The Metamorphoses of Caroline Walker Bynum

272 International Journal of the Classical Tradition /Fall 2004 a hymn of the type called 'kletic'. "22 For Sappho fr. 18 V. she somewhat unconvincingly argues that, since capitalization is an editorial decision, the capitalized <II>dLv in the first line of the fragmentary poem refers to "Pan the god of goats, wild space, pipes and the silence of noon" (363). For Sappho fr. 31 V., perhaps the most celebrated archaic Greek lyric poem in m o d e m times, Carson provides subtle comments on our main source for the poem: Longinus' mode of sublimity, in the context of his quoting the poem, suggests that while Sappho's body falls apart, Longinus' body is brought together, "a drastic contract of the sublime" (364). Interestingly, in fragment 38, Carson intervenes in the rendering of the ancient text: "you burn me" instead of "us". The five elisions and two correptions in fragment 105a V. are accounted for metaphorically in terms of the poem's self-correction in lines 2-3 which emphasizes desire's infinite deferral (374). Carson persistently pays attention to details and attempts to throw light on the tiniest intricacies of the fragments. Greek sources and texts that are insightfully explored in her notes range from Poseidippos, to the literary critic Demetrios, to Gregory of Nanzianzus and Libanios, to Maximos of Tyre, to Gregory of Corinth, to Yannis Ritsos. Carson employs excerpts from, among others, Samuel Beckett, Emily Dickinson, and George Eliot to elucidate issues relating to specific poems by Sappho. In conclusion, I would like to stress that, although perhaps not intended for classical scholars, Carson's translations are generally reliable and most elegant. Compared to Mary Barnard's celebrated, but overly poeticized and 'reconstructive', renderings, or David Campbell's literal and accurate translations, 23 Anne Carson's If Not, Winter stands out as a major attempt at recapturing the polysemic state of Sappho's fragments, as well as the echo of their original language. It is perhaps the most significant recent contribution to the long tradition of 'translations' (in the broadest sense) of Sappho in Europe and the United States. Dimitrios Yatromanolakis Department of Classics Johns Hopkins University The Metamorphoses of Caroline Walker Bynum Caroline Walker Bynum, Metamorphosis and Identity (New York: Zone Books, 2001), 280 pp. Few medievalists have written so interestingly as Caroline Walker Bynum. Though her work falls broadly into the area of medieval spirituality, it nicely resists categorization into a modern academic specialism, ranging as it does between social, cultural and intelprinted by Lobel/Page [Poetarum Lesbiorum Fragmenta, above, n. 7], translated and discussed by Page in his Sappho and Alcaeus [Oxford 1955: 39], and adopted and translated by Campbell [Greek Lyric I, below, n. 23]). 22. For a different methodological approach, see D. Yatromanolakis and P. Roilos, Towards a Ritual Poetics, Athens 2003, 43-59. 23. M. Barnard, Sappho: A New Translation, Berkeley 1958; D.A. Campbell, Greek Lyric I, Cambridge, Mass. 1982. Marenbon 273 lectual history, literary criticism and---in the theme if not the matter and manner of her latest book--the history of philosophy. Metamorphosis and Identity, a study of the concept of change (and related ideas of identity and wonderment) mainly in twelfth- and thirteenth-century texts, is not Bynum's best work, but for this very reason it offers an excellent vantage point from which to survey her methods and ideas: her achievement, its limitations and the challenges which they offer. Bynum's Achievement Bynum's two major works, Holy Feast and Holy Fast. The Religious Significance of Food to Medieval Women and The Resurrection of the Body in Western Christianity 1 seem to have sharply different subjects and to belong to different modern disciplines--the first to cultural and social history, the second to the history of theology. Yet the surprising conclusions she reaches in each have a surprising amount in common. In Holy Feast, Bynum argues that the extravagant fasting practised by female ascetics in the later Middle Ages should not be interpreted, as it had usually been, in negative terms as an expression of a dualistic hatred of the corporeal, and indirectly, as a product of the medieval hatred of women and conception of them as not just weak, but sensual, bodily creatures. Not only was it a way in which women, from their position of weakness, could manipulate their families and register their dissent from social and ecclesiastical conventions. Their ascetic practices were also linked, Bynum shows, with a wide range of positive images of women, as providers of food and even as food themselves--images which, in a way that often involves a deliberate and daring reversal of sexual roles, underline their special closeness to God. (Already, in an essay called 'Jesus as Mother and Abbot as Mother', Bynum had called attention to the way in which female descriptions were fitted to Christ. 