with time, the various practical approaches applied to manuscript illumination distilled into new theoretical principles. Weitzmann’s theory, which he called picture criticism, formed an individual chapter of his major publication, Illustrations in Roll and Codex: A Study of the Origin and Method of Text Illustration (1947).2 The model of picture criticism came from philology: text criticism was elaborated for studying various text versions of the Bible by German and English academics. It was also a working method in the US, especially in New Testament studies at the University of Chicago from the 1930s onward, led by the quest for the ur-text of the Gospels. Weitzmann was in close contact with this project and implemented his method under the strong influence of the Chicago Lectionary Project.3 The method of picture criticism, which was intended to be the most objective and precise method of art history, helped Weitzmann to establish proper chronology and determine the locality of formerly unknown or unpublished manuscripts he came across at various expeditions to Mount Athos and elsewhere. It also assisted him in tracing the genealogy of illustrations from a so-called archetype or prototype, to track imprints from alterations, modifications or substitutions over various copies, and to follow the migration of compositions.4 Weitzmann firmly believed that antique book illustrations, in the form of rolls, had once blossomed as a rich and extensive genre but after a while disappeared because of the ephemeral materials of which these books were made; this belief, however, was based on a very small number of surviving fragments and on Pliny’s references.5 The force behind it was an ambition aimed at deciphering medieval art and tracing it back to antiquity. (Weitzmann [1947] 1970, p. 184). Weitzmann might not have proclaimed it as such, but this is in accordance with the Warburgian program of “the study of the continuation, interruptions and survivals of the classical tradition” (Ginzburg 1989, p. 16). Furthermore, picture criticism offered a grand narrative for manuscript illumination in the age when grand narratives began to be deconstructed.

No matter how grandiose a vision the genealogy of the manuscripts was, it is essential to ascertain that picture criticism is a forceful theoretical construct relying on little evidence, and thus has evoked diverse reactions between praise and serious disapproval. Weitzmann’s ideas were acknowledged and disseminated by his students, first of all Herbert L. Kessler and George Galavaris, and was followed by many others. At the same time, his theory was put to the test and was questioned first by his contemporaries, art historian Hugo Buchtal, Robin Cormack, and others,6 and more recently by Mary-Lyon Dolezal and John Lowden. Yet, studies that have been entirely or largely devoted to Weitzmann’s methodology are scarce. The most important is Dolezal’s paper, Manuscript studies in the twentieth century: Kurt Weitzmann reconsidered (1998), about the evolution of the method. Her main concern is that the method gives “a too restrictive view of the total manuscript” (Dolezal 1998, p. 240) because it lacks the proper context of the historical, religious, and cultural milieu and “focus[es] on hypothetical original sources rather than on the existing Byzantine books themselves.”7 Lowden’s refusal approaches Weitzmann’s ideas from the point of works of art: observing the surviving illuminated Octateuchs, he came to the conclusion that picture criticism is “an unsuitable method” for their study because the sources were more diversified and included monumental cycles (Lowden 1982, p. 125). Even more telling is why Weitzmann’s impact is still palpable: because “art historians have shown a marked reluctance to formulate alternatives” (Dolezal 1998, p. 220, footnote 16). Another aspect of criticism would address aesthetic quality and artists’ autonomy (Walter 1971). Weitzmann took little heed of the actual conditions of artistic production; however, deviation from the pure iconographic source, if that superior original has ever existed, does not necessarily result in works unworthy for study. There are also semi-critical voices with different strategies: ones that emphasize Weitzmann’s seminal role but proving with meticulous care that it was more akin to wishful thinking,8 or those that avoid taking sides in the debate on genealogy, changing the focus of their study or applying distinct methodologies.9 However, the vast quantity of scholarly works referring to Weitzmann’s publications oppose his ideas on a case-by-case basis and elaborate further, refine, or correct the statements he claimed about certain works of art, for example, the wall paintings of Dura Europos. Without a doubt, picture criticism meant a dividing line. Recently, it was indicated that a post-Weitzmann era is a valid concept for Byzantine art history, because it brought novelties in methodology earlier missing from the palette: the function of images in theology, liturgy, and society, the production or reception of images, or the questions of gender and female patronage (Kalavrezou and Tomaselli 2017).

Lesser known is the fact that Weitzmann left a deep impact on the study of pictorial storytelling: not independently from the method of picture criticism, he advanced another method, a narrative analysis, also explicated in the Roll and Codex. The heart of his narrative analysis was an evolutionary classificatory system for storytelling images, the basis of which came from the elusive narrative categories of archaeologist Carl Robert and of Vienna School art historian Franz Wickhoff. Weitzmann’s classification elevated narrative analysis to be, along with stylistic analysis and iconography, an equally important art historical approach to Byzantine illuminated manuscripts. The method had been introduced into the university curriculum in Princeton and to Byzantine art history by Weitzmann; in addition, his taxonomy had wider coverage and was applied by archaeologists and art historians working in the expanded field of world art history.

When Weitzmann came to Princeton, the concept of visual narration was hardly studied on its own right by US scholars and especially not in art history. By the time he died, humanities and social sciences experienced yet another influential movement, the narrative turn, which brought a renewed interest in storytelling across many disciplines.10 If we interpret narrative turn to be “the institutional study of narrative for its own sake, as opposed to examination of individual narratives” as it was suggested in The Routledge Encyclopedia of Narrative Theory (Kreiswirth 2005, pp. 377–78), Weitzmann’s focus on storytelling means that the narrative turn in art history happened earlier than in literary theory.11 This explicit attention on visual narration was unique at a time when narratology as a discipline has not yet evolved in a modern sense, and that makes Weitzmann’s narrative analysis in many aspects analogous with Vladimir Propp’s narrative theory of fairy tales. Weitzmann’s classificatory system played an inevitable role in later taxonomies in Western and Buddhist art and in the disciplines of archaeology, semiotics, aesthetics, and narratology.12 Few scholarly articles have touched upon Weitzmann’s relation to pictorial storytelling; Jacobus Bracker shortly examined the Robert–Wickhoff–Weitzmann line in archaeology (Bracker 2015, pp. 316–26), but most used his ideas as a starting point, like Vicky A. Foskolou’s study on narrative strategies in the monumental painting of the Comnenian and Palaiologan periods of Byzantine art, or Nektarios Zarras’s paper on New Testament cycles in Byzantine churches (Foskolou 2019; Zarras 2016). Understanding the thematic change and the methodological shift Weitzmann’s narrative analysis caused in art history, notably in Byzantine and medieval art history, means recognizing the strength and influence of the method he proposed.

The focus of this paper is Weitzmann and his method of narrative analysis: its personal and institutional background, the way it was formulated, its context, dissemination, and legacy. I claim that Weitzmann’s narrative analysis was developed through his pedagogical work and lecturing and that it had a key influence on the style of his autobiographical writings. Regarding the broader context, in the historiography of narrative research, Weitzmann had a tripartite role: first, he formed an intellectual connection between the early generation of German-speaking art historians interested in visual storytelling and the next generation of scholars of visual narration in the US, many of whom were educated by him and associated with Princeton University and Byzantine studies; second, the evolutionary model of typology Weitzmann proposed in 1947 constituted the common ground for all succeeding narrative taxonomies and, in this sense, he was able to dictate the way visual narratives have been approached in recent decades; and third, the narrative classification he borrowed from archaeology and redesigned for Byzantine art, together with his narrative analysis, served as a method of transmission between Eastern and Western art.

