What is art history as a discipline: Insights gleaned from some principal methods of art history

It has been convincingly argued that art history as a university discipline began in Germany, more precisely, at the University of Göttingen in 1799. In 1966 Reference KultermannUdo Kultermann (1966), a German art critic, published his noteworthy Geschichte der Kunstgeschichte, a history of art history from Giorgio Vasari to Ernst Gombrich, an unprecedented publication at the time and still very useful today.

As Reference ChastelAndre Chastel (1987: 10) appropriately diagnosed, art history, which had achieved its height as a discipline during the 1910s with the formalism of Alois Riegl and Heinrich Wölfflin, receded into the background later on, and in the 1980s fell into a state of confusion. But along with Chastel I hold out hope that this confusion has not become a chronic condition. It was most likely engendered in part by the domination of the so-called “linguistic model” in the human sciences, or by the penetration of the language-based disciplines into the realm of art historical study, an academic phenomenon that emerged in the 1960s. Some important language-based art historical methods are iconography/iconology, Marxist or social art history, feminism, psychoanalysis or the analytic/art-psychological approach, semiotics and structuralism, and so forth.^{Footnote 2}

This influx could also be explained by the fact that in recent years art historians have expanded the range of visual phenomena they deal with, from the fine arts like painting, sculpture, architecture and minor arts to other media such as advertising posters, comics, animated cartoons, television, computer games and so on. On the one hand, the expansion of art history's sphere is driven by the assumption that hitherto existing methods of art history, including the language-based ones, are sufficiently applicable to a wider range of objects. Yet it can also be attributed in part to the rise of the study of “visual culture.” Some historians of culture such as Reference BurkePeter Burke (2001) and Reference GinzburgCarlo Ginzburg (1994, Reference Ginzburg1996) have, without any difficulty, thrown their weight around in the realm of art historical practice.

Before going further, let me define in my own terms the work of art, the object normally dealt with in the practice of art history. A work of art consists primarily of its form, contours and colour. Some may claim that it does not have meaning until it is beheld. Nevertheless, I would argue that a work of art has an intrinsic meaning, or in some cases, a number of intrinsic meanings. This is so even though it is true that these meanings are not objectively accessible to us before we view and decipher or interpret the work of art in terms of both its visual appearance and the historical evidence relating to it. For the simplified discussion that follows, Old Masters paintings will be the subject of inquiry.

Before most Old Masters paintings are taken up in art historical practice, they require identification; they are subject to a quest for authorship or authenticity (to render them distinguishable from their copies or fakes), as well as their geographical and historical point of origin. Roughly contemporary with the foundation of art history as a Western university discipline in the 19th century, connoisseurship emerged under the influence of figures such as Giovanni Morelli and Bernard Berenson. In my view, the exercise of connoisseurship is still today the primary and indispensable foundation for training in art history. This is because the traditional methods of connoisseurship, those that place primacy on “the eye,” have been expanded to constitute a more rigorous visual examination of a work of art. Such examination can be applied to works that newly reappear on the art market or that were poorly identified and poorly classified in the past. In fact, by using new technical aids and other scientific means of examination, such as dendrochronology, X-ray photography, infrared reflectography, and chemical and spectral analysis, this approach continues to find ever more useful data concerning works of art for further comparison. Despite these contributions, however, connoisseurship is in decline, with the exception of certain regions of the world where the basic data of art history are still being collected and sorted. In a sense, the decline results from the domination of language-based methods in the field of art history, since iconography, psychoanalysis, and other increasingly lively approaches such as Marxist and feminist criticism, structuralism and semiotics are not predicated upon direct access to the art object. Because art history has always been an interdisciplinary enterprise, the contributions that the insights and theories of other disciplines within the humanistic sciences in particular have made to the development of art history should not be underrated.