2) In The Resurrection of the Body, Bynum shows that medieval thinking about death was not, as many historians had taken for granted, dominated by the idea of the soul being released from the body. Belief in the resurrection of the body placed a strong emphasis on fleshly, physical continuity and medieval interpretations of the scriptural passages on which the doctrine was based tended to reject the metaphorical interpretations and notions of metamorphosis which Paul's language suggested. Even the sophisticated theory elaborated by Aquinas using the Aristotelian notion of substantial form involves a large degree of continuity between earthly and heavenly bodies. Here too, in an area where scholars have been quick to detect a devaluation of the bodily of a sort that seems alien to most people now (though for some it is part of the period's peculiar appeal), Bynum argues that medieval writers came to a rather sophisticated understanding of the central role of human bodilyness. 1. 2. Holy Feast and Holy Fast. The Religious Significance of Food to Medieval Women (Berkeley, Los Angeles and London: University of California Press, 1987); The Resurrection of the Body in Western Christianity, 200-1336, Lectures on the History of Religions sponsored by the American Council of Learned Societies, New Series, 15 (New York; Columbia University Press, 1995). The essays contained in Bynum's Fragmentation and Redemption. Essays on Gender and the Human Body in Medieval Religion (New York: Zone Books, 1991) are linked to the concerns of both these monographs. Originally published in 1977, it was reprinted in expanded form in a collection to which it gives its name: Caroline Walker Bynum, Jesus as Mother. Studies in the Spirituality of the High Middle Ages (Berkeley,Los Angeles and London: University of California Press, 1987). 274 International Journal of the Classical Tradition / Fall 2004 In b o t h books (and in most of the essays linked with them), B y n u m comes to her unexpected b u t t h o r o u g h l y convincing conclusions t h r o u g h the same, u n d e r l y i n g method. W h e t h e r the text she is using is a saint's life, a chronicler's anecdote, a p o e m or a theological treatise, she is rarely content to stay w i t h the surface narrative or argument. She looks into the language and images the writers use, teasing out a set of concerns that are not i m m e d i a t e l y obvious, b u t seem v e r y clearly to be implicit, given the w e i g h t of evidence B y n u m can bring to bear. 3 This m e t h o d of reading texts links u p especially well with visual evidence, and a feature of her books is usually a set of plates with captions that m a k e the medieval pictures or carvings into vivid illustrations of the line she is arguing. But once the evidence has been gathered so richly and perceptively, it n e e d s to be interpreted. In The Resurrection of the Body, this interpretation can take the rather simple form of a chronological thesis, about the replacement of an earlier m e d i e v a l m o d e l for bodily resurrection with a scholastic one, based on Aristotelian metaphysics. In Holy Feast, B y n u m gives a many-levelled interpretative discussion, w h i c h could serve as a m o d e l for historians on h o w to avoid b o t h reductionism and superficial literalness. W h a t meaning, she asks, do certain ways of conceiving s o m e t h i n g (in this case, fasting) h a v e within their c o n t e x t - - t h a t is to say, the context of medieval social life, values a n d cultural practices? B y n u m is always concerned b o t h to avoid an a p p r o p r i a t i o n (or, worse, a p e r e m p t o r y judgement) of medieval ideas in m o d e r n terms and yet to m a k e these a p p a r e n t l y strange w a y s of thinking and behaving relevant to a w i d e range of readers now, and not just those w h o already h a v e a quasi-scholarly fascination with the Middle Ages. B y n u m ' s w a y of achieving this seeming conjunction of opposites is b y a m e t h o d of analogy 9In an essay dating from 1995 ('Why All the Fuss about the Body? A Medievalist's Perspective'4), w h e r e she talks explicitly about her m e t h o d s and aims, she argues that 9 if w e situate our o w n categories in the context of o u r o w n politics, w e m u s t situate those of the Middle Ages in theirs 9The relationship b e t w e e n then and n o w will thus be analogous and proportional, not direct. It seems to me, that is, that the fruitful question to explore is not likely to be, H o w is Origen (or Christine de Pisan or Aquinas) like or not like [Judith] Butler (or Spivak or Foucault)? Posed in this simple way, the a n s w e r . . , is almost certain to be, not v e r y much. It is far more fruitful to think along the lines: Origen is to Origen's context as Butler is to Butler's. 5 Putting this ideal into practice involves sticking to the contextualizing a p p r o a c h described in the last paragraph, whilst maintaining an explicit awareness, especially w h e n she a n n o u n c e s her point of d e p a r t u r e or comes to d r a w her conclusions, of m o d e r n ques- 3. 