Introduction
Transcendentalism and the Rise and Fall of Metaphor
Twentieth-century philosophy is fascinated by the phenomenon of language. Russell and the logical positivists saw formalized language as the logical matrix for all our knowledge of the world. And they argued that it would be the philosopher's task to reduce language by formal analysis to its logical core and, furthermore, that a thorough analysis of that logical core would show us how all reliable (i.e., scientific) knowledge is built out of its elementary, atomistic constituents. Carnap gave the logical positivist's thesis a polemical edge when he added that metaphysics—and metaphysics embraced, in his view, the greater part of Western philosophy—originated in the philosopher's ignorance of the proper syntactic rules for the logical constitution of the world. Hence, logical analysis, as advocated and practiced by the logical positivists, would dispel most of the problems that had been discussed in the tradition of Western philosophy. Thanks to a logical analysis of the language that had been used for stating these problems, they would not be solved in the proper sense of the word but would be shown to be pseudoproblems.
At a later stage, the Wittgenstein of the Philosophical Investigations, Ryle, Austin, and many others would turn to the social dimension of language; for Wittgenstein, the many different languages we use on different occasions could best be compared to the playing of a game. Playing a game requires that all the players involved accept the rules of the game, and it would be no different with the speakers of a language. Language was no longer a logical calculus but a social practice. And natural language henceforward replaced formalized language as the proper focus of philosophical interest. Carnap's rejection of metaphysics now was abandoned in favor of Strawson's peculiarly Nietzschean thesis that the most general syntactical structures of natural languages determine the metaphysical struc-
ture of our world.[1] Thanks to this sociological (or, as is ordinarily said, linguistic ) turn, philosophy now received the task of developing a descriptive metaphysics that would account for these metaphysical structures of the world.
But what all these philosophies of language had in common—despite their many diversities or even outright oppositions—was the assumption that language is the principal condition for the possibility of all knowledge and meaningful thinking, and that therefore an analysis of language is of as much importance to the contemporary philosopher as an analysis of the categories of the understanding was for the Kant of the first Critique. Precisely because of this obvious similarity it has often been pointed out that contemporary philosophy of language can best be seen as a new and more fruitful phase in the transcendentalist program that was inaugurated two centuries ago by Kant.
Two intimately related assumptions underlie contemporary philosophy of language. (I hasten to add that these two assumptions are of primary importance merely from the point of view of what I want to say in this introduction; it certainly is not my wish to make any general claims with regard to the practice of philosophy of language.) The first assumption is a methodological one that harks back to the so-called resoluto-compositional method that was adopted by early modernist philosophers like Descartes and Hobbes.[2] This method requires us to divide complex problems into their simpler components. It is recommended that the philosopher start with the simpler problems and then slowly and carefully work his way up to the larger and more complex issues. The "assumption behind this assumption" is that nothing essential to the larger and more complex issues will be lost when this method is applied. The acceptance of the resoluto-compositional method in the practice of contemporary philosophy of language resulted in the almost universally shared conviction that philosophy of language ought to start with an investigation of the behavior of logical constants, proper names, et cetera, and of the meaning of words and propositions. Obviously this assumption must have an elective affinity with the logical atomism that was described at the beginning of this introduction. Hence, though logical atomism as a philosophical position has been discredited for over half a century, contemporary philosophy of language
[1] P. F. Strawson, Individuals: An Essay in Descriptive Metaphysics , London, 1971.
[2] Descartes proposed four rules for the discovery of truth. The second ran as follows: "de diviser chacune des difficultés que j'examinerais, en autant de parcelles qu'il se pourrait, et qu'il serait requis pour mieux les résoudre" (to divide each of the problems that I would investigate into as many parts as possible and as would be required for better solving them). (R. Descartes, Discours de la méthode, Paris: Flammarion, 1966, 47)
is still "atomist" as far as its method is concerned. We are confronted here once again with the peculiar paradox that philosophers, always so focused on methods used in other disciplines, are largely indifferent to their own methods and their implications.[3] In short, with the help of an investigation of propositions (either singular or universal) and their constituent components, or of simple transparent conjunctions of propositions, philosophers of language hoped to discover the transcendental conditions for truth and meaning.
As soon as this method is accepted it will not be hard to appreciate the plausibility of a second assumption of twentieth century philosophy of language. According to this second assumption, the problem of how language might account for a complex reality in terms of texts rather than of individual propositions (the professional concern of the historian!) is regarded as a nonproblem; that is to say, one was unwilling to expect problems here that would not be reducible to the kind of problems encountered in the analysis of propositions and their parts. Most of the fortunes and misfortunes of contemporary philosophy of history can be explained from this perspective. With regard to the misfortunes, it must be pointed out that philosophers of history were often tempted to superimpose this assumption on philosophy of history. Thus in the fifties and sixties, philosophy of history preferred to focus on the elements of the historical text, like singular statements about historical states of affairs, statements expressing causal connections, or on the temporal perspective of statements about the past (Danto's "narrative sentences"). The historical text as a whole was rarely, if ever, the topic of philosophical investigation. This is all the more to be regretted since the fortunes of philosophy of history self-evidently lie with the historical text and not its parts. Only a philosophy of history concentrating on the historical text as a whole could contribute importantly to contemporary philosophy of history and go beyond a mere application of what had already been discovered elsewhere. History is the first discipline that comes to mind if we think of disciplines attempting to give a truthful representation of a complex reality by means of a complex text. Hence, what is so interesting about the historical discipline is that it so clearly suggests the limitations of the resoluto-compositional method. Considered from this perspective philosophy of history could have provided philosophy of language with a wholly new departure, resulting in a philosophy of language that would pose new and interesting problems, both unstatable and unsolvable within the parameters of existing philosophy of language. In this way Collingwood's prophecy—that it would be the main business of twentieth-century philosophy to come to terms with
[3] See chapter 5 of this volume.