This paper relies to some extent on the accounts of Weitzmann’s professional life, which is well documented both in written and material forms—although critical voices suggest it is also important to note what is missing from these accounts (Dolezal 1998, p. 230, footnote 45). Weitzmann’s posthumously published long memoir, Sailing with Byzantium from Europe to America: The Memoirs of an Art Historian (1994), recounts the events of his professional life, as do, in a shorter form, the two interviews he recorded, one in 1983 for the Rutgers Art Review and the other in 1992–93 for the Getty Oral History Program (Weitzmann 1984, 1994b). For the material part, Princeton University’s Firestone Library hosts the Kurt Weitzmann Archive, which contains his correspondence, notes, lectures, and other manuscripts,13 and Dumbarton Oaks holds a relatively small Kurt Weitzmann Papers collection.14

2. The Development of Weitzmann’s Narrative Analysis

2.1. Institutional Background and Early Influences behind the Narrative Method

The interest in narrative genres, especially in illuminated manuscripts and frescoes of the medieval East and West had been present from the very beginning of Weitzmann’s career and had a decisive role in turning to the theory of narration. Trained as a general art historian, his first paper on the lavishly illuminated Paris Psalter in 192915 and his first book, Die byzantinische Buchmalerei des 9–10. Jahrhunderts, published in Germany in 1935 were meant to be serious entrées into the field of Byzantine studies. Because of the large number of objects and the wide range of topics they presented, Weitzmann regarded manuscript illumination as “the most complex material in Byzantine art” (Weitzmann 1994a, p. 528), but the key to the choice of illustration probably lay in the idea of narrative, since “the art of storytelling in pictures became inextricably linked with the history of book-illumination” (Weitzmann 1957, p. 83). When Weitzmann was invited to Princeton to contribute to the main project of the Department, to publish the corpus of the illustrated Septuagint manuscripts initiated by Morey and to write a volume on Greek manuscripts, the Octateuchs, he was ready to engage in larger projects (Weitzmann 1994b, p. 176).

Colleagues, historical circumstances, and the specific milieu of Princeton had equally vital consequences on Weitzmann’s methodology. Two colleagues at the department had a crucial influence: Albert M. Friend broadened his methodology,16 and Morey’s research in iconography affirmed the choice of book illustration because Morey assumed that this genre was fundamental in developing Christian iconography (Dolezal 1998, pp. 230–31; Weitzmann 1947, p. 411). Erwin Panofsky, who came to Princeton a few years earlier and at that time was already a permanent professor at the newly founded IAS, was delighted by the working conditions and research facilities Princeton offered, discovering that “certain aspects of mediaeval painting and book illumination could be more exhaustively studied in this country than in Europe (…) owing to a series of historical accidents” (Panofsky 1954, p. 12). The beneficial outcomes of these historical accidents for Weitzmann must have included an opportunity to travel and participate in excavations, the Manuscript Room, a special research and library room for art books and photographs established by Friend at the department, and the Index of Christian Art,17 an influential force behind his narrative classifications. The Index was founded by Morey in 1917 and was intended to be an exhaustive compilation of the images of medieval works of art, a standard thematic archive that classifies images according to their iconographic type.18 Weitzmann himself contributed to the Index: his first major project was an expedition to Mount Athos in 1936 that enriched the collection with more than two thousand images. He surely realized that classification functions well for handling large amounts of newly discovered works, mostly ivories, icons, and manuscripts, which needed a first-hand approach to be described, analyzed, and integrated into the total body of works that constituted art history. Weitzmann’s theoretical purpose in Roll and Codex was to elaborate one more such classification, a narrative one.

Apart from Princeton, in these early years, Dumbarton Oaks was another institute which offered Weitzmann a platform to develop and test his ideas on narratives; their cooperation began even before the institution was officially inaugurated as a research center in 1940. Dumbarton Oaks accommodated a yearly symposium aiming at bringing together scholars of Byzantine and Early Christian art. Part of these symposia, in 1938, Weitzmann lectured on the Principles of Byzantine Book Illumination; in 1943, he gave a series of four lectures in a day covering four themes, The Roll Illustrations in the Hellenistic and Roman Period, The Transformation of Roll into Codex, and two others with special emphasis on the relation of images and texts in manuscripts, the ideas formulated in the Roll and Codex; in 1945, he gave lectures on the iconography of Dura Europos Synagogue’s fragmented wall paintings, supplementing the theologian Carl Kraeling’s approaches. He also helped build up the collection and the library and contributed to developing it into a center for Byzantine art.19

2.2. Narrative Analysis Developed through Teaching and Lecturing

For Weitzmann, teaching and lecturing happened to be the most powerful catalysts for developing and disseminating his ideas on pictorial storytelling and publishing them in a written form in the Roll and Codex in 1947. Narrative illustrations were fundamental topics in both courses he was teaching at Princeton.20 First was the Manuscript (MSS) seminar, a graduate course on illustrated manuscripts initiated back in 1931 by Morey.21 Weitzmann was affiliated with the IAS that time, but Friend invited him as early as 1935 to co-teach the course. Each covered half: Friend talked about portraits of the Evangelists, while Weitzmann spoke about narrative book illustrations. The topic on narrative was suggested by Friend, whose special interest lay in the early development of manuscript illuminations. This joint seminar usually took place in the spring term but was suspended for the years between 1943 and 1945 because of the war; by the time it restarted in 1946, Friend had already begun his long collaboration with Dumbarton Oaks, and the seminar was run solely by Weitzmann covering all fourteen lectures.22 Princeton was well known for the so-called preceptorial system, according to which lectures were accompanied by small-group seminars for discussions; Weitzmann eliminated testing and instead introduced essay-writing; over time, the fourteen-lecture semesters turned into ten-lecture courses, giving time for students’ presentations.

The second was the Medieval course, a comprehensive undergraduate core course he began teaching in 1945 (Weitzmann 1994a, p. 59; 1984, p. 74). Between 1935 and 1945, Weitzmann was a full-time member of the IAS and was also teaching at the Department with Friend. He took over this Medieval course from Morey when he retired in 1945 and Weitzmann was appointed as research associate in art and archaeology. This associate position was turned into full professorship in 1950, lasting until his 1972 retirement, but he ran the MSS seminar for two more years.23 He also kept his half-time membership at the IAS. Weitzmann’s life was regulated by these different duties: teaching in spring semesters and research and travel in fall semesters. Relying on the proximity and the collection of the Manuscript Room, the MSS and the Medieval course offered visual material on book illustration not available anywhere else at that time.

Weitzmann’s teaching was new in methodology, and it was new thematically as well. He introduced current ideas on visual narratives into his teaching curriculum and organized these courses around specific narrative-related topics and certain types of narratives. For example, the MSS course, apart from the discussion on text and picture criticism, stylistic criticism, and text-image relations, discussed narrative-related questions in individual lectures; these were the ancient or classical ‘system of illustrations’ in roll and in codex, the narrative (and liturgical) gospel cycles, and the most influential of all, the continuous frieze.24 Elements of the later typology, like the ‘monocyclic and polycyclic images’, were also distinguished. In these courses, studying the complex structures of narrative picture cycles, such as the Dura Europos frescoes, and manuscripts, such as the Paris Psalter and the Joshua Roll, or the various narrative cycles on the Bible, were essential. Topics related to narration were integral parts of the course, Weitzmann kept them in the curriculum for decades. Wickhoff’s Wiener Genesis, on which he built his narrative taxonomy, and some contemporary writings on narrative images were recurring items on the list of assigned readings. Unlike many art historical seminars that gave out essay questions for students focusing on one particular work of art, Weitzmann asked for broader, narrative-related topics: written tasks included ‘the history of the illustration of the Old Testament’, ‘the various modes of representation, the symbolic, narrative, dogmatic and liturgical’, or ‘the role of illuminated manuscripts in medieval art’.