It is no coincidence that the historical critique or reevaluation of the achievements of the abovementioned figures and connoisseurship in general incited heated debate during the phase of confusion in art history.^{Footnote 3} Old-fashioned methods of connoisseurship were said to be tainted by an over-emphasis on the eye, intuition and skilled practice rather than on intellect and knowledge of the text. This critique applied to the ahistoricism represented by the so-called Morellian method in particular. Giovanni Morelli practiced a sort of detachment as the bearing of the experimental scientist in a double sense. First he detached the form of the work of art from its content (in fact, he showed little interest in the subject matter and the iconographical meaning of the work). Second, he searched for marginal and irrelevant details – insignificant traits like drapery, earlobes or the shape of a hand, elements on which the artist exerts the least conscious control – as revealing clues for attribution, rather than more weighty artistic features like composition or expression or color distribution and so on.

It is not surprising that connoisseurship was the first method to be abandoned by the so-called “revisionist” art historians who examined their own assumptions by adopting methods from related language-based disciplines. However, it is also true that connoisseurship has fostered several skills central to the activity of art historians: a practiced eye, visual memory, sensitivity to quality, and an ability to reconstruct the processes of artistic creation. One's eye, a good eye in particular, requires incessant training and practice just as that of a connoisseur. Visual memory then provides the means for internalizing comparisons with other works in the mind, guiding perception and classifying the work of art to build a concrete foundation on which the conceptual structure and practice of art history can be based. Discrimination and a sense of quality are indispensable for art history as long as we intend to distinguish art history from other kinds of history. Reference OffnerRichard Offner (1951: 24) underscored this point when he observed that the “history of art is not a verbal structure but a physical one,” even if it must be published and taught with words. It should be noted that everyone looks at the same work of art differently from one another, with their own discriminatory sense of quality. Consequently, just as a work of art probably has as many meanings as there are viewers, it can also be interpreted just as diversely. At this point art history can set out to work toward a better and more comprehensive understanding of the work in its intellectually reconstructed historical context, by adopting one or more art historical methods.

If a historian of culture attempted to view and interpret a work of art that had previously been examined and interpreted by art historians, a dispute regarding the interpretation of the work might emerge between the two types of scholars. A notable example of such a dispute was the “caso Guidoriccio” (the case of Guidoriccio), which consisted of two divergent points of view of the “Guidoriccio at Montemassi” in the Palazzo Pubblico, Siena, and the “New Town” which in 1981 was uncovered lower down on the same wall in the Sala del Mappamondo. This controversy was instigated in 1977 by Gordon Moran, an American scholar of medieval art history whose findings were supported later on by Michael Mallory and continued to be commented upon by art historians and historians until the end of the 1980s.^{Footnote 4} Reference Mallory and MoranMallory and Moran (1986: 250) declared that “the Guidoriccio is not a documented work of 1330, not by Simone [Martini], and not of the Trecento.” They reached this conclusion exclusively through a reexamination and reinterpretation of the written material and evidence, and by putting forward newly discovered evidence of their own. The international congress of Simone Martini studies held in Siena in 1985 became a heated stage for the Guidoriccio debate. In concluding its proceedings, Reference Bologna and BellosiFerdinando Bologna (1988: 240) declared in a meaningful and most memorable manner that we art historians rely primarily on this petition of principle: “It is not the documents which authenticate works of art, but works of art which authenticate the documents.”

Interestingly enough, Japan experienced a similar controversy regarding the date and authorship of the most famous example of a genre of screens depicting the city of Kyoto (known as rakuchu rakugai zu, 洛中洛外図, literally “Scenes In and Out of Kyoto”). The controversy took the form of a debate between the historian Imatani Akira and the well-known art historian Tsuji Nobuo during the 1980s. This pair of screen paintings had generally been dated by most Japanese art historians to the 1560s or 1570s and considered to have been created by Kano Eitoku, the primary painter in service of Oda Nobunaga. Imatani proposed, however, that the Uesugi Screens reflected the actual layout of the capital from the year 1547, criticizing the art historians’ interpretations of the written material and evidence surrounding the screens; he argued that they should be understood as shinkei zu (真景図), a “true view” of the actual landscape. By contrast, Tsuji insisted that even shinkei zu could incorporate capricious and imaginary renderings of the late medieval townscape, and placed primacy on the discriminatory quality of the works, in the process reinterpreting the related material with insight and flexibility (Reference ImataniImatani 1984; Reference TsujiTsuji 1986b; Reference ImataniImatani 1988.)