4. 5. Bynum's approach makes her, at least to some extent, a literary critic. Yet, despite her determination to look at language carefully, she lacks the finest critics' sensitivity to linguistic specificity--a defect which is indicated by the failure ever to give more than the occasional word in the original language. To see how material and themes not unlike Bynum's are handled by a genuinely literary scholar, it is interesting to compare Peter Dronke's Imagination in the Late Pagan and Early Christian World. The First Nine Centuries A.D. (Florence: SISMEL--Edizioni del Galluzzo, 2003), especially Chapter 4 on earthly paradises. CriticalInquiry 22 (1995), pp. 1-33. Ibid., p. 29. Marenbon 275 tions and modern theories (more usually anthropological, sociological, feminist or literary, rather than strictly philosophical) that try to answer them. 6 The final essay in Metamorphosis and Identity, 'Shape and Story', is an especially good example (perhaps because it was written for a large, popular audience) of Bynum's ability to link medieval texts with m o d e m concerns in a way that enriches both. She begins by looking at some of the meanings that 'identity' has for us, different and even opposite, since we think of 'identity' as what individuates us, but we also have a group identity, as a man or a woman, for instance, or as a Jew, an Indian or a Russian. She then passes on to look at three different literary accounts of that most troubling form of metamorphosis, of man into werewolf, choosing an author from antiquity (Ovid--whose Metamorphoses became popular in the later twelfth century--on Lycaon), the Middle Ages (Marie de France's Bisclavret) and our own time (Angela Carter). Although change is their theme, all these stories, she explains, are also about continuity---otherwise there would be no story at all, but just 'discrete vignettes'. They do not offer 'a single meaning or message' but provide a powerful way of thinking about identity, as 'the shape (or visible body) that carries story.' Carter's idea--it is a suggestive rather than clear-cut one---is that m y 'shape', m y visible body, complete with its fixed and its ever-changing characteristics, the scars, the wrinkles that deepen, the hair that thins and goes grey, is the bearer of m y story. She then develops this theme by looking at some passages from Dante: the metamorphoses in Hell which succeed in obliterating any individual shapes or stories, and heavenly resurrection which preserves and transforms them. By the end of the piece Bynum has not only made known to her readers some great works of literature (for many of her audience on the occasion it will have been a first acquaintance with Marie de France, Ovid or even with Dante; for me it was an introduction to the marvellous and disturbing world of Angela Carter). She has also vindicated her claim to have given us a way of thinking about identity and change: metaphors and stories which, rather than allowing us to think of ourselves as (p. 188) 'monsters and h y b r i d s . . , stuck together from our own sense of the incompatibility of aspiration and situation, culture and genes, mind and b o d y . . , help us to imagine a world in which we really change and yet really remain the same thing.' Bynum's Limitations There are plenty of other felicities in Metamorphosis and Identity--an exceptionally sensitive investigation of Bernard of Clairvaux's language, for example (in 'Monsters, Medians, and Marvelous Mixtures'), and a discussion of wonder which contributes to the growing body of literature analyzing medieval concepts of the emotions. Yet overall the book is, at least in comparison with her earlier achievements, a failure, and it is instructive to consider the two main reasons why. The first of them is, to some extent, an incidental matter, but it is linked to the second, which reveals a problem fundamental to Bynum's very approach, at least as it has been developing in recent years. 6. For example, three of the essays collected in Fragmentation and Redemption are explicitly designed to 'use major twentieth-century intellectual figures as a means to better understand late medieval religion' (p. 14): the anthropologist Victor Turner in 'Women's Stories, Women's Symbols', the sociologists Max Weber and Ernst Troeltsch in 'The Mysticism and Asceticism of Medieval Women' and the art-historian Leo Steinberg in 'The Body of Christ in the Later Middle Ages'. 