It was during these years at Princeton when Weitzman met students who were interested and got involved in the study of visual narration during their career; he was either teaching them a course or supervising their theses. Among them the most important were Richard Brilliant in 1954,25 George Galavaris in 1955, Thomas Hoving and Ihor Ševčenko in 1956, Charles Dempsey in 1960, Herbert L. Kessler in 1962, Shigebumi Tsuji in 1962, Archer St. Claire in 1966, and Gary Vikan in 1970.26 Brilliant, Kessler, Tsuji, and others formed the next generation of scholars writing about pictorial storytelling in the Byzantine context.

2.3. Publishing the Method in Roll and Codex: Narrative Analysis for Book Illustration

From Weitzmann’s lecture notes, it is evident that he not only incorporated the topic of visual narration into the curriculum but treated it as equally important compared to other methods of art history, which was a novelty. An analysis of the materials studied in the MSS and Medieval courses were built on historical evidence and were primarily approached through the traditional visual methods developed for art history, stylistic analysis and iconography. These approaches, however, could not describe the specific characteristics of narrative cycles or narrative book illustration. Weitzmann regarded objects as unique and full entities, and following Adolph Goldschmidt, he aspired to reach a comprehensive approach: “His [Goldschmidt’s] method was simply to make you look at a work of art and make you think of every possible angle. This means this was neither purely stylistic nor purely iconographical. Every aspect had to be taken into account, also the historical aspect. Everything.” (Weitzmann 1994b, p. 23) The aspect he added was narrative. All these ideas were reflected in Roll and Codex in which Weitzmann introduced his theoretical observations on narrative objects in the first chapter, The General Relation Between Literature and the Representational Arts. This book came into being because of the Manuscript seminar taught at Princeton; it was used as a general handbook for the course, was translated to other languages, and attracted students to Princeton, including Galavaris and Tsuji.

From the standpoint of visual narratives, Weitzmann’s main achievement in this book was the dual approach: formulation of an evolutionary typology for narrative objects, which standardized the vocabulary, and a method of interpretation based on narrative structures and text-image comparisons. The typology classified storytelling images by their visual structure against the Aristotelian time-space coordinates, allocated the types to consecutive artistic periods, and delineated an alternative history of narrative imagery. The first and earliest type is the simultaneous method, which describes several actions in one single scene without repeating the characters (before the 5th century BC). The monoscenic method was predominant during the 5th and 4th centuries BC, which displayed only one single action to which all the poses and gestures were related. The cyclic method was invented to define the characteristics of the Hellenistic period (4th to 2nd centuries BC), where the scenes show consecutive but individual scenes, each repeating the characters (Weitzmann [1947] 1970, pp. 12–46, especially pp. 13–18). It is here that Weitzmann introduced polyscenic images as opposed to monoscenic images and created a subcategory for summarizing, epitomized cycles. Weitzmann eventually did not invent the idea of narrative typology but reworked two rather rudimentary late 19th-century typologies that were mostly known to a relatively small community of German-speaking art historians: Carl Robert introduced three narrative types for ancient Greek art in Bild und Lied in 1881, which was developed further and applied to Late Roman and Early Christian art by Franz Wickhoff in Die Wiener Genesis in 1895.27 Weitzmann regarded his own classification closer to that of Carl Robert’s and criticized more of Wickhoff’s, but the latter helped him reach his main focus, Byzantine art.

The real strength of Weitzmann’s narrative taxonomy lays in the fact that it was elaborated consistently enough to become a practical method which worked well on a large number of unclassified images. Weitzmann standardized and expanded the formerly existing typologies and popularized the new system to a larger English-speaking scholarly community in the US. This opened up a pathway for a long chain of further clarifications and applications that were all rooted in Weitzmann’s typology: this meant numerous attempts in archaeology, for example, by Paul Meyboom (1978) who proposed a revised system of Weitzmann’s typology, or Stansbury-O’Donnell (1999) who completed the table along the line of Aristotelian expectations; in other areas of the discipline, such as in medieval, Renaissance and early modern art by Marilyn Aronberg Lavin (1990) who, based on a huge database, identified numerous spatial arrangements in Italian narrative frescoes, or Vidya Dehejia (1990), who expanded Weitzmann’s taxonomy for the study of Buddhist art; and in other fields, like narratology and semiotics. Indeed, these additions, redefinitions, and clarifications are still being formed even today.28

While narrative classifications were elaborated to deal with huge corpora of images rather routinely and impersonally, Weitzmann’s narrative analysis, looking from the perspective of both the analyzer and the work of art, was a more individual method for studying storytelling images. The analysis included various phases, depending on the relevancy for that particular (group of) artworks: organizing images according to narrative types, investigating the changes of narrative modes, locating geographical characteristics of narration, recognizing period style in storytelling, comparing the plot in images and texts, or analyzing text and image relations. In practice, these common methods can be seen in a mature form, for example, in the individual descriptions of Weitzmann’s Late Antique and Early Christian Book Illumination (1977): each explanation aims to establish the literary source, identify all the narrative episodes and characters in the depicted scenes, match the plot in the written and graphic form, analyze the narrative solution compared to other narrative, compositional, and structural models, and emphasize the distinctive narrative features of that particular work. The focus of this comparative text-image analysis enlightens the core question of illustration in Weitzmann’s perception: to what extent has the proper visualization of the literary content been fulfilled, and are there any additions, omissions or modifications?29

3. The Context, Dissemination, and Impact of Weitzmann’s Narrative Analysis

3.1. Specificities of Weitzmann’s Narrative Method

Weitzmann’s typology bridged the differences in subject matter of Robert’s and Wickhoff’s classifications and developed into a method that worked well for Eastern Byzantine and Latin Western art. Robert studied the narrative structures of the representations of Greek mythological stories, while Wickhoff was interested in the transition between Roman and Early Christian art and was not afraid to refer to examples from Egyptian to Renaissance art to justify the use of continuous narration (the method repeating the dramatis personae) (Wickhoff 1895, pp. 8–10), which operated quite differently from the so-called pregnant moment prescribed by the 18th-century aesthetic taste. Weitzmann returned to Greek examples and worked the way up systematically through Roman and Early Christian periods to the art of Byzantium. His typology helped him classify the masses of newfound book illustrations, which found their literary sources most often in the Bible, as, for example, in the Book of Genesis or the Gospels. It also meant that by the time narrative art began to flourish in the Byzantine empire, all these ancient Greek and Roman narrative methods were available for use as models and ready to be blended. The afterlife of the Roll and Codex has clearly shown that not only did the idea of narrative classifications travel with Weitzmann from the early German scholarship to the West and spread worldwide, but the subjects of the method themselves followed a similar path: from ancient Greece to Byzantium and to the art of the Latin West, up to the point when they became a world art phenomenon.

Weitzmann was judgmental: the evolutionary typology he created was hierarchic in order to validate the superb nature of Byzantine book illumination. Robert’s typology was chronological; the evolutionary progression was inherited directly from Wickhoff, where the continuous method was the most appropriate one perfected by Roman art (Horváth 2016, pp. 253–54). Evolutionism in art history has been present for a long time, from Pliny the Elder through Giorgio Vasari to Hegel and Winckelmann, but Darwinism gave it a new comeback in the late 19th century when it supported theories of stylistic change.30 Weitzmann not only claimed that the learning process in narration was evolutionary, but asserted that the progression was organic and culminated with the cyclic method, in which the Aristotelian unity of time, space, and action was completed. The simultaneous method had shortcomings and was regarded as ‘primitive’, and the monoscenic method had ‘limitations’, but the cyclic method was ‘genuine’ (Weitzmann [1947] 1970, pp. 15, 17, 22). Obviously, the greatest quality was assigned to Weitzmann’s favorite works, “the extensive narrative cycles of the Bible, because it was here that Byzantine book illumination excelled. It has been pointed out that the earliest cycles [the Cotton Genesis or the Vienna Genesis] are the richest.” (Weitzmann 1975, p. 50) The value-oriented view of narration was definitely the heritage of the Vienna School, but this judgmental aspect has disappeared from subsequent classifications.