Bologna's statement about authentication encourages the conviction that students of art history should commence with the close examination of works of art in order to develop a practiced eye, and, in so doing, foster their visual memory and discrimination of quality. Needless to say, artists or art-makers never begin from nothing. This basic fact enables the student of art history and connoisseurship to acquire further knowledge of each work of art, not only by carrying out morphological and stylistic comparison with other well-identified works, but also by ultimately attempting to incorporate works into the geographical and/or historical context in which they were most likely created.

The discrimination or judgment of quality in works of art is certainly neither a means nor an end limited exclusively to art history; it is also central to the philosophical discipline of aesthetics. The differences between the two disciplines, however, are not inconsiderable. While one intends to incorporate the results of such discrimination into a historical or geographical context, the other, which is more interested in the universal characteristics of the perception of beauty, as well as the history of taste, tends to apply it in a more conceptual or ahistorical manner.

In a word, art history is simply a discipline of the humanistic sciences which, starting from a careful viewing of an individual work of art or visual object, works toward the building up of a historical and geographical and, if possible, worldwide map of art. Nevertheless, one wonders if art history can truly achieve a comprehensive map of art on a global scale, to be viewed without prejudice from any historical and geographical perspective. It has been pointed out that every method adopted by an art historian depends on the interests that he or she brings to a particular work of art and the questions he or she wants to ask of it. He or she will inevitably rely on different methods for different purposes. And because no method can ever provide a comprehensive account of a work of art, the creation of such a worldwide art map would appear to be a virtually impossible task.

From the 1980s onward, that is to say, during the era of confusion, a bipolar tendency in the study of art history has emerged. On the one hand there is the emergence of a globally oriented scholarly focus that could be called world art studies, or even world art history; on the other, a regionally oriented practice that could be categorized as a traditional form of art history based upon the analytic, interpretive and narrative methods of Western art history. This is so even if each region has increasingly adopted indigenous technical terms, ways of thinking and historiographical writing methods for the analysis of regional or local artworks. Even now, however, art historians from different countries vary in what they study, and a wide range of interpretive methods is employed. Iconography remains a principal or default mode of scholarship.

Quite recently, James Elkins attempted to articulate five arguments in favor of and five arguments against the idea that art history is, or could become, a single coherent worldwide enterprise. This attempt was made in the context of an informal introduction to the art seminar Is Art History Global? held as a roundtable conversation in the University College Cork (Reference ElkinsElkins 2007: 3–23). He concluded that the five points in favor of art history as a global enterprise are more compelling than those against, and that art history is indeed becoming a worldwide enterprise. I would never deny this prospect. By the same token, however, Elkins is also correct in arguing that there is no non-Western tradition of art history, if by that is meant a tradition with its own interpretive strategies and forms of argument.

It is important to acknowledge that both a globally oriented and regionally oriented art history can engender a number of scholarly limitations. As long as art history is introduced into a country or region such as East Asia or Australia as a university discipline, even a regionally oriented art history can never work well without making use of common Western art historical methods or approaches such as formal analysis, iconography and periodization. On the other hand, as Elkins points out, a globally oriented approach runs the risk of dissolving art history into image studies, or the kind of visual culture studies mentioned earlier. As a result, art history runs the risk of being done away with as an academic discipline by national and regional governments that embrace the latter.