276 International Journal of the Classical Tradition/ Fall 2004 The first problem with Metamorphosis and Identity is its lack of coherence. This deficiency may be the result of how the book is constituted: only the chapter on Bernard was written specially for it; the other three are based on lectures given at different times and occasions. Bynum herself, however, seems to admit the need for a unifying theme, because in her Introduction she proposes a general, wide-ranging thesis. It contrasts the view of change predominant in the mid-twelfth century, in which it is conceived 'not as replacement but as evolution or development, as alteration of appearance or mode of being' (p. 23), with a 'a new understanding--a new model' which emerged towards the end of that century: 'radical change, where an entity is replaced by something completely different' (p. 25). As presented, this thesis is clearly a chronological one, which w o u l d be borne out by finding few instances of 'radical change' discussed before about 1200, and many after then; indeed, Bynum says that 'the question for historians is w h y ways of conceptualizing might themselves change, w h y many people around 1200 might entertain, with fascination and fear, understandings of change that were infrequent only fifty years before' (p. 26) and speculates about the extent to which the availability of 'new intellectual materials' provides a response. She even connects this change in ideas of change with other developments historians have pointed out in the years around 1200, such as the growth of misogyny and the scape-goating of Jews and lepers. Yet in the chapters which follow Bynum provides no such chronological contrast. Indeed, she does very little to draw attention to the chronology of her texts, most of them from around the second half of the twelfth century, and her concern is particularly to point out the complex attitudes to change within each passage. So far from there being--at least on her account of the material--a sharply marked contrast between accounts from the middle of the twelfth century and those from its end, Bynum's analyses tend to bring out the ambivalence about radical change shared by almost all the accounts, an apparent unwillingness to accept it even when the material seems to call for its recognition. If one asks what Bynum would need to have done in order to provide the chronological analysis her thesis demands, the second, more fundamental defect of the book starts to emerge. One of the main intellectual changes between the mid-twelfth and the early thirteenth century, by which she might have ordered and arranged her material, was the growth of the Universities of Oxford and Paris and the development of a n e w type of Aristotelianism, different in kind from the Aristotelianism of the twelfth century. Yet the whole world of the schools and universities remains a shadowy, ill-defined background presence in Bynum's book. But is it not her prerogative to leave it like that, a justifiable decision about where to place her emphasis? For the Bynum w h o wrote Holy Feast and Holy Fast, it would be easy to answer affirmatively: at that stage, she was writing as a social historian, if one unusually sensitive to the aesthetic and intellectual dimensions of her source-material. But Bynum herself has metamorphosed, though without abandoning her former identity. Already in the Resurrection of the Body, social and cultural history takes second place to the history of ideas. Now, in making her central concern change and identity, Bynum has moved towards the most abstract aspect of the theme she had already tackled. Certainly, the philosophical nature of the subject she has chosen does not preclude her from tracing it in a whole range of writing, but it does mean that, as an historian who aims to make the ways in which people at the time felt and thought her starting point, Bynum should not ignore the technical apparatus developed in the schools and universities to think about change and identity. The theologians of the schools and universities, such as Peter the Lombard, H u g h of St Victor, Roger Bacon and Thomas Aquinas himself, do indeed appear quite frequently in her pages. Bynum recounts their views about particular issues--for example, whether an- Marenbon 277 gels can take on bodily shape, but she does not look into the framework of concepts they used to think about change. There are occasional, general references to Aristotle's On Generation and Corruption, but nothing detailed enough to introduce readers to the variety of complex theories about change which were proposed in the context of the different Aristotelianisms