twentieth-century history—could be realized.[4] History would then be as important to contemporary philosophy as science was to seventeenth- and eighteenth-century philosophy. (As we shall see below, giving philosophy of history this task does not in the least imply that philosophy of history should be pitted against philosophy of science, or history against science.)
It is sad that philosophers of history are still disappointingly reluctant to meet this challenge. If philosophy of history presently is in such poor shape that one might well ask whether it still exists at all, this has much to do with the unwillingness of philosophers of history to explore the philosophical gold mine that is their exclusive possession. Two factors may (partially) serve to explain this reluctance. First, philosophers of history in the recent past have tended to downgrade the significance of the distinction between historical research (the results of which are typically expressed in terms of individual statements about the past) and historical writing (which has integrated the results of historical research within the whole of the historical text) and refused to attribute to the latter a certain autonomy and independence with regard to the former. The result has been that most philosophy of history has been a philosophy of historical research. The thesis of the theoryladenness of empirical facts has most often justified the rejection of the distinction. Needless to say, insofar as the integration of the results of historical research in historical writing does not merely aim at a confirmation or a reproduction of the relevant theories determining description, this thesis will inevitably fail to justify the rejection of the distinction.
But, more important, it can be shown that texts logically differ from (individual) propositions and that, consequently, historical writing (on a par with the historian's text) can never be completely reduced to (the resuits of) historical research (on a par with individual propositions about historical states of affairs). For suppose we have a text on, for example, the French Revolution. We should note, then, that it would be impossible to clearly distinguish between those elements in the text that refer purely to the French Revolution without describing it and those elements that ascribe certain features to the French Revolution without referring to it. There is no clear border between these two, and it might even be argued that the referential elements completely coincide with what is ascribed to the purported object of reference.[5] Here, then, we observe what texts have in common with paintings from a logical point of view. If we look at Goya's painting of the Duke of Wellington, it is no less impossible to distinguish between what merely denotes the Duke and the features that Goya wished
[4] R. G. Collingwood, An Autobiography, Oxford, 1970, 79.
[5] See my "Statements, Texts and Pictures," in F. R. Ankersmit and H. Kellner, eds., The New Philosophy of History , London, 1994 (forthcoming).
to ascribe to him with his painting. In the case of statements, however, the distinction is entirely unproblematic and coincides with the functions of the subject-term and the predicate-term of the statement. Thus, we discover here the insurmountable barrier between the statement on the one hand and the historical text or the painting on the other. And because of this logical barrier the necessity of distinguishing between historical research and historical writing cannot be doubted. It follows from this, in turn, that something essential will he lost if we reduce the historical text as a whole to its constituent parts, as the resoluto-compositional method would have us do.
The second factor that may help to explain the reluctance of philosophers of history to develop a philosophy of language devoted to the historical text is the following: Since philosophy of language did not provide the philosopher of history with usable insights, the most obvious strategy was to turn to literary theory. For, since literary theory was accustomed to dealing with texts as a whole (i.e., novels) it seemed reasonable to expect that some intellectual instruments could be found here that might help the philosopher of history to analyze the historical text. Yet one may justifiably have one's doubts about literary theory as a surrogate for this (non-existent) kind of philosophy of language. It is a bad omen that narratology, as developed by Genette, Bal, and others—that piece de résistance of contemporary literary theory—has done little to further our understanding of historical writing.[6] Analogies to the literary devices used in the novel—and of professional interest for the literary theorist—can undoubtedly be found in historical writing (no one could deny this), but this is insufficient for justifying the claim that literary theory will substantially deepen our insight into historical writing. For, despite such analogies, the aims and effects of literary narrative do not necessarily coincide with those of historical narrative. Think, for example, of Zola's Les Rougon-Macquart cycle and let us assume, for the sake of argument, that the cycle does indeed give a correct picture of social life in France under Napoléon III. In this case we might decide to read the cycle if we want to be informed about social life in that period. But it will be obvious that the information we are looking for is presented in the cycle in a different way than in, for example, Zeldin's History of France (perhaps a somewhat eccentric example, I admit). The cycle would require a specific kind of reading: we would have to read the cycle in such a way that the relevant knowledge could be deduced from the cycle—whereas it is the pretension of history books to present their readers with that kind of knowledge in a straightforward way. The
[6] Illustrative of the unsatisfactory relationship between narratology and the writing of history is D. E. Polkinghorne, Narrative Knowing and the Human Sciences , Albany, 1988, chap. 3.
difference is analogous to that between the clue for a word in a crossword puzzle (the novel) and the intended word itself (history). And naturally this difference must have its consequences for the narrative organization of either novel or historical text.