There is little doubt that currently, as Reference ElkinsElkins (2007) insists, art history is dependent upon Western conceptual schema. He cites several of the most ambitious recent attempts to challenge the question of the “Westerness” of art history from widely differing perspectives. Among them is one book worthy of special consideration, David Summers’ magnum opus Real Spaces (Reference Summers2003). As Elkins himself states, the book is the only recent attempt to write about the entirety of world art history without relying on chronology as a fundamental ordering principle, while at the same time introducing or newly formulating concepts that can be applicable to art created anywhere in the globe. Accordingly, it is written at a high level of abstraction. It attempts to create a conceptual framework that enables the treatment of all cultural traditions on an equal footing. At the same time this framework attempts to accommodate and understand opposition and conflict both within and between cultures. The core of the author's argument is a proposal to replace the modern Western notion of the “visual arts” with that of the “spatial arts”; the concept of “space” is further subcategorized as coordinate space, metaoptical space, metric space, personal space, real space, social space, viewer space and virtual space. As for Summers’ concept of real space, which lends the book its title, Elkins wonders if it should be used at all for pre-Renaissance and non-Western art, and rightly concludes that space as a concept is a problematic starting point for an art historical analysis intended to bridge Western and non-Western cultures.

Even more problematic is the author's argument, based on the so-called “linguistic relativity hypothesis,” that language has a shared structure that provides a basis for a shared experience of the world. In reality art historians are forced to pay continual attention to the languages of different cultures in order to better understand works of art through concepts that would have been familiar to their makers and initial viewers. Summers’ book, even if it is multilingual, is almost entirely unconcerned with non-Western languages. This serves as a limitation to an otherwise ambitious book that purports to find “the means to address as many histories as possible nearly enough in their own terms to permit new intercultural discussion” (Reference SummersSummers 2003: 12).

Another notable proposition toward world art studies has recently been put forward by John Onians. He has named it “neuroarthistory.” Onians has always been much interested in the search for the origins of artistic creation by human beings on the basis of neural biology, or neuroscience. This interest stems from his belief that culture can never explain why art came into being in the first place, and leads him to argue that the roots of art clearly lay in the art-maker's unique biology, and particularly in his or her unique neurobiology. Onians was encouraged by the recent and remarkable development of neuroscience, with discoveries such as the recognition of several groupings of neurons and their “receptive fields,” the brain's neural “plasticity,” and so on, to dedicate a guidebook to neuroarthistory (Reference Onians2007). On another occasion he articulated the reasons why using neuroscience as an aid to the study of art is highly advantageous:

This theoretical aspect of Onians’ argument is of great interest. As he himself states, Onians’ art historical position is influenced and encouraged by that of his teacher Ernst Gombrich, as well as that of fellow Gombrich disciple Michael Baxandall. Both were interested in the psychology of perception and studied the physiology of the eye and the retina, as manifested in publications such as Gombrich's Art and Illusion (Reference Gombrich1960) and particularly The Sense of Order (Reference Gombrich1979), and Baxandall's aforementioned Painting and Experience in Fifteenth Century Italy. In introducing his concept of “the period eye,” based on the notion that individuals are differentiated by their neural apparatus, the latter wrote,

This conviction allowed him to assert in the preface that “the visual skills evolved in the daily life of society became a determining part of the painter's style.” On the other hand, Onians goes much further inside the brain, or to the two neural networks that link the sensory and the motor, as well as the chemical links to the viscera and the plasticity of the cortex. A sort of ecological determinism supported by the advanced study of neuroscience merits a serious consideration. The new discoveries on the so-called synaptic neuronal plasticity demonstrate that prolonged exposure to certain forms of visual conditioning will lead to a subtle reorganization of synaptic connections and could establish certain patterns of visual skills. It has been proven that the brain is not only plastic in its early phases during ontogeny, but is also continually tuned and adjusted in response to visual and other kinds of experience. The new research of neuroscience has strongly suggested that the ways we view and interpret objects are enabled and conditioned by a distinct neuronal architecture, and that at the same time they influence that architecture itself. This synaptic plasticity supports the notion of specific cognitive styles, such as what Baxandall's “period eye” could engender, that can in turn be associated with particular cultures, within specific age groups and even within certain strata of society. Even with its many merits, however, in the context of world art studies neuroarthistory has intrinsic limits. Ironically it runs the risk of underrating the existence of local traditions or period eyes associated with specific places and periods within larger abstract categories such as “Western,” “Oriental,” “Chinese” or “Japanese” art. And neuroscientists have noted that on a subconscious level we are mentally inclined to see what we would like to see.