For Weitzmann, the narrative image was, in its most perfected form, an illustrated text. He was quite conservative in suggesting that narrative illustration is a medium of derivation and is subordinate to textual meaning: what makes picture cycles better is the faithfulness to the textual source (Weitzmann [1947] 1970, p. 185). Here, text is the norm. The act of scanning pictures is consistently designated as ‘reading’ (Weitzmann [1947] 1970, pp. 17–19), and, as we often see in narratology, the communicational capacities of images are expressed through linguistic terms.31 The text-illustration relationship is also modeled as being strict, hierarchic, and value-based. The narrative illustrations of the archaic, classical, and Hellenistic periods were ‘imperfect’ in their use of textual sources, but with time artists turned away from oral traditions and referred more precisely to the literary sources; this improvement reached perfection, just as above, in Early Christian illustrations. According to this theory, this was a “slow, evolutionary process” (Weitzmann 1957, p. 83, my italics), in which any departure from the written source was seen as “(conscious) deviation” (Weitzmann [1947] 1970, pp. 15, 183), an “iconographic negligence and corruption” (Weitzmann [1947] 1970, p. 11), or “an error in pictures” (Weitzmann [1947] 1970, p. 184). This approach differs strikingly from the evaluation of art in the Latin West, where such alterations in the Renaissance were seen as signs of creativity or innovation.32 On rare occasions, Weitzmann steps out of the context of medieval art and acknowledges those qualities of visual storytelling that exist beyond the support of a textual source: he refers to the “cinematographic character” of narrative picture cycles in connection with the Vienna Genesis or the four Gospels, and recognizes these values as being antecedents to contemporary film or comic strip (Cf. Weitzmann 1975, p. 3; 1971, p. 250; 1977, p. 10).

The fundamental steps of Weitzmann’s narrative analysis have remained valid up until today, and some art historians with more traditional attitudes still continue to follow his methods closely and rather enthusiastically. However, the concept of narrative analysis began to change in the 1990s, when more complex and visual-oriented text and image relations, introduced by Critical Inquiry and W.J.T. Mitchell at the University of Chicago in the 1980s, began to take effect; the scope of narrative works were expanded to include products of visual culture, and narratology, especially transmedial narratology in the 2000s, began to offer noticeably different approaches.33

3.2. The Broader Context of Weitzmann’s Narrative Analysis: Iconology

The focus on classification and storytelling relates Weitzmann’s investigations to two powerful contemporary theories in which narratives and typologies also play a fundamental role, Erwin Panofsky’ iconology and Vladimir Propp’s morphology.

Juxtaposing Weitzmann with Panofsky seems obvious because of the parallels of their lives; nonetheless, they shared a similar line of thinking about the need to examine narrative structures. With a few years’ difference, Weitzmann and Panofsky attended Goldschmidt’s medieval seminars in Berlin and were deeply attached to the German tradition of art history writing. They both had to restart their carriers in the US as “transplanted Europeans” (Panofsky 1954, p. 7) and integrate themselves into a rather different academic culture. As colleagues at the IAS, they were much aware of each other’s work, but they carefully avoided interfering in the other’s intellectual circle. The distance can be well captured in Weitzmann’s memoir, where Weitzmann tells endless stories about people he knew, but has no story to share about Panofsky. There is only a brief reference to their working relationship, a telling remark about the Tolnay case, and a short note about an unusual case of a ‘divided loyalty,’ a student who studied both with Weitzmann in the Manuscript Room, and later with Panofsky (Weitzmann 1994a, pp. 87–88, 93, 467), respectively. The parallel is present in their approach: they both picked up a highly intellectual method from the early German scholarship and, while being aware of the new context, developed it into a more practical art historical method and then introduced it into a wider, English-speaking scholarly community.

Panofsky’s iconology, though wider in scope and influence as compared to Weitzmann’s, was also developed to be applied to narrative art.34 Panofsky’s famous essay, the Iconography and Iconology: An Introduction to the Study of Renaissance Art (1939), which mostly concerned fifteenth-century Northern Renaissance painting, offered a three-step method for describing and understanding the subject matter and meaning of works of art; the second level, the iconographical analysis, is where meaning is decodable through one’s deep knowledge of literature, “whether acquired by purposeful reading or by oral tradition” (Panofsky [1939] 1955, p. 35). Because of the special treatment of narrativity in the visual, the two gave detailed attention to literary sources and regarded images as objectified reformulations of certain literary texts. Eventually, Panofsky and Weitzmann shared the belief that the ultimate meaning of an image is an external idea lying beyond the realm of the work of art. In iconology, it is the intrinsic meaning of the individual work of art: it is “essential, unconscious, and accessible only to subjective understanding” (Hasenmueller 1978, p. 291) and needs to be recreated because of the lost original context. In Weitzmann, the meaning derives from the way an illustration fits into a long chain that leads back to the original and often lost prototype. Whether the meaning is obtained through subjective interpretation, as Panofsky thought, or through a more rational method used by Weitzmann, in both cases, the supporting theory was more of a construction than a reconstruction.

Similarities appear with regard to taxonomies, too: the method of iconography also created a classificatory system that relied on the history of types. These types form the nucleus of an enquiry into the specific problem of identification and the distinction of certain thematic groups and their history. The idea appeared in Panofsky’s 1931 Kiel lecture, but in the revised and published 1939 version, he gave a more precise definition framing the thematic-based typology (Cf. Panofsky [1939] 1955, pp. 38–39). Here, the history of type means an “insight into the manner in which, under varying historical conditions, specific themes of concepts were expressed by objects and events” (Panofsky [1939] 1955, p. 41). The actual iconographical typology was not elaborated further in his methodology, yet the classification was clarified through such famous examples as the iconographic types of Judith as opposed to Salome.

Yet, their influence was different. Beyond the different scope of their theories, their dissimilar personalities also played a role. A neat remark from Gary Vikan, Weitzmann’s former student, reveals his teacher’s character: he “was not the elegant intellect that Panofsky was, but rather an almost compulsively methodical archaeologist of iconography, modelling his approach on that of 19th-century text critics” (Vikan 2016, p. 24). Panofsky’s iconology was articulated as an independent and compact theory and had wider relevance; in the subsequent decades, it became one of the most fundamental methods of general art history. Weitzmann’s narrative analysis was not intended to become a comprehensive theory for art; it was more of a theory in the making, which attained validity through proper examples and served as primary classification. While Panofsky had become the “praeceptor Americae, at least as far as art history was concerned” (Van de Waal 1972, p. 229), Weitzmann influenced a rather isolated circle of Byzantologists and medievalists and only after a while did this include those art historians and archaeologists who engaged with pictorial storytelling.

3.3. The Broader Context of Weitzmann’s Narrative Analysis: Narratology

Weitzmann’s life did not come across with Vladimir Propp as with Panofsky, but the theoretical investigations in the Roll and Codex and in Propp’s Morphology of the Folktale (1928) strongly correlate in many aspects: both focused on narrative forms and structures, developed a fundamental terminology and a classificatory system that intended to serve as a first stage of narrative analysis, and both connected their work to biology. Propp’s typology was elaborated on a corpus of Russian folk tales; he identified seven character types (for example hero, villain, or false hero) and thirty-one types of actions, or functions as he called it (like interdiction, violation, interrogation, etc.). This scheme is independent of the singular qualities of the characters but observes the role they occupy in the structure of the story. It is a point of connection with Weitzmann, whose classification examines the characters of a story in the time-space coordinates of the pictorial space separately from the plot. By means of these components, they developed a rather impersonal, formalistic structural analysis that proved to be replicable.