What has been said up till now provides us with the appropriate background for assessing the achievements and merits of Hayden White's historical theory. A moment ago we observed that narrativism (the term I shall henceforth use for referring to a philosophy of language analyzing the historical text as a whole) is only possible if the distinction between historical research and historical writing is recognized and respected. And this certainly is the case in White's historical theory. Though White concedes that it may be difficult and sometimes even impossible to distinguish between fact and interpretation,[7] a recurrent theme in his earlier work is his insistence on the cognitive gap between annals and chronicles on the one hand and historical texts, in the proper sense of the word, on the other. He thus urges us "to confront the conventional, but never fully analyzed, distinction between 'mere' chronicle and the history properly so-called."[8] His argument is that we are mistaken in believing that history is simply hidden in the facts and that telling the story is merely a matter of making explicit what is already there. But telling a story (or writing a history) is a construction we impose on the facts. This is already the case at the basic level of our personal life ("no one and nothing lives a story,"[9] writes White), and when we turn to the abstract entities that are the topic of the historian, the autonomy of the text or story (historical writing) with regard to the facts (historical research) is even more obvious.
White adds to his defense of the distinction in question the paradoxical insight that the abandonment of medieval chronicles, for the kind of historical writing that we are now accustomed to, was accompanied by a loss of certainty. For there would be no point in doubting the annalist's statement that in 732 "Charles fought against the Saracens at Poitiers on Saturday," however naive and imperfect that statement may be,[10] whereas it is quite hard to argue convincingly why, for example, Furet's account of the French Revolution is better or more adequate than that given by Labrousse or Soboul. In a certain sense, therefore, the coming into being of modern narrativist (or historist[11] ) historical writing could be regarded
[7] H. White, Tropics of Discourse , Baltimore, 1978, 107.
[8] White, Tropics, 111. See also White, Tropics , 91, and H. White, The Content of the Form , Baltimore, 1987.
[9] White, Tropics , 111.
[10] Quoted in White, Content, 7.
[11] Narrativism could well be seen as a nominalist version of classical historism. This has been one of the main theses of my Narrative Logic: A Semantic Analysis of the Historian Language , The Hague, 1983. I add here a note on the terminology that will be adopted in this collection. The term historism will be used to refer to the kind of historical theory that has been developed by Ranke and Humboldt, for example, and whose main theoretical statements are collected in G. G. Iggers and K. von Moltke eds., The Theory and Practice of History , New York, 1973. With Mandelbaum, "historism" can be defined as "the belief that an adequate understanding of the nature of any phenomenon and an adequate assessment of its value are to be gained through considering it in terms of the place that it occupied and the role which it played within a process of development." See M. Mandelbaum, History, Man & Reason , Baltimore, 1971. "Historism" as understood in this way is, therefore, by no means identical to, historicism , which, in Popper's sense of this term, refers to conceptions of history aiming at predicting the future. Speculative philosophies of history are historicist and the gap between historism and historicism is as deep as the one between modernism and postmodernism. In fact, as is argued in the last chapter of this collection, postmodernism relates to modernism as historism relates to the Enlightenment.
as a movement against truth—as a movement, that is, that invites us to risk ourselves outside the safe sphere of the Pyrrhonist truths of the annalist and the chronicler and to enter the more interesting but also less certain world of historical writing. This paradoxical movement against truth requires our special attention since it supplies us with another argument in favor of the distinction between historical research and historical writing. For if the latter were merely an extension of the former, why should historians risk entering that dangerous territory at all? Clearly they only do so because insights can be gained here that historical research will never be able to give.
But undoubtedly an appreciation of White's use of literary theory will be of more direct importance for an assessment of his place in contemporary philosophy of history. Because of White's versatility and the impressive theoretical scope of his writings this is a vast subject and I necessarily restrict myself to what is of relevance from the perspective of this introduction. The crucial datum from that perspective is that an interesting ambivalence can be discerned in White's use of literary theory. This ambivalence can be elucidated in two steps in the following way. We should first note an ambivalence with regard to what is the real issue in his (early) work. If we consider Metahistory we might, at first sight, be inclined to say that what White offers there is a theory of historical writing in the proper sense of the term. However, Metahistory is not primarily a book about how historical truth can be attained and tested, et cetera (the main preoccupation of philosophy of history in the fifties and sixties), but about how we should read history books. It was part of White's enterprise to read the great texts of nineteenth-century historians as if they were novels—something no theorist had ever done before.[12] And by doing so he created—together with Lionel Gossman, whose studies on Michelet and Thierry
[12] H. V. White, Metahistory: The Historical Imagination in Nineteenth-Century Europe , Baltimore, 1973. Roland Barthes's book on Michelet anticipated in several respects White's Metahistory.
drew on similar inspirations—a new and exciting form of historiography that was unlike anything that had previously been done in the field. The books written by authors like Bann, Kellner, Orr, Partner, and others can be located within this disciplinary matrix for a new historiography that was created by the joint effort of White and Gossman.[13]
Since historiography answers historical instead of philosophical questions, it might seem that Metahistory had no bearing on the kind of topics discussed by philosophers of history. Nevertheless, Metahistory did also imply a theory of history in the traditional sense; as is already clear from the different kinds of reactions to Metahistory, the book is a theory on historical representation as much as one on how to proceed in historiography. It is true that the major theses of this theory (one may think here of White's thoroughgoing relativism, his advocacy of a linguistic turn for historical theory, and the way he argued his views) pointed toward a new phase in the history of historical theory; but as such these theses undeniably fell within the scope of what traditionally was perceived as the task of philosophy of history. Hence, Metahistory was ambivalent in that it tended to render historiography more philosophical and philosophy of history more historiographical; the borderlines between the two disciplines were effectively blurred. Self-evidently, the qualification can or even must be added that we can only speak of this ambivalence of Metahistory if we take for granted the regime of disciplines antedating Metahistory. White might, for good reasons, reject the label of ambivalence and claim coherence for Metahistory, while, at the same time, criticize as schizophrenic the position from which Metahistory is perceived as "ambivalent." It would not be difficult to account for this chiasm on Kuhnian terms. And this would enable us to appreciate the "revolutionary" character of White's work.