For Onians, neuroarthistory does not replace other art historical approaches, but supplements them. In arguing the interactive nature of the “dialogue” between reliquary figure and devotee in the medieval West, Reference Montgomery, Akiyama and TomizawaScott B. Montgomery (2009: 64, 112–115) cited as evidence of his argument Bernard d’Anger's famous description of animation in the reliquary figure of St Gerald at Aurillac: “So strikingly was the face of the human figure portrayed that it seemed to several people as if it were fixing its beholders with a piercing glance, as well as sometimes graciously granting the petitions of the supplicants with a twinkle of the eye” (Montgomery quotes from Reference DahlDahl 1978: 177). This description, seemingly a simple rhetorical exaggeration, could be credible if interpreted within the framework of neuroscience. Imagine a medieval devotee viewing the reliquary figure of St Gerald; it could indeed appear lifelike when viewed on an altar, with candlelight flickering off the reliquary‘s gold in particular. Such lighting would enable the devotee to experience St Gerald's eyes as twinkling, strengthening his or her mental inclination to see what they were subconsciously eager to see. Examples such as this one support Onians’ argument about the usefulness of neuroscience as a supplemental approach to world art history.

Art history in Japan: Past and present

Reference AgostiGiacomo Agosti (1996: 16) drew our attention to the process by which art history was introduced to Italy, noting that the discipline was first taught not in universities, but in the art academy; this anomaly might have derived partially from the nationally urgent matter of classifying and cataloguing Italy's extremely rich artistic patrimony soon after the unification of the country. A similar development occurred in Japan. After the Meiji Restoration of 1868, Japan was faced with the urgent need to register its cultural and artistic patrimony, scattered in old temples. This enterprise was undertaken, with great enthusiasm, by the American scholar Ernest F. Fenollosa, by Okakura, and by others. Fenellosa, who was invited to teach at the University of Tokyo in 1878, was the one who introduced modern art history in Japan, Alongside his instruction of politics, economy and philosophy at the university, he was also highly active as an art critic, delivering lectures on art and, from 1884 on, writing book or exhibition reviews in newspapers published in English such as the Japan Weekly Mail.

On November 12, 1898, for example, Fenollosa authored an intriguing article entitled “The Present Exhibitions of Painting.” It was a review of the first exhibition of paintings planned and realized by the painters of the newly founded Bijutsu-in Society (美術院) during the previous month. Fenollosa compared the praiseworthy effort of Bijutsu-in painters to the two-hundred-year development of the Italian Renaissance from Giotto to Michelangelo. He concluded his review by making use of a rhetorical expression of admiration rooted in Renaissance historiography: “Easily he [=Shimomura Kanzan] has forged to the front, and now stands to the new school of Japanese art as the young Giotto stood to the total achievements of Italian … It is the solid outcome of the prophecy uttered by the lamented Kano Hogai before his death in 1888; for, if Hogai was the Giunta Pisano of this movement, and [Hashimoto] Gaho his Cimabue, here again I say is our Giotto” (Reference FenollosaFenollosa 1898: 489). This degree of admiration has never subsequently been manifested in the historiography of Japanese art history.

Okakura Kakuzo (better known as Tenshin), a disciple of Fenollosa, was the first scholar to teach modern art history in Japan, representing the subject with courses on both the “History of Western Art” and the “History of Japanese Art” from 1890 through 1892 at the Tokyo School of Art (now known as the Tokyo University of Fine Arts and Music). From this point forward, art history as a distinct university course in Japan has enjoyed a relatively long tradition. Although art history was included in the first lecture on aesthetics delivered by Fenollosa in 1881, the first chair for aesthetics to which art history was subordinated was founded later, in 1893, at the University of Tokyo, and the chair for art history proper later still, in 1914. From the beginning, at both the Tokyo School of Art and the University of Tokyo, two different types of art history were taught and practiced, namely Japanese or East Asian art history and Western art history. Their methods and terminologies were and still are quite different from one another. Although each discipline has a long tradition and has undergone considerable scholarly development, interdisciplinary exchange or comparative study has rarely taken place except on an individual level.