Similarly to Weitzmann’s analysis, Propp’s seminal theory was not just replicable, but it was also transferable to other genres; narratologists and literary scholars referred to it as a grammar or a syntax of a special genre that “offered a model for the establishment of other grammars (…) the system he devised revealed to the next generation some universals of narrative grammar.”35 All this came only after Propp’s Morphology was translated into English in 1958 and generated a debate with Claude Lévi-Strauss and other French structuralists. Structuralism and post-structuralism often reached out to art: Gilles Deleuze was engaged with Francis Bacon, Michel Foucault with Velázquez, Julia Kristeva with Hans Holbein. However, at the time when Weitzmann described his narrative analysis in the 1940s, researching narrative aspects of works of art was not a subject for any existing approach in art history, and there was certainly no method for its study ready for common use. Nor was this yet an established method in literature or in other disciplines in the humanities before French structuralism, from which narratology sprang out in literary theory, was formulated (see Onega and Landa 1996, pp. 12–34). Propp quickly earned a position of a prestructuralist and was seen as a predecessor in the historical evolution of narrative studies. His theory indeed had a great impact on anthropology, philosophy, and linguistics, but it did not have that great of an impact on art history. This is mostly because formalism, which involves stylistic, structural, and compositional analysis, was inherent to art history since the late 19th century, the period from which the first classification for narrative images, which Weitzmann refined, originated. It was not before the 1980s that narratology began to filter through more traditional methods in art history.

When Propp turned to biology, he emphasized how the structural approach to narration is natural and self-evident, just as Weitzmann did when referring to evolution. The title of the book, Morphology, already calls attention to this phenomenon and also draws a parallel with botany (Propp 2009, p. xxv). Plots are compared to organic forms, living things, and organisms: “The sequence of events has its own laws. The short story too has similar laws, as do organic formations” (Propp 2009, pp. 22, 78). When explaining the need for classifications, Propp turns again to sciences for validation and accentuates that classification is the “first and most important step of study”, just as in sciences and in Weitzmann’s analysis (Propp 2009, pp. 4, 11). The idea that narratives own an organic nature and are like living organisms is now widespread, but this is due to Roland Barthes’s oft-quoted opening passage of the Introduction to the Structural Analysis of Narratives, in which he declared that narrative “is simply there, like life itself.”36 While Propp probably drew this idea from the fact that stories replicate, vary, and spread, Weitzmann relied on the Vasarian–Hegelian concept that had long been present in art history. Yet, they both followed empirical methods and analyzed a large number of specimens, as biologists would do. They also disregarded the social, religious, or cultural context of the stories because they aspired to follow a ‘pure,’ structural approach. It is the unprecedented importance they gave to the phenomenon of storytelling that made them pioneers in their field.

3.4. The Dissemination of Weitzmann’s Theory of Narratives

Before the 1980s, when the interest significantly raised toward the topic of narrative in the US, the activities in visual narrative research concentrated around three institutions: the Department of Art and Archaeology at Princeton, where Weitzmann held a chair and was teaching from 1935 up until the mid-1970s; Dumbarton Oaks, where he was a frequent visitor and played an indispensable role in the formation of the Institute; and the University of Chicago, with which Weitzmann was also closely associated through the Lectionary Project. In other words, Weitzmann propagated the study of visual narratives directly to his students through his almost four-decade-long teaching career, to his colleagues through an effective personal influence, and to a broader group of art historians with his extremely productive oeuvre. As his personal and professional correspondence reveals, he made a huge effort to establish and expand a professional community, which actually meant a relevant audience for his ideas. This was an active, responsive, and critical audience consisting almost exclusively of male voices, a network of colleagues, fellow art historians, other German refugees, and most importantly, his former students.

One early Princeton case of impact is Weitzmann’s colleague, Otto Pächt, a Vienna School art historian whose explicit interest in pictorial storytelling appeared during his stay at IAS in 1956–57. The following year, Pächt and Weitzmann gave a joint lecture in Stockholm. Pächt’s research was published as The Rise of Pictorial Narrative in Twelfth-Century England (1962), the theoretical part of which relies exclusively on Weitzmann’s ideas in the Roll and Codex. Nonetheless, Pächt attributed even less capacity to images than Weitzmann (Pächt 1962, p. 1).

The archaeologist and theologian Carl Kraeling was fascinated by Weitzmann’s investigations into continuous friezes in The Joshua Roll (Weitzmann 1948), because it gave him an exemplary method to treat the Dura Europos Synagogue’s narrative murals,37 a shared interest of the two that resulted in several distinguished publications. At that time, Kraeling served as a professor of New Testament studies at Yale University, and in 1950 he became the director of the Oriental Institute at the University of Chicago, the institution that hosted the first ever conference on visual narration in 1955. The topic grew out of Kraeling’s strong interest in the method of continuous narration. At the early phase of organizing, it was referred to as a symposium on “continuous method,”38 a title later changed to Narration in Ancient Art.

The main approach was art historical with a special emphasis on storytelling: themes in narrative art were expanded to cover the central geographical areas of ancient art and early Christianity in the Mediterraneum.39 Helen J. Kantor was talking about the development of narrative representation in reliefs and illustrated papyri in Egyptian culture; Ann Perkins aimed to set up the history of narrative art in Babylonia by analyzing the major narrative works; Hans G. Güterbock offered a comparative analysis of narrative art across the three regions of Anatolia, Syria, and Assyria; George Hanfmann wrote about generalizations in the representation of actions and gestures that artists in ancient Greece used to make their story recognizable; and Peter H. von Blanckenhagen studied the method of continuous narrative in the context of Hellenistic and Roman art, analyzing friezes and other narrative objects, including Trajan’s Column. As the last speaker, Weitzmann addressed the relationship of literature and its pictorial representation, the perfection of which relation he saw also as an evolutionary process in the Early Christian context (Weitzmann 1957, pp. 83–91). The proceedings were all published in a special issue of the American Journal of Archaeology in 1957.40 This conference publication was the first systematic and comprehensive study of artworks conveying historical and mythological narratives across the Mediterranean region and has had an enormous influence on further research. The six papers addressed almost all the topics that in the following decades emerged concerning visual narration in general: the relationship of the pictorial story to its literary source, evolution of narrative types, regional and cultural differences in visual storytelling, the comprehension of pictorial stories, and the role of the beholder.

The fact that the majority of the speakers, such as Blanckenhagen, Güterbock, Hanfmann and Weitzmann, were German refugees meant two things. For Weitzmann, it “showed clearly the impact of the German emigration to the United States in the pre- and post-war years” (Weitzmann 1994a, p. 212). It also revealed that in the mid-1950s, visual narration still primarily attracted scholars for whom the early German scholarship, in which the topic of narrative art first appeared in the discipline of art history, was rooted in personal experience or personal heritage. Weitzmann’s crucial achievement in the area of narrative studies is that he was able to incorporate early German art historical tradition into modern scholarship and opened up the gates to this new knowledge to spread all over the US. Weitzmann, along with Otto Pächt and Carl Kraeling, and the participants of the 1955 Chicago conference formed a group that we may call the first generation of US scholars involved in the research on visual narration.