of the twelfth and thirteenth centuries. Thinkers of this long period shared a general broad understanding, according to which individual things belonging to a natural kind (this man or that horse, for instance) were each seen as a substance of a particular sort (human substance, equine substance) which was affected by various accidents, such as quality, quantity, relation, place and time. Suppose Socrates is sitting in the forum at sunset (to elaborate on a favourite medieval example), then we can speak of a substance (in this case, an individual human being) with accidents of posture, place and time---sitting in the forum at sunset--and other accidents, such as being, let us say, light-brown, five-foot tall and sad, perhaps, to contemplate the close of another day. Socrates is for ever changing in his accidents, as when he stands up cheerfully to greet the friend who approaches, or his skin becomes whiter in wintertime or he shrinks a little as he ages. Through all these losses and gains, the substance that by which he is a h u m a n being--remains the same. But one day a different, more drastic change will affect Socrates. He will die, and what was until then a living, ensouled human body--that is to say, the individual h u m a n substance Socrates--will become a different sort of thing, a corpse. This substantial change (something akin to Bynum's radical change) is still change, because there is an unchanging substrate, matter. Medieval Christian theologians (especially in the later Middle Ages), but not Aristotle, also allowed, in principle, for a change even more drastic, a change so drastic that it would not be change at all. God, though no other being, might decide to annihilate Socrates, so that nothing would remain. Bynum is right to see Aristotle's On Generation and Corruption as one of the sources for this general conception of change, but the development of Aristotelianism in the twelfth and thirteenth centuries is much more complex than she suggests. On Generation and Corruption was not in common use until the mid-thirteenth century, but a somewhat different version of the Aristotelian metaphysics of substance and accident had been known for four centuries already, mainly through Aristotle's logical works. 7 Twelfth-century thinkers were fascinated by the various paradoxes which arise in connection with the change from human being to corpse, and they explored these in the context of Aristotelian logic. They also touched on the very paradox which underlies the discussion of radical change: that, in a sense, nothing can change, because if A is changed, it is now something different, B, and so A has not changed, but has been replaced by B: the nominales, followers of Peter Abelard, defended as one of their characteristic theses the view that nothing grows. 8 Yet, unlike their thirteenth-century successors, they knew very little about Aristotle's theory of the soul as the form of the body; they tended to view the soul, rather, in Platonic terms, as a substance distinct from the body. Among the questions, unasked by Bynum, that her topic raises, is whether knowledge of Aristotle's theory about the soul altered views about change; to what extent and how writers such as those Bynum mostly studies--highly educated but in most cases not actually working as school or uni- 7. 8. See J.M.M.H. Thijssen and H.A.G. Braakhuis (eds.), The Commentary Tradition on Aristotle's 'De generatione et corruptione': Ancient, Medieval and Early Modern, Studia artistarum 7 (Turnhout: Brepols, 1999). See Christopher J. Martin, 'The Logic of Growth: Twelfth-Century Nominalists and the Development of Theories of the Incarnation', Medieval Philosophy and Theology 7 (1988), pp. 1-15. 278 International Journal of the Classical Tradition /Fall 2004 versity masters--used the Aristotelian distinctions as a basis for or a w a y of conceptualizing their more intuitive thoughts about identity and metamorphosis; and what alterations Christian requirements (the need to allow God absolute power, even to annihilate; the special Eucharistic change, so important for theologians) demanded of the Aristotelian picture. Superficially, Bynum's earlier monograph on The Resurrection of the Body might seem proof against such criticisms, since Aquinas' Aristotelianizing account of spiritual bodies features as one of its central ideas. Yet a closer look shows Bynum unwilling to probe these arguments of Aquinas--in the manner undertaken recently, for example, by Robert Pasnaug--to look into their metaphysical context in any depth or to give a sophisticated idea of the many shades of opinion, often linked to differences on fundamental ontological problems, about this topic among academic theologians in the thirteenth and early fourteenth centuries. At the basis of the difficulty, there