Let us examine the specific case of the introduction of the Morellian method of connoisseurship to Japan. The first mention of this method, to my knowledge, appeared in an essay “An Urgent Artistic Business” by Okakura in issue 93 of 時事新報 (Current Affairs: An Arts Weekly), on February 19, 1908:

Iwamura Toru, lecturer and then professor of Western art history at the Tokyo School of Art from 1899, was also active as an art critic. In June of the same year (1908) he published an explanation of Morelli's methodology entitled “Morelli's method of authenticating old paintings” (“ モレリの古畫鑑定法 ”) in the first volume of the journal 光風 (Kofu: Reference IwamuraIwamura 1913). The article was instigated by a lecture given recently by Okakura (most likely the one resulting in the abovementioned publication) in which Okakura described Morelli's connoisseurial method. Yet Iwamura claimed that he himself had known about it for some fifteen years (thus from around 1893), and commenced to explain Morelli's method at length. His familiarity almost certainly derived from his reading of the English translation of Morelli's Italian Painters, Critical Studies of Their Works, published in London in 1892. A copy of the translation was kept in his library, and his reference to it is clear from his citation of an English technical term, “art morphology,” in place of the corresponding German word found in the original.^{Footnote 5}

Some Japanese scholars have argued that Morelli's method could be successfully applied to the study of Japanese painting. From around 1930 onward, for example, art historians such as Tsuchida Kyoson and Doi Tsugiyoshi employed it as a way to individualize and discriminate between the hands of Kano Sanraku and Hasegawa Tohaku, both famous painters of the 16th century, especially in the case of the brushwork of tree branches in the panel paintings of temples such as Daikakuji, Chishaku-in, and Myorenji in Kyoto (Reference TsuchidaTsuchida 1982: 143, 307). These tentative attempts, however, do not seem to have taken root in Japanese art historical studies in the years that followed, primarily due, as Okakura pointed out, to the highly developed indigenous traditions of connoisseurship that had existed in Japan long before the transplantation of modern art history during the Meiji era.

By contrast, a unique and more effective approach was adopted by Matsuki Hiroshi, who employed the Morellian method to articulate the distinction between authentic ukiyo-e (浮世絵 : “image of the floating world”) woodblock prints by Sharaku and inauthentic ones made by his shadow painter, whose true identity has yet to be identified. Reference MatsukiMatsuki (1985), inspired by the brief and stimulating description of the method in Edgar Wind's Art and Anarchy, put forward his own proposal for the discrimination of the physiognomy of the ear in Sharaku's figural prints, through which he attempted to resolve the controversy over the identification of Sharaku's true identity. According to this scheme, the ears depicted in middle-sized ukiyo-e prints of theater actors or sumo wrestlers around the eleventh month of 1794 are distinct from those of Sharaku's authentic prints produced during other periods.

Although some scholars of Japanese or East Asian art history have sporadically adopted the terminology or approaches of Western art history, these attempts have not resulted in any remarkable contribution to scholarship in Japan. As Inaga Shigemi has pointed out, “a theoretical resistance to the fallacy inherent in an attempt at globalizing art history was clearly formulated by the pioneer of Japanese or East Asian art history [i.e., Okakura Tenshin] at the beginning of the 20th century” (Reference Inaga and ElkinsInaga 2007: 255^{Footnote 6}).

I am particularly drawn to a proposal concerning the comparative history of world art made in an impressive paper delivered by Reference Bialostocki and VayerJan Bialostocki (1982) on the occasion of the 24th International Congress of History of Art, CIHA, in Bologna in 1979. Entitled “A Comparative History of World Art, Is It Possible?” it urged the construction of a common research platform for collaboration and exchange between scholars from many countries to build toward a so-called world art history. In order to realize this, Bialostocki proposed as methodological models several adopted in the comparative study of literature and religion.