3.5. The Afterlife of Roll and Codex: The Second Generation of Scholars in Narrative Research

The second generation of art historians examining pictorial storytelling undeniably emerged from their seeds: most of them were students of Weitzmann and were once associated with the Department of Art and Archaeology at Princeton University. The method introduced in the Roll and Codex soon became a common approach in art history for analyzing the relationship between narrative illustrations and their literary source: first for Byzantine and medieval manuscripts, then in many other genres of visual culture. Going global could not have happened without the help of Weitzmann’s devoted pupils who took up the topic of visual narration and advanced it not only in different universities and research institutions across the US, but in lands as far as Japan. The impact of this second generation, most importantly that of Herbert L. Kessler and Richard Brilliant, can be apparently felt from around the 1980s when they started off their careers.

Weitzmann still cultivated excellent connections with Dumbarton Oaks scholars, remaining a frequent visitor at the Institute for many decades and promoting research on narratives: in 1971, his lecture at Dumbarton Oaks addressed a special type of visual narratives, the cyclic narration.41 In exchange, Weitzmann had an unvoiced privilege each year of naming one student as a Junior Fellow, which included many interested in visual narration, like Kessler, Gary Vikan, Shigebumi Tsuji, or Hans Belting. For them, Dumbarton Oaks meant a secure ground for launching their career. The University of Chicago was the other institution where Weitzmann’s name circulated for these decades due to his connection to Kraeling and the Lectionary Project. In 1961, Weitzmann was offered a professorship in which a former student, Joshua Taylor, then a professor at Chicago, had an initiative role.42 In this rather piquant story, Weitzmann used this offer to secure his own position with Princeton, reduce his teaching hours and ‘adjust’ his salary. By the time in 1968 when Chicago offered Weitzmann an honorary degree, he already had another enthusiastic supporter there, Herbert L. Kessler.

Kessler was one of Weitzmann’s favorite students and his most important heir in the study of visual narratives. He earned his doctorate under Weitzmann on Bible illustrations and returned to his hometown, Chicago, in 1965 as an assistant professor. Their shared interest in pictorial storytelling certainly played a role in that Kessler became Weitzmann’s protégé; later, they developed a deep friendship. For Kessler, this topic has meant a life-long engagement. His own volume, Studies in Pictorial Narrative (1994), is dedicated to Weitzmann. It is a collection of essays on various topics of medieval art, some already well-known, such as the Cotton Genesis or the Dura Europos frescoes, some surprisingly new in approach, such as the one on the utilization of pictorial narratives in pastoral pedagogy (Kessler 1994).

Another student was Richard Brilliant, who attended Weitzmann’s course during his Yale year in 1954–55. Brilliant’s book, Visual Narratives: Storytelling in Etruscan and Roman Art (Brilliant 1984), focuses on four distinct narrative art forms: Etruscan urns, Iliac tablets and narrative frescoes, the Column of Trajan, and mythological sarcophagi. Addressing the problem of conveying and acquiring knowledge through the visual, Brilliant placed narration in a communicative context, where the artist-narrator deploys special skills, gestures, for example, to make the story comprehensible. Unlike Weitzmann’s, Brilliant’s approach gives visual and verbal forms of storytelling equal importance. His paper on the Bayeux Tapestry, in the presentation of which an oral tradition was envisioned, was also dedicated to Weitzmann (Brilliant 1991).

The Roll and Codex established bidirectional connections: it attracted students from afar and promoted the idea of narrative illustration in remote countries where such approaches were not part of the vernacular scholarly vocabulary. This was the case with George Galavaris, who came to Princeton from Greece, attended the famous Manuscript seminar in 1955, then wrote a thesis with Weitzmann on the illustrations of Gregory of Nazianzus’s Homilies. Afterwards, Galavaris became Weitzmann’s assistant and a collaborator of the volumes on the Sinai and Athos manuscripts.

A similar case is that of Shigebumi Tsuji, who arrived from Tokyo to study book illustrations but widened his interest toward the not-so-obvious topic of narrative landscapes, a legitimate theme in the mirror of Japan’s narrative art traditions. Weitzmann opened Tsuji’s path to Dumbarton Oaks. In the summer of 1980, Tsuji organized a symposium in Osaka on narrative book illustration in honor of Weitzmann, who was supposed to give the inaugural lecture. Although a great admirer of Far Eastern art, he considered this trip to be too risky for his age.43 Most papers of the symposium were published in the Studia Artium Orientalis et Occidentalis in 1982 and 1985 (Kimura 1982, 1985). The Osaka symposium focused on the relation of texts and their illustrations and wanted to shed light on “illustrations based on less established, less standardized texts” (Tsuji 1982, p. v). The reason for this was the different state of canonization of texts and the different attitudes in the adherence to these texts in Eastern and Western cultures.

Few years later, in 1984 in Baltimore, Marianna Shreve Simpson and Kessler organized another scholarly event on visual storytelling, which indicated the growing interest toward visual narration. The papers appeared in a separate volume (Pictorial Narrative in Antiquity and the Middle Ages, 1985) (Kessler and Shreve Simpson 1985), constituting the third large collection of thematic essays on the topic of pictorial storytelling. Following the endeavor initiated by Kraeling in the 1950s, to accommodate as wide range of narratives as possible, this collection was even more ambitious and covered artworks from the Far East, Near East, Egypt, Byzantium, and medieval Europe. There was also a crucial change in attitude: narrative pictures were examined not as subordinate to their textual sources but as independent vehicles of knowledge, and the emphasis was given to the complex networks of narrative works.

Hans Belting, as a young Byzantine scholar, also investigated medieval storytelling. Belting did not belong to Weitzmann’s school but got in touch with him through Dumbarton Oaks, which he repeatedly visited from the late 1950s. He cultivated a close working relationship with Weitzmann, was able to integrate himself very well into the circle of Byzantinists in the US, and his non-standard research questions were appreciated by this community.44 Belting actively participated in both the Osaka and Baltimore symposia and was, to my knowledge, the first scholar who directly addressed narrative in Italian Trecento painting by focusing on the change in storytelling that began in Giotto’s era (Belting 1980, 1985).

While Princeton, Dumbarton Oaks, and the University of Chicago remained important centers for the research on narratives, since Kessler took up a position in 1976 at the Department of the History of Art at John Hopkins University, Baltimore also began to show a certain cumulation of the second generation scholars of pictorial storytelling. Charles Dempsey, who was invited there by Kessler in 1980, had twenty years earlier attended Weitzmann’s Manuscript course twice, because of the methodological investigations concerning the narrative work and its textual sources, which were utilized in his studies on Renaissance and Baroque art theory. Gary Vikan, also from the Weitzmann school, participated in the Osaka symposium and was attracted by Byzantine narrative art, but changed his mind and pursued a museum career; his recently published memoir describes his early years at Princeton.45 Professors Henry Maguire and H. Alan Shapiro, also from Baltimore, have contributed to the study of narratives in Byzantine and in ancient Greek art, respectively, but they were not strictly associated with the Weitzmann school.

Weitzmann had a dynamic relation with his students and owed them a lot. The long pages devoted to them in the Sailing reveal several microstories about the often close relationship between a successful professor and his students; importantly, only those were mentioned whom he found worthy of mentioning. The very few who matched Weitzmann’s expectations, like Kessler or Galavaris, were raised into the honorable position of a collaborator, the tradition Weitzmann inherited from Goldschmidt (Weitzmann 1994b, p. 176). Galavaris remembers this practice of selection concerning students “who were destined to be his future collaborators” (Galavaris 1994, p. 13). Vikan recalls their relationship from the other side and describes how they, “little Weitzmanns in the making”, idolized him (Vikan 2016, pp. 23–26). There were a few students who had an enduring interest in narratives and turned out to be great storytellers themselves, such as Thomas Hoving, the former director of the Metropolitan Museum and author of the historical thriller, King of the Confessors (1981), or Thomas Baird, who became an author of some ten novels. Weitzmann’s intention to establish his own school of students was, in the end, successful.