But a more interesting and important ambivalence in White's early work has its source elsewhere, namely in his theory on the roles of the tropes in historical writings. As is well-known, according to Metahistory, historical writing is always informed by one of the four tropes—metaphor, metonymy, synecdoche, or irony. And this confronts us with the ambivalence in White's early theory of history that most deserves our attention within the framework of this introduction. For, on the one hand, this theory of the tropes undoubtedly achieved a rapprochement between history and literature: the use of figurative language is what both have in common. And this is also why Metahistory was so severely criticized by most commentators. White could simply not be right, it was argued, since his theory of the tropes left no room for notions like the truth and testability
[13] For an exposition of the similarities and the differences between White and Gossman, see my review, "Lionel Gossman, Between History and Literature," CLIO 21 (1992): 175-176.
of historical writing and thus seemed to inspire disrespect for the cognitive responsibilities of the historian himself. Tropology seemed to blow White's ship out of the safe port of the sciences and onto the treacherous seas of literature and art. Metahistory transformed historical writing into literature.
On the other hand, we should recall that Max Black already recognized the significance of metaphor for the sciences; and Mary Hesse even went so far as to assert that concept-formation in the sciences is essentially metaphorical.[14] In other words, when focusing on tropology White happened to single out precisely that aspect of historical writing which one, for good reasons, could say is shared by literature and the sciences. This, then, is the ambivalence in White's theory that will preoccupy us for the remainder of this introduction: surely Metahistory inaugurated the swing of historical theory toward literature, yet it managed to do so in such a way as not to preclude a scientistic interpretation of historical writing. One might object, at this stage, that the ambivalence is merely apparent. That is to say, arguments like those of Black and Hesse should not be interpreted as an indication of the scientism of metaphor (and, hence, of White's tropology) but rather as an indication that philosophers of science are now prepared to recognize "literary" elements even in the sciences. What we see in the arguments of Black, Hesse, and White is, it might be said, an unequivocal agreement to move away from science and toward literature. Certainly this objection makes sense. However, in reply to this objection, I now want to point out that in White's own view tropology does not necessarily mean a radical break with science and scientistic cognitive ideals and, moreover, that an independent argument can be conceived to show that tropology even lies at the heart of these scientistic cognitive ideals.
First, with regard to White's own relevant declarations, one of his characteristic assertions is that the only instrument "the historian has for endowing his data with meaning, of rendering the strange familiar (my emphasis), the mysterious past comprehensible, are the techniques of figurative language."[15] The implication is, clearly, that historical insight and meaning are only possible thanks to the use of the tropes, and that therefore,
[14] M. B. Hesse, "Models, Metaphors and Truth," in F. R. Ankersmit and J. J. A. Mooij, eds., Knowledge and Metaphor , Dordrecht, 1993.
[15] White, Tropics , 94. Elsewhere, White writes:
Understanding is a process of rendering the unfamiliar, or the "uncanny" in Freud's sense of that term, familiar; of removing it from the domain of things to be "exotic" and unclassified into one or another domain of experience encoded adequately enough to be felt to be humanly useful, nonthreatening, or simply known by association. This process of understanding can only be tropological in nature, for what is involved in the rendering of the unfamiliar into the familiar is a troping that is generally figurative (White, Tropics , 5).
For further elucidation see White, Tropics , 86, 88.
precisely tropology can show us how the discipline of history truly is part of the Western, Faustian effort to conquer cognitively the physical and the historical world we live in. In one word, tropology is for history what logic and scientific method are for the sciences. White is even quite specific about how the common ground between history and the sciences has to be defined. He thus hazards the suggestion that each of the four tropes corresponds to one of the four stages that Piaget discovered in the cognitive development of the child. And as this cognitive development is conditional for the possibility of doing scientific research, so the tropes are conditional for the possibility of historical meaning and insight. What is of special interest in this suggestion is the following: As is well known, Piaget's description of the cognitive development of the child is in many ways similar to and to some extent even inspired by Kant's transcendental analysis of the human mind, as expounded in his first Critique. I am convinced that the link between tropology and Kantian transcendentalism that is thus hinted at should be taken quite seriously. Indeed, this seems to be in conformity with White's own explicit intentions: he incidentally compares his own tropology with the Kantian enterprise.[16] And still more illustrative of the Kantian character of White's tropology is how White sums up the aims and purposes of his magnum opus:
One must try to get behind or beneath the presuppositions which sustain a given type of inquiry [i.e., history] and ask the questions that can be begged in its practice in the interest of determining why this type of inquiry has been designed to solve the problems it characteristically tries to solve. This is what metahistory tries to do. It addresses itself to such questions as, What is the structure of a peculiarly historical consciousness? What is the epistemological status of historical explanations, as compared with other kinds of explanation that might be offered to account for the materials with which historians ordinarily deal? What are the possible forms of historical representation and what are their bases?[17]
Think, furthermore, of the obvious similarities between the way in which the tropes organize historical knowledge and how the manifold of human experience is organized by the Kantian categories of understanding. What becomes clear from all this is that we would be justified in attributing to White (no less than to Dilthey) the wish to develop a quasi-Kantian critique of historical knowledge and to closely associate his own theory of history with that impressive culmination point of Western scientistic thought. Apart from the literarization of historical writing Metahistory is no less an endeavor to provide us with a quasi-Kantian, epistemological investigation of
[16] White, Tropics , 22. The analogy with Piaget is developed here.