I will briefly discuss the usefulness of such a comparative art history. Although intended as a supplementary approach, I consider it no less effective than those hitherto mentioned in bridging the gaps between the globally oriented and regionally oriented art histories, Japanese art history in particular.

Until recently only a few scholars, especially those of Western art history, have seriously argued for a comparative history of art or something related in Japan. In March 2003 I organized a one-day conference titled “Visioni dell’Aldilà in Oriente e Occidente: arte e pensiero” at the Library of the Uffizi Palace in Florence (Reference Osano and OsanoOsano 2003). This convocation provided a good occasion to advance comparative research, thanks to the 21st COE (Center of Excellence) Research Program “The Construction of Life and Death Studies Concerning the Culture and Value of Life” conducted by the Graduate School of Humanities and Sociology at the University of Tokyo. Needless to say, death is not a subject confined to the field of biology; it is a cultural issue that varies in its make-up depending upon time and place. For the study of the rich visual cultures of both East and West, it seems most productive to adopt well-established methodologies in the fields of comparative religion and literature.

From a thematic perspective I thus attempted a comparison of the images of the hells described by both Dante and Genshin, a Japanese Buddhist exegete of the tenth century (Reference Osano and OsanoOsano 2003). In particular, I tried to draw a clear comparison between the illustrations of Dante's Hell by Sandro Botticelli and hanging scroll paintings of the so-called rokudo or “Six Realms of Existence” (六道) inspired by Genshin's descriptions in his Essentials of Salvation (往生要集). This comparison began with their geography and structure, and proceeded to analyze their respective torments and causes. This pairing was not without justification because Genshin's Essentials of Salvation, as Reference WaleyArthur Waley (1960: 12) once stated, was a Dante-esque work that described its subject in a manner similar to the visual arts. Dante's hell in the form of an upside-down cone consists of nine large concentric circles, whereas the Japanese hells of Genshin consist of eight large ones and additional smaller sites of torture. The punishments in Dante's hell, as the inscription on its entrance gate declares, are endless, in contrast to those in Genshin's. In both East and West they are delivered to the damned according to the law of contrappasso (by analogy or by contrast). Dante regarded crimes committed by man's instinctive appetites as deserving of only light punishment in the upper circles of hell, whereas those occasioned by human ill will deserving of heavy punishment in the lower circles. By contrast, the damned described by Genshin are destined to be variously punished because in their life time they could not emancipate themselves from all worldly passions in order to attain to nirvana.

I also referred to the ritual practice of the “deathbed ceremony” (臨終行儀,rinju gyogi), minutely described and recommended by Genshin in particular, in comparison with the Western ars moriendi. The Illustrated Biography of Monk Honen (法然上人絵伝), a National Treasure (Fig. 1), shows one such ritual in which the monk Ryuken Risshi is sitting in prayer, on the verge of departing for the paradisal Western Pure Land. In order to be conducted there by the Amida Buddha, Ryuken's joined hands are connected to a picture of Amida with five color threads. In fact, evidence of this ritual practice can be found in the traces of five color threads found in the hole on the joined hands of the Amida in the famous screen painting Amida Emerging from the Mountain (山越阿弥陀図, Fig. 2).

Figure 1. Scene of the “deadbed ceremony” practiced by the monk Ryuken Risshi, Detail of the Scroll of the Illustrated Biography of Monk Honen, 14^th century, Chion-in Temple, Kyoto (Photo credit: National Museum of Kyoto)

Figure 2. Screen painting of Yamagoe Amida zu, 14^th century, Konkai-Komyo-ji, Kyoto (Photo credit: National Museum of Kyoto)

We can also compare Western art and East Asian art in terms of their lexicons of description, aesthetic categorization, and parameters of judgment. Here I would like to discuss their respective representations of the shadow. The Japanese word 影 (kage: shadow) not only meant the soul or the spirit of the deceased, as it had ab antiquo as found in The Tale of Genji (源氏物語, in particular chapters 34 – 若菜（上): Wakana, 1 – and 49 – 宿木 : Yadorigi), but also referred to the figure of the absent person, brought to the mind of those who pined their absence. Such meanings immediately conjure up the notion of similitude of the keepsake or portrait. The word 遺影 (i’ei), derived from 影, signifies “the shadow left for posterity” or the portrait in memory of the dead.