3.6. The Impact of Visual Narratives on Weitzmann’s Autobiography

Weitzmann was a great storyteller: he did not just analyze stories, but also told and wrote them. His special sensitivity for narration found its way not only into his research, but into his written memoir, Sailing with Byzantium from Europe to America (1994). In the introduction, Meeting Kurt Weitzmann, the former-student-then-collaborator Galavaris recounted his memories about the end-of-semester beer-drinkings with Weitzmann sharing his life experience, and emphasized two characteristics of his storytelling: being amusing and instructive, the latter clearly with a pedagogical aim (Galavaris 1994, pp. 10–13). The history of art history is not abundant with autobiographies: there is Bernard Berenson’s diary, Rumor and Reflection (1952), and Kenneth Clark’s memoirs that reflect on it (Another Part of the Wood, 1974; The Other Half, 1977), the unfinished and yet unpublished memoir of the Russian spy, Anthony Blunt, the memoir of Weitzmann’s great mentor, Adolph Goldschmidt’s Lebenserinnerungen, published posthumously in 1989, and recently the Princetonian Gary Vikan’s Sacred and Stolen (2016). Weitzmann was writing Sailing in his final years, and, like Goldschmidt, never saw it published. The text is a first person narration and is arranged roughly chronologically. It reconstructs events of scholarly interests, such as the activities Weitzmann participated in, initiated, or organized, and describes the people he got in contact with as a researcher, traveler, and professor, just as it is pointed out in the foreword: “I shall, thus, describe art historical events as they affected not only my own life but also that of those with whom I came into close contact” (Weitzmann 1994a, p. 29). Defining it as a scholarly memoir, Weitzmann presents the major storyline of his life, giving a somewhat narrowed focus to the idea of what a personal autobiography would cover.

Convincingly, it is a vita activa driven by a search for knowledge, fulfilled by work, and formed around major changes. The reader first meets a curious schoolboy interested in the visual arts, who becomes a student, and in time, a young researcher. Due to his persistent work, his personality, and some fortune, he gradually gets higher and higher, reaching the status first of a well-established scholar, then a professor, an adviser, a member of prestigious institutions, and finally a curator. As his physical capacity decreases, his scholarly reputation grows. Carolyn A. Barros argues that the readers of an autobiography expect substantive changes of the life narrated, which indeed lies at the core of the genre: “Change is presented as transformative, a significant mutation in the characteristic qualities and societal relationships of the principal persona (…) Change is then the operative metaphor in autobiographical discourse” (Barros 1998, p. 2). In Weitzmann’s intellectual history, changes of locations further emphasize the transformations of his professional self and denote the changing substance of his scholarly identity: his early years were spent in different towns of Germany; the young researcher lives for almost a decade in Berlin; the highly appreciated art historian is affiliated to Princeton University; the adventurous researcher finds inspiration in the remote monasteries of Mount Athos and Sinai; the knowledge of the scholar is disseminated through various trips to Europe and the US; the curator works for the Metropolitan Museum. In the autobiography, just as its title Sailing suggests, he easily crossed regions of East and West just as he traversed the art of the Byzantine East and the Latin West in his research. Life, as lived by Weitzmann, happened both in time and space.

Weitzmann’s Sailing is certainly closer to the genre of hagiography than to a Bildungsroman, as it is not a novel of self-formation, but rather a constructed net of events, persons, and acts, and the consequences of acts. Dolezal rightly cited Panofsky about the genre of autobiography being “a construct rather than a mirror of the past” (Dolezal 1998, p. 223) and properly noted that “despite the copious detail, the autobiography does not incorporate a more interpretive or reflective view of Weitzmann’s scholarly life.”46 However, the question still remains: why is there so much focus on events and no details of deeper motivations and feelings? Among the few reasons, an obvious one is surely the way it was written. Weitzmann mentions twice that he did not keep a diary and wrote his autobiography based on the entries of his personal calendar and administrative lists of students, courses, and travels, and finally “around this skeleton I filled in details from memory” (Weitzmann 1994a, pp. 25, 487)—which also meant his wife, Josepha Fiedler’s memories. We don’t know whether he was familiar with Hayden White’s dilemma of the modern historian approaching and interpreting annals or chronicles, in this case not medieval but his very own ones, when he was dressing them up for representing the official story of his life (Cf. White 1980). But he must have been well aware of the literary genre that could also serve him as a model, hagiography. Martin Hinterberger describes hagiography as a genre founded on mutual relations that is also true for the autobiography of high-ranking persons: the text “usually relies on the existence of a recognised saint, but at the same time it promotes his/her recognition.” The Sailing can be paralleled with the subgenre of ‘Life (vita)’ because “a vita treats the life of a person who reaches sainthood not by means of death, but by his/her way of life, of their life in Christ” (Hinterberger 2014, pp. 28–29), except that in this case, Christ is substituted with his modern counterpart, academia, or more generally, knowledge.

Weitzmann, however, handles events and characters in his memoir in a very visual way, which suggests that he acquired his storytelling knowledge from the narrative images he studied throughout his life. The density of the narrative, the action-centric storytelling, the repetition and the dynamism are exactly the qualities he was looking for and appreciated in book illustrations. The most essential element of Weitzmann’s narration is the intentional action. Almost all the paragraphs in the book can be labeled with action verbs, like traveling, visiting, meeting, organizing, examining, teaching, lecturing, writing, or publishing, and the paragraphs are formed around these central actions. Weitzmann, the protagonist, appears countless times: he stands behind the events, he is the source of knowledge, he is the center of a huge network. This accumulation of actions is a very visual way of treating stories. In the introduction to the memoir, Galavaris had an apt remark on this dynamic flow of action, saying that “some of the stories may provide material for a film scenario. Others may be suitable for a comic strip” (Galavaris 1994, p. 21). Weitzmann regarded events and actions as being the most crucial narrative elements through which stories in images are constructed and perceived. The density of the actions and scenes he observed was an extremely significant component of illustrated manuscripts, a characteristic that was often taken to the extreme by ancient illustrators, as in the “individual books of the Bible, for the obvious reason that the scenes follow in such quick succession and are so numerous that, on this scale, the illustration of the entire Bible would have been a practical impossibility” (Weitzmann 1977, p. 15). This congestion of scenes and actions is provided by some of the most wonderful early Christian manuscripts, the Vienna Genesis, the Cotton Genesis, or the Ashburnham Pentateuch.

Dynamism and repetition give the impression that Weitzmann is the necessary source of action. Turning the pages of his memoir is very much like looking at the consecutive scenes of a grand work of narrative art, a picture cycle, which uses the method of continuous or cyclic narration showing the protagonist in each scene during an action. It is as if the readers would look at the Emperor on Trajan’s Column preparing for battle and gaining victory, or William, Duke of Normandy, and Harold II, King of England, on the Bayeux Tapestry, or following the protagonists of the Joshua Roll. Similarly, when Weitzmann is writing about his trips to Mount Athos, we learn that he traveled there five times, lived with the monks, took photographs, studied manuscripts, fell ill, had difficulties with acquiring permissions, recognized thefts, etc. He does not dwell on details of the environment, the location is as denotative as a text inscription would be on a manuscript page: the breathtaking view of Mount Athos was not even mentioned. However long the memoir is, the action-based approach and the simplicity of narration establish an intensive flow of events. We, as readers follow him, action by action, paragraph by paragraph, location by location. The words with which Weitzmann described the “general principle” of late antique book illustrations, “[these books] rendered as many events as possible in concise, frameless scenes which follow one to the next” (Weitzmann 1977, p. 10), could have been written about his memoir, the Sailing.