[17] White, Tropics , 81.
the cognitive foundations that support historical representation and meaning. This, then, is the ambivalence in tropology that no reader of White's earlier work can afford to ignore.
I shall now proceed to consider more closely the equivalence or, at least, the close relationship between Kantian transcendentalism on the one hand and White's tropology on the other. I shall thereby restrict my exposition to metaphor. I am aware that this restriction is not without its problems. Several writers in the recent past have stressed the profound difference between the individual tropes, and it has even been argued that the contrast between modernism and postmodernism coincides with the difference between metaphor and irony.[18] However, it suffices to note here that this is not how the tropes function in White's Metahistory. Here the tropes all have comparable cognitive functions—a fact that is reflected by White's insistence that even an internal logic can be discerned in the sequence of the tropes that more or less naturally will lead us from one trope to another (including irony). Within the whole of White's tropology, we will nowhere encounter an insurmountable barrier separating one (or more) tropes from the others.
When focusing exclusively on metaphor, then, the first thing to be noted about metaphor is the following: By proposing that we see one thing in terms of another, metaphor is essentially equivalent to the individuation of a (metaphorical) point of view, from which we are invited to see part of (historical) reality (such is the theory of metaphor that will be adopted throughout this volume).[19] For example, the metaphor "the earth is a spaceship" invites us to see the earth from a point of view that is defined by the interaction (to use Black's terminology) between the concepts earth and spaceship. It should be observed, next, that this view of metaphor is in fundamental agreement with the main inspiration of Kantian transcendentalism—and this is why metaphor is a continuation of scientistic cognitive ideals, rather than being in opposition to them. Or, to be more precise, there are two similarities that we must bear in mind in this connection.
First, the two function, cognitively speaking, in a similar way. For both enable us to organize (our knowledge of the chaotic manifold of) the world. Both the transcendental subject and the metaphorical point of view do this organizing by withdrawing themselves from the world that is organized by them. Think of how Kant defined the transcendental subject. On
[18] 1. Hassan, The Postmodernism Turn: Essays in Postmodernism Theory and Culture , (Ohio State University Press, 1987), 91-92.
[19] See my Narrative Logic , 209-220; and my "Reply to Professor Zagorin," History and Theory 29 (1990): 275-297.
the one hand, the transcendental self organizes the chaotic manifold of noumenal reality into a phenomenal reality that is accessible to our understanding. But, on the other hand,
the transcendental self remains forever an unattainable entity, for through this I or me or it (the thing) which thinks, nothing further is represented than a transcendental subject of the thought = x. It is known only through the thoughts which are its predicates, and of it, apart from them, we cannot have any concept whatever, but can only revolve in a perpetual circle, since my judgment upon it has always already made use of its representation.[20]
Hence, what we can truthfully say about the transcendental subject only gives us access to its predicates but not to itself. Much the same can be said about the metaphorical point of view. In a brilliant and justly famous essay on what he referred to as the usure of metaphor in Western philosophy, Derrida demonstrated that the use of metaphor provides us with an intellectual or mental entity that functions both as an "organizing center" and as a "blind spot," that is, a spot that sui generis cannot be aware of itself.[21] In order to explicate Derrida's view of metaphor, let us start with the former part of his claim. Taking once more the metaphor, "the earth is a spaceship," as our example, it will be obvious that the metaphorical point of view calls for quite a specific organization of the knowledge we have of our ecosystem: the organization must be such that it clarifies the vulnerability of that system. As for the latter part of Derrida's claim, the point of view necessarily remains a blind spot to itself because of its incapacity to objectify itself (a quality it shares with the transcendental self). For, each attempt at objectification would temporarily require us to abandon the point of view. Points of view obey the logic of the center, and of the center we can say that it cannot be "decentralized" by looking at it from the perspective of another center, without robbing it of its defining characteristic of being a center. The conclusion follows that, cognitively speaking, the transcendental self and the metaphorical point of view fulfill identical functions. Transcendental philosophy is intrinsically metaphorical, and metaphor intrinsically transcendental.
But there is a second, less formal—but perhaps precisely because of this, even more important—similarity between transcendentalism and metaphor. I said above that White's tropology was, in White's own words, in harmony
[20] I. Kant, Critique of Pure Reason , trans. N. Kemp Smith, London, 1978, 331.