As Pliny (Hist. Nat. xxxv, 43) recounts, Butades, a potter of Sicyon, invented the method of producing a clay portrait by encircling a shadow cast by his daughter's young lover on a wall. A similar story is described by Kurokawa Doyu in his diary 日次記事 (Hinami kiji) in 1685. According to Kurokawa, when the noted Zen monk Gensan Daishi died, his disciples commissioned the painter Awataguchi Ryuko to create his portrait. That night the deceased appeared to him in a dream. Upon waking, he found the figure's shadow left on a shoji (障子) or paper sliding door, and traced it to make the portrait. From the beginning of the nineteenth century, some portraits were created by tracing the contour of the human shadow. These may have been intended as keepsake images, as in the self-portrait of a shadow by Tani Buncho dated to 1834 (Fig. 3). In terms of artistic effect, these images are similar to those made for Johan Gasper Lavater, but their intentions differ conspicuously. The shadow cast by the human figure or object was rarely depicted in Japanese painting, with the exception of some strikingly impressive or exaggerated cases. Yet a genre known as “shadow pictures” (影絵), in which a silhouette of a motif was shown through or on a paper sliding door, was popular among artists, in particular ukiyo-e (浮世絵) artists, reflecting a playful spirit that is an essential characteristic of Japanese art (Reference TsujiTsuji 1986). The depiction of flickering silhouettes on paper sliding doors was suggestive of the image, reflecting another characteristic of traditional Japanese painting.

Figure 3. Tani Buncho, Self-portrait of a shadow, 1834, Private Collection

A more typical suggestiveness is manifest in a category of genre painting known as “Whose sleeves?” (tagasode byobu, 誰が袖屏風). Such works were folding screen paintings depicting only robes (such as 小袖, kosode, or 袴, hakama) hung on a kimono rack or a folding screen without any human presence (Fig. 4). Inspired by traditional Japanese poems (和歌, waka) and tales (物語, monogatari) such as The Tale of Genji, these works could serve as a pictorial means of recollecting the absent or deceased through a viewing of the kosode robes they used to wear. In this regard, tagasode byobu recall van Gogh's famous still life painting Shoes, or Old Shoes with Laces (Fig. 5). Reference Schapiro and SimmelMeyer Schapiro (1968: 203–209) revealed that this painting was doubtless identical with the artist's work of a pair of shoes Martin Heidegger had in mind in writing his famous essay on “The Origin of the Work of Art”. Criticizing Heidegger's purely philosophico-aesthetic interpretation of this painting, Meyer Schapiro, adopting a semiotic point of view, successfully demonstrated that for van Gogh the shoes depicted therein represented a piece of his own life, “a portion of the self” (in the words of the novel Hunger by Knut Hamsun from the 1880s). Schapiro did so by referring specifically to a deeply affecting story related to van Gogh's shoes, described in Gauguin's reminiscences of van Gogh when he shared living quarters with the artist in Arles in 1888. Both tagasode byobu and van Gogh's Shoes function as a formidable visual metonymy of an absent person or the owner of the depicted object.

Figure 4. Tagasode byobu (left pendant to the pair of Screen paintings), 17^th century (?), Suntory Museum of Art, Tokyo (Photo credit: Suntory Museum of Art)

Figure 5. Vincent van Gogh, Shoes, 1886, Van Gogh Museum, Amsterdam. (Photo credit: Van Gogh Museum Amsterdam (Vincent van Gogh Foundation))