might seem to be twin-stories of Two Cultures. Bynum, an admirably wide-ranging and inquisitive m o d e m intellectual who, as already observed, is familiar with anthropological and sociological thinking, feminism and the whole range of discussions embraced in literature and humanities departments under the title of 'theory', manifests not the slightest curiosity for what is going on in the office of the Department of Philosophy just down the corridor. Similarly, the analytical philosophers who are gathered there are unlikely to have much interest either in Bynum or in any part of the whole range of m o d e m thinking that stimulates her. Turn back eight centuries and there seems to be a similar split, between the logically-trained, highly technical theologians of the schools and universities, and a wider b o d y of more or less educated writers, steeped in a wider humanistic culture. And so is it not only natural that modern humanists should look back to their humanistic predecessors, and medieval scholastic subtleties should be left to the analytical philosophers, supposing any of them should have the historical bent to consider such material? But, in fact, the parallel is misleading. The historian of contemporary thought, seeking to understand Bynum, needs to know very little about what is going on in the Philosophy Departments, except perhaps their isolation from the other arts and humanities. The historian of medieval thought, however, can afford no such narrowing of his perspective when he tries to understand strands in intellectual life other than the technical theology of the schools and universities, because there was no split, like the modern one, between an analytical philosophy and theology and other, less rigorous ways of thinking. Even an extreme example such as Bernard of Clairvaux extreme, that is to say, in his antipathy for the schools--was conversant, if in a befuddled manner, with the technical, logical arguments about doctrine proposed b y his arch-enemy, Abelard. John of Salisbury, the twelfth-century humanist, was taught by the leading logicians of the day (Abelard included), and Dante himself had acquired, in some w a y or another, a variety of complex scholastic learning. There was was no split in the Middle Ages between technical scholastic theology and philosophy, and a wider, more literary and imaginative approach to the same problems, in part because to a great extent anyone who was being educated followed a similar syllabus, and in part because--as Bynum's own recourse to the scholastic theologians illustrates--the role of theology faculties in defining and supporting Church doctrine meant that their highly sophisticated and abstract philosophizing had a social importance which few modern analytical philosophers would dream (or per- 9. ThomasAquinas on Human Nature (Cambridge: Cambridge University Press, 2002), pp. 380-93. Marenbon 279 haps even desire) for their speculations 9Bynum, then, for all her width, is to an extent guilty of imposing on her medieval sources a certain modern narrowness, which sets specialists in literature and history apart from their colleagues in analytical philosophy. Bynum's Challenge This criticism comes, of course, from someone whose own specialism is the technical medieval philosophy and theology which Bynum is being accused of treating superficially 9 And its source betrays its weakness. Who could be more guilty of imposing a modern narrowness on medieval thinking than those who view the thought of the period from the perspective of analytical philosophy? Yet, apart from this approach, what work is going on (in English, at least), except for the type of desiccated historical scholarship that rates accuracy above understanding and ends by providing what is fundamentally no more than a meaningless inventory? Moreover, B y n u m - - a n d a constellation of other historians of late ancient and medieval culture and religion (Peter Brown and Jacques Le Goff are examples)--have won themselves an audience: they have made people want to find out about medieval ways of living, feeling and worshipping, not just out of curiosity, and not to imitate, but as something from which to learn. By contrast, as it has been pithily put, "II n'est rien de plus inactuel que la philosophie mddidvale'--nothing is further removed from current concerns, nothing matters less to us, than medieval philosophy. 