[21] J. Derrida, "White Mythology: Metaphor in the Text of Philosophy," in Derrida, Margins of Philosophy , Brighton, 1986, 228. Whereas my argument does not go beyond the (already quite substantial) claim that transcendental philosophy is essentially metaphorical, Derrida makes the more comprehensive claim that all philosophy is metaphorical. See also my "Davidson en Derrida over de metafoor," in R. T. Segers ed., Visies op cultuur en literatuur , Amsterdam, 1991, 221-229.
with the Enlightened, Faustian endeavor of the knowing subject to "appropriate" the world or to "familiarize" what is initially experienced as strange and unfamiliar in that world. This striving to familiarize the unfamiliar, to make us "feel at home" in this world, to effect a stoic oikeioosis ("attempt to make the unfamiliar familiar"), can also be attributed to both transcendentalism and metaphor. Transcendentalism poses no problems here: the transcendental self transforms noumenal reality into a phenomenal reality that has adapted itself to the structure of the transcendental self. Reality is thus "appropriated" by the transcendental self. Next, the familiarization or appropriation of reality by metaphor is no less pronounced and can even be said to be the actual purpose of metaphor. If we recall our example of metaphor, it is clear that the metaphor is directly related to our attempts to protect the ecosystem against pollution by industry, stockbreeding, transport, and so on. In other words, the metaphor organizes our knowledge of these aspects of our world and it does this in such a way as to enable us to make this world a better and safer place for both ourselves and our children. The metaphor tells us how to fix up "our natural home" (recall the stoic notion of oikeioosis just referred to).
Speaking generally, metaphor has been remarkably effective in organizing knowledge in ways that may serve our social and political purposes (and this also explains why the social and the political and, hence, the historical world is metaphor's favorite domain). Metaphor arguably is the most powerful linguistic instrument we have at our disposal for transforming reality into a world that is adaptable to human aims and purposes. Metaphor "anthropomorphizes" social and sometimes even physical reality and, by doing so, enables us to appropriate and to become familiar with that reality in the true sense of these words. And finally, what is even more suggestive than metaphor's capacity to make an unfamiliar reality familiar is the fact that metaphor always invites us to see a less familiar system in terms of a more familiar system. Familiarization truly is the heart of metaphor.
Having arrived at this stage, we should note that there is a profound difference between White's earlier work and the essays that have been collected in The Content of the Form (1987), from the point of view of the present discussion. White does not explicate this difference himself, but his apparent change of mind is no less important for that. No reader of The Content of the Form can fail to be struck by the fact that the tropes are all but absent from it. A clue to White's change of mind can be found in what is, in my opinion, the most fascinating essay in the collection: the essay on what White refers to as "the politics of interpretation." White starts the essay by observing that the development of historical writing since the beginning of the last century can best be seen as a process of disciplinization—with all the Foucauldian connotations of that word. white suggests
that this process of disciplinization was far from being that discovery of cognitive innocence for which historians always so strenuously strive. It certainly did not permit historians, for the first time in the history of the discipline, to achieve a "realist" disclosure of the past "as it actually was"—though this undoubtedly had been the hope and expectation of all the historians and philosophers of history who have been involved in the process of disciplinization. To be more precise, it was hoped and expected that disciplinization would permit the historian to correct ideological and political distortions that were believed to be the major obstacle to the "realist" interpretation or representation of the past. White correctly demonstrates the futility of the effort. For, what this "political appeasement" amounted to, in practice, was the universal acceptance of a thoroughly anti-utopian historical writing. And one can say many good and positive things about anti-utopianism, but that it is an antipolitical position cannot possibly be maintained.
If, then, disciplinization of historical writing was realized in the last one-and-a-half centuries, and if disciplinization cannot be equated with depoliticization, we shall have to look elsewhere for what transformed history into a discipline. White now considers the suggestion that disciplinization ought to be identified with the attack on rhetoric by the founding fathers of history as a discipline. Eighteenth-century historical writing was still openly rhetorical, and, as Gossman has demonstrated on several occasions, the writing of history was conceived of as being part of the world of letters and literature; the quest for historical truth that was inaugurated in the nineteenth century, and which inspired the disciplinization of historical writing, required the abandonment of rhetorics and literary effect, since these were believed to stand in the way of historical truth.[22] However, though White does not say so himself in so many words, derhetorization will not be much more helpful than the option of depoliticization if we wish to grasp the nature of the disciplinization of historical writing. For, as depoliticization in practice resulted in the acceptance of a certain kind of political position, so derhetorization brought about the universal acceptance of a new, but nevertheless, merely different kind of rhetoric. Quoting Paolo Valesio, White speaks here of the "rhetoric of antirhetoric."[23]
So neither depoliticization nor derhetorization can explain the disciplinization of history in the course of the nineteenth century, and that brings us to White's third and decisive proposal for how to conceive of the disciplinization of historical writing. In conformity with the whole of his theoretical oeuvre, White focuses here once again on narrative and he be-
[22] L. Gossman, "Literature and Education," in Gossman, Between History and Literature , Cambridge (MA), 1990.
[23] White, Content , 66.
gins by pointing out that the historian must use his imagination if he wishes to integrate the results of his historical research into a historical text: imagination "is operative in the work of the historian at the last stage of his labors, when it becomes necessary to compose a discourse or narrative in which to represent his findings."[24] Needless to say, this is the kind of statement we might have found in Metahistory or in Tropics of Discourse. The same can be said of White's assertion that the imagination achieves narrativization by adopting a specific style and that, therefore, our question of the disciplinization of historical writing comes down to the question of "the nature of a disciplined historical style."[25] But when answering that question White makes a move that brings him outside the tropological framework. For, White now relates the issue of the disciplinization of history and of historical style to a dilemma that presented itself in eighteenth-century aesthetics as developed by Burke, Kant, Schiller, and Hegel.