Experiencing this rhythm of life corresponds with the viewing experience of a system of illustrations, be it as a roll, a long tapestry, a glass window or a painted ceiling. Richard Brilliant, one of Weitzmann’s students, describes this experience in regard to the Bayeux Tapestry, but it is equally valid for the pages of this memoir: “The directional flow of scene to scene and their progressive cohesion into episodes were the products of an overriding design that guided the eye towards the viewer’s right along the continuum of the narrative from the beginning to the end. (…) That very measure of narrative time passing coincided with the visual passage over the Tapestry’s spatial fabrication of time, thus uniting narrative and experimental time in the act of perception” (Brilliant 1991, p. 117).

In the memoir, even the description of people takes the form of a narrative, which creates parallel storylines and embedded episodes. Weitzmann loved to be connected to interesting people. A great observer, he was quickly able to discern people’s potentials and highlight fundamental events that made up their lives and identity. A wide range of people feature in the memoir: professors who influenced him (Goldschmidt), colleagues and other art historians (Morey, Friend, Kraeling), people related to research institutes and museums (the Blisses), German refugees (Richard Ettinghausen), his most beloved students, other famous Princetonians or family members, and many more minor characters. Weitzmann specifies the circumstances in which he met them and offers short remarks about their most intrinsic characteristics, which are often enriched by short stories about their personality or life-changing episodes, be they sad, funny, tragic, or deeply human. It is a long inventory of acts and characteristics.

The people who were subjects of these microstories were also meant to be the first audience of the memoir; they defined Weitzmann’s scholarly identity and also reassured the standard of his professional qualities. Susan Engel’s consideration about the shared moments illuminates Weitzmann’s engagement: “Many personal recollections that contribute to one’s identity unfold in highly motivated and charged situations – where there are other people (…) These situations then end up shaping one’s life story as it emerges across time and place. In this way context plays a huge role in determining the self one knows through one’s stories about the past” (Engel 1999, p. 87). This group of people formed the referential world of Weitzmann’s scholarly life that determined his self.

The convergence of Weitzmann’s approach in his research on narrative illustration and his written narratives seen in his personal autobiography is what makes him unique from a narratological point of view. Suzanne Preston Blier called attention to the significance of autobiographies that supplement scholarly research claiming their vital part in historiography: “the lenses offered by particular theoretical perspectives (…) necessarily also remain a vital part of the self-history of art, even when such theories are depassé, or as is equally often the case, filled with as many problems as provocative vistas of academic insight” (Preston Blier 2001, p. 39). Thus, the events described in Weitzmann’s memoir that are relevant in the formation of a diachronic sense of self induce the process that reflects on the central topic of his studies, pictorial storytelling. The density of the narration, the action-centric perception of storytelling, the repetitive patterns in the construction of his life-story, and the dynamism of the flow of events are the characteristics that constitute the parallels with the way pictures handle stories in the visual.

4. Conclusions: Kurt Weitzmann’s Importance and Legacy

In the Roll and Codex, Weitzmann developed a methodological apparatus for studying narrative book illustration with two important features: an evolutionary narrative typology that paid attention to the narrative strategies the painter chose for presenting a story, and a comparative narrative analysis that observed stories in illustrations in relation to their textual source. The evolutionary view was the heritage of the art historical past. Weitzmann elaborated further the typology for ancient Greek art and subsequently adapted it for Byzantine and medieval art, thus determining the direction of future research. The special attention to narratives at the time was quite a novelty. For all these fields, narrative analysis rapidly became an essential methodological tool. In the pre-Weitzmann era, manuscript illuminations had their style, iconography, and genealogy, with Weitzmann, they acquired explicit narrative qualities. Weitzmann’s ideas were clearly built upon earlier accounts of narratives found in German art history and aimed at the further refinement of the already existing terminology and theory.

Therefore, the Roll and Codex constituted a link not just between the art of the Byzantine East and the Latin West, but between the late 19th-century European art historians and their US successors in the research on pictorial storytelling half a century later. In the dissemination of ideas on visual storytelling, his extensive scholarly network was instrumental. The enduring effect of Weitzmann’s methodology can be traced from the mid-20th century: first in the research of his colleagues, Otto Pächt and Carl Kraeling, and afterwards, applied and renewed by his disciples, like Richard Brilliant and Herbert L. Kessler, who formed the second generation of scholars examining this topic. Weitzmann’s circle made Byzantine studies the leading field for research into visual narratives over the period from around 1940 to 1980. From the 1990s, Weitzmann’s influence declined in Byzantine studies, because new research focused more on individual narrative works and the spatial arrangements of narrative cycles, but his effect is still prevailing in archaeology. Primarily due to Weitzmann’s activities, the interest in visual narratives remained continuous throughout the century.

Due to his position, Weitzmann was a principal authority in Byzantine studies for decades, and that meant a favorable way of promoting his new methodology. The procedure of canonization followed the process David Carrier described: the ideas were refined through teaching and lecturing, then they were published in a book, and were accepted by certain institutions: “these institutions guarantee that there will soon be an elaborated literature on that problem. When research is thus institutionalized, post graduate students want thesis topics; younger professors need to publish; and Slade lecturers need to find original ways to describe familiar works” (Carrier 1989, p. 339). In the US, all the institutions that promoted narrative interpretations of Byzantine art had productive scholarly connections with Weitzmann: Princeton University, Dumbarton Oaks, the University of Chicago, and through his former students, the John Hopkins University at Baltimore. These centers also provided the primary context in which the change in the perception of narration took place.

The more than seven decades that have passed since the publishing of Roll and Codex offer a historical perspective on Weitzmann’s accomplishments. His position allowed him to change the standards, and he did change profoundly the discourse on manuscript studies and visual narration. The criticism received for narrative theory was not so harsh as in the case of picture criticism, but it was more diverse and constructive. His methodology for text-image comparisons was challenged in the 1980s by the interdisciplinary approaches of W.J.T. Mitchell and his colleagues, while the problem of narrative typologies, of which Weitzmann’s has become the common denominator for almost all subsequent ones, is still a vital issue in art history and archaeology. Weitzmann’s methodology implemented a significant shift in the analysis of narrative imagery, which at the time worked against the hegemony of iconography.

Weitzmann was not only a story analytic, he was also a storyteller. The highly charged bond between Weitzmann’s life, work, and writings is reflected in his scholarly autobiography, the Sailing. As an art historian’s testimony, it goes beyond a documentary account of his life and consciously positions and reassures his professional self. It is as if Weitzmann took Michel Foucault’s thought literally: “From the idea that the self is not given to us, I think that there is only one practical consequence: we have to create ourselves as a work of art” (Foucault 1984, p. 351). He indeed created his professional self so as to resemble the narration known from Roman artworks and Byzantine and Early Christian manuscripts.

Examining Weitzmann’s work beyond Byzantine studies, in the critical discourse of 20th century art history and narratology, it is clear that one of Weitzmann’s strongest roles was an initiator: his ideas fostered the establishment of visual storytelling as a distinctive area of study. He was certainly aware of it. “I am only one of the actors on the stage” (Weitzmann 1994a, p. 29)—the quotation from the foreword of his memoir shows both the way he regards himself as being part of a larger scene of a scholarly network and a narrative approach toward his self-apprehension. The history of the research on pictorial storytelling is yet to be written. In this historiography, many well and lesser known theoreticians, from Alberti and Leonardo to Lessing and Franz Wickhoff, would deserve an individual chapter. Surely, Weitzmann is one of them. Yet, in this story, Weitzmann is not only one among many actors, he is a major protagonist.