1~ Cultural historians do start with an advantage over historians of philosophy 9Precisely because their evidence is not shaped into rigorous arguments, they are able both to respect its medieval setting and yet shape the questions they ask of it according to a modern set of interests 9Scholastic discussion of a point of metaphysics does not easily lend itself to such treatment. To do such arguments justice an interpretation must enter into their terms and details in a way that forces it into a medieval mould, and away from modern intellectual concerns. The apparent kinship between some techniques and topics of analytical philosophy and scholastic medieval ways of thinking seems to offer a path of escape from this trap, but it is a road which leads often to the distortion of the arguments in the text (the kinship being merely apparent), and even at best it brings those who follow it to a disappointing destination, the ghetto inhabited by the tiny group of those analytical philosophers who will interest themselves in such material 9 In so far as Bynum's interests have developed so as to take her into the history of medieval philosophy and scholastic theology, she faces to some degree the same problem as the specialists 9Can she bring the same qualities she has used to understand medieval social history to this less amenable material? At the end of her essay on 'Wonder', Bynum uses the concept of admiratio she has found in her medieval sources to characterize an historiographical ideal: 9 we write the best history when the specificity, the novelty, the awe-fulness, of what our sources render up bowls us over with its complexity and its signific a n c e . . . For the flat, generalizing, presentist view of the past encapsulates it and 10. The comment was made by Pierre Alf6ri, who has written on William of Ockham. It is quoted by Alain de Libera (Penserau moyen ~ge [Paris: Editions du Seuil, 1991], p. 56), who is one of the small band of specialists in medieval philosophy to have faced up and reacted to the challenge this comment presents. 280 International Journal of the Classical Tradition / Fall 2004 makes it boring, whereas amazement yearns toward an understanding, a significance, that is always just a little beyond both our theories and our fears. (pp. 74-75) Bynum has shown her readers h o w to wonder constructively at many aspects of medieval experience. The challenge for her, as she is led by her interests to the more explicitly intellectual aspects of the Middle Ages, and the challenge for all w h o specialize in this field, is how, here too, we can learn to move from our initial bafflement not towards a boring encapsulation, but into amazement. John Marenbon Trinity College Cambridge The Letters of Erwin Panofsky Erwin Panofsky, Korrespondenz 1910 bis 1968. Eine kommentierte Auswahl in fiinf Biinden, hrsg. von Dieter Wuttke (Wiesbaden: Harrassowitz Verlag): Bd. I, Korrespondenz 1910 bis 1936 (2001), LIII + 1142 pp; Bd. II, Korrespondenz 1936-1949 (2003), XXVIII + 1363 pp. Erwin Panofsky was born in Hannover in 1892, the only child of a well-to-do businessman who had sold the family lime-burning interests in what was then Upper Silesia and retired at the age of forty-two, a couple of years before his marriage. When Erwin was ten years old, the family moved to Berlin, where he had his subsequent early schooling, with a good grounding in Greek and Latin. A brief experience as a student of law in Freiburg i. B. in 1910 was transmuted into the preoccupation with the history of art which was to last until his death in 1968 at just short of seventy-six in Princeton, N e w Jersey. The two volumes under review pass lightly over his early years, to bring into focus his transformation from the precocious Dr. phil. and brilliant young Privatdozent to widely recognized pre-eminence in Germany, before exile made it necessary to reachieve that preeminence in another country and another language. The initial phases of Panofsky's university life were passed chiefly in Freiburg; he never ceased to acknowledge the importance for him of Wilhelm V6ge, his Doktorvater, and of Walter Friedl~inder, who was then making his Habilitation there. His early experience in Berlin, which had become his parental home-town, might on the other hand be seen as equally influential, if not even more decisive. From Adolph Goldschmidt's Berlin seminar he was despatched to Aby Warburg in Hamburg and his Institut f~r Ausdruckskunde orf//r methodologische Grenzidberschreitung; when he visited it for the first time late in 1915, he had already supplied Warburg with learned references. The same seminar also introduced him to Dora Mosse, Berlinerin and art-historian, some years his senior, w h o m he married in 1916. When he was called to Hamburg at the end of 1919 by Gustav Pauli, Director of the Kunsthalle and honorary professor in the new University, their two sons, both of them later distinguished scientists, had already been born. Panofsky's contact with Warburg was followed up in correspondence from 1916 onwards; from the end of 1919, there is a tactful and sympathetic, but tragically premature, letter of good wishes on recovery from the mental illness which was to dog Warburg for