The dilemma White has in mind here is that of the sublime and the beautiful. In aesthetic theory the beautiful was associated with "order," "sense, meaning and meaningful action"; the sublime, on the other hand, confronts us with what escapes and transcends our attempts to impose meaning and, therefore, in the words of Schiller, with "the terrifying spectacle of change which destroys everything and creates it anew and destroys again."[26] The terrifying spectacle of a continuous creation and destruction brings us to a realm that lies beyond our cognitive, historical, and political grasp and that successfully resists all our attempts to master it intellectually. To put it in the terminology that I have been using in this introduction, the beautiful is what can be intellectually appropriated by means of the tropes and lends itself willingly to our attempts at tropological appropriation; the sublime both escapes and even undoes our most strenuous efforts at appropriation. White himself uses the word domestication instead of appropriation and formulates the contrast between the beautiful and the sublime in the following way: Historical facts are "domesticated [in terms of the beautiful] precisely insofar as they are removed from displaying any aspect of the sublime that Schiller attributed to them in his essay of 1801."27 This, then, is what according to White essentially was at stake in the process of the disciplinization of historical writing: a striving for a taming, domestication, or appropriation of history by stripping the past from everything that might not fit into the tropological explanatory patterns that Western man has devised for making sense of sociohistorical reality. Obviously, we are tempted to exclaim that we could not possibly expect his-
[24] White, Content , 67.
[25] White, Content , 68.
[26] White, Content , 69.
[27] White, Content , 72.
torical writing to do anything else; what else could we expect from the historian's text than that it succeeds in making an unfamiliar past familiar to us? Put differently, how could we possibly refrain from making use of figurative language since metaphor and figurative language are our ultima ratio in the task of transforming the unfamiliar into the familiar? However, it is precisely this category of the sublime that reminds us that the tropological appropriation of the past is not the only option that is open to the historian: representation—and even historical representation—leaves the historian the possibility of presenting the terrifying strangeness and sublimity of the past to his readers.
I will not enter into a discussion of the plausibility of White's view that the disciplinization of historical writing mainly consisted in an exchange of the sublime for the beautiful in historical representation. Surely many historians who wrote during the centuries before the disciplinization of historical writing in the nineteenth century felt little affinity with the historical sublime. And yet, historians like Gibbon, Carlyle, and Michelet, who are in White's eyes the last great historians before the disciplinization of history took place, undoubtedly preserved an instinct for the historical sublime. It is true that after them the past became more common, more domesticated, more a variant of an eternal present. In the wake of institutional and socioeconomic historians, bourgeois rationality has been projected onto the strangest and remotest pasts, and hermeneutic theorists like Collingwood or Dray have offered a compelling and, to many historians and historical theorists, convincing justification of this effort at domestication. But the following is of more interest: When White favorably contrasts the historical sublime with the beautiful (and with tropology), this is self-evidently a move against tropology, but, at the same time, it is a move merely within but not against Kantian transcendentalism. For, as White amply points out himself, the sublime still has its logical place within the schematism of the Kantian system. Within the overall architecture of Kantian criticism the beautiful neatly corresponds to the categories of understanding (Verstand ), while the sublime is on a par with the higher faculty of reason (Vernunft. ) Within the Kantian system, the experience of the sublime can therefore be explained as the experience of a reality effectively resisting subsumption within the categories of the understanding.[28] And insofar as this is a reality lying outside the grasp of those categories, the experience of the sublime could, in a no less Kantian manner, be described as the experience of a noumenal reality. Hence, though the sublime surely does test the Kantian system to its limits, the sublime can still be accounted for within Kantian assumptions. Thus, to put it metaphorically, White's obvious fascination with those aspects of the past that
[28] As is explained by White himself. See White, Content , 70.
resist domestication has provoked him to rattle at the doors of Kantian criticism, but has not yet shown him the way out of that well-ordered house of Kantian transcendentalism.[29]
In the remainder of this introduction, I want to pursue the path suggested by White's transition from the kind of views that were presented in Metahistory to those we may find in his Content of the Form. That is to say, following White, I want to explore the possibilities and the nature of a form of historical writing that breaks with the Kantian, Enlightened tradition that always strove for a domestication or appropriation of the past. However, in contrast to White, I shall try to do so by attempting to effectively break the spell of Kantian, transcendentalist patterns of argument. My motivation for looking for an anti- or a-Kantian argumentation will be obvious: in the foregoing, we have seen that the intellectual function of both transcendentalism and of metaphor has always been to effect an appropriation of the relevant parts of reality. Hence, the avoidance of appropriation in our approach to reality can only have a chance of success to
[29] With all the more interest we may turn to another antitropological movement that can be discerned in The Content of the Form . In the last chapter of the collection, White distinguishes between a "linguistic theory of the text" and a "specifically semiological conception of it," and it is clear that his sympathies lie with the latter. A linguistic theory of the text is a theory "that takes specifically lexical and grammatical categories as elements in its analytical model" and can best be associated with the kind of work that has been done by Russell, Wittgenstein, Austin, or Chomsky. Next, the semiological model is described by White as follows:
A semiological perspective, on the other hand, treats the text less as an effect of causes more basic or as a reflection, however refracted, of a structure more fundamental than as a complex mediation between various codes by which reality is to be assigned possible meanings. It seeks, first of all, to identify the hierarchy of codes that is established in the process of the text's elaboration, in which one or more emerge as seemingly self-evident, obvious, natural ways of making sense of the world. (White, Content , 202.)
It will be obvious that this semiological view of the origin of (historical) meaning owes a lot to Barthes's notion of myth as developed in the latter's Mythologies .
We will note, therefore, in White's sympathy for the semiological model another instance of his ambivalent attitude toward the Enlightenment tradition. For, on the one hand, Barthes's notion of myth underlying the passage just quoted is profoundly anti-Kantian. The opacity of language that is presupposed by both White's semiological theory of the text and Barthes's notion of myth would be as alien to Kantianism as the suggestion of the fundamental and unavoidable opacity of the Kantian categories. It is exactly the purpose of so essential a part of the first Critique as the transcendental deduction of the categories to show the transparency of the categories and why the categories of the understanding form a reliable foundation of scientific knowledge and truth. On the other hand, Barthes proposed the notion of myth in order to demonstrate how language can be even more successful in appropriating reality than we are ordinarily aware of—and the same would be true of White's semiological model of the text.
the extent that we know how to resist the temptations of transcendentalism and of metaphor. I want to emphasize, furthermore, that my interest in the development of a non-Kantian theory of history that avoids appropriation is more than an invitation to solve some intriguing intellectual puzzle. One may think here of the following four considerations: First, as is demonstrated by White's own position, because Kantian transcendentalism is so intimately and closely related to the movement toward appropriation, the effort to avoid appropriation will necessarily require, within the Kantian system, some quite radical and dramatic steps. So much is already clear from White's introduction of a category such as the historical sublime and of what he associated with that in the course of his argument.[30] Speaking generally, this is the hidden danger of all Kantianism: it accommodates so much, is so much in tune with all the rationalist tendencies in Western thought, and is so much the epitome of all our cognitive efforts at mastery of the world that it cannot fail to give a tremendous radicalist impetus to each attempt to step outside its sphere of influence. One is reminded here of the image of an extremely heavy mass, such as the sun, and how it may enormously accelerate the speed of objects entering its gravitational field. It seems worthwhile, then, to investigate whether we can attain White's goal with the help of more modest instruments.
This brings me to my second consideration. I shall argue in the last chapters of this collection that several variants within contemporary history of mentalities can be interpreted as implementing the movement against appropriation. Needless to say, these recent forms of historical writing have nothing of the extremism that is (unavoidably) part of White's Kantian conception of the historical sublime. We may expect, next, that developing such a non-Kantian historical theory against appropriation may help us better understand what actually is at stake in the variants of the history of mentalities that I have in mind. I am convinced that we should not credit these new variants with merely having discovered a number of new potential topics of historical investigation, though that is true as well. The history of mentalities—at least some of it—is much more than that. For, it should primarily be seen as a break with most of the historical writing that was produced in the nineteenth and the twentieth centuries insofar as it always aimed at the appropriation of the past, at rendering familiar what was considered unfamiliar and strange in our past and was in this way part of the Enlightened, Kantian enterprise.
A third consideration directly follows from this one. Appropriation was the primary goal of all the cognitive efforts within the several scientific disciplines that have come into being since the Enlightenment. From this perspective the significance of the history of mentalities, as conceived here,
[30] White, Content , 74-75.
must not be restricted to historical writing. We have all become so Kantian that we find it hard, if not impossible, to think of a discipline that does not aim at appropriation. Obviously, if we can discern an example of an actually existing disciplinary practice (hence, no mere theoretical model) that contradicts or does not fit within the all-encompassing Kantian tradition, such an example may function as an entryway into a new intellectual world that is so difficult for us to imagine because of our being conditioned by the Kantian paradigm of knowledge and meaning. It is here that contemporary historical writing—not unlike the historism of the end of the eighteenth century—may give us an inkling of the new intellectual universe that lies ahead of us.
Fourth, and last, the non-Kantian model of a form of historical understanding which does not aim at appropriation is the background of the present collection when taken as a whole. Whereas the first four chapters still operate on the basis of Kantian assumptions, the last three chapters—though each in a different way—explore the possibilities of such a non-Kantian, nonmetaphorical form of historical writing and of historical consciousness. This is why this collection could be said to give an exposition of both the rise and the fall of metaphor in historical writing, as is implied in its title.
If, then, we consider this project of developing the rough outlines of an alternative to a Kantian theory of history, we should begin with the recognition that Kantian transcendentalism is primarily a theory of experience and of how experience is transformed into knowledge. So historical experience, the experience of the past, will be our natural point of departure. At first sight, the appropriateness of this aperture seems to facilitate our enterprise considerably. For, a good deal of existing theory of history possesses the characteristics of a theory of historical experience. Hence, we may expect that existing historical theory will offer us some useful clues. Yet, on closer inspection, it quickly becomes clear that we have been too optimistic in hoping that existing historical theory might function as a useful guide. The first problem we encounter is that most of contemporary historical theory is based on the assumption that the past can never be an object of experience, for the simple but decisive reason that experience always takes place in the present and that an experience of the past is therefore ruled out almost by definition. Admittedly, the historian may base his knowledge of the past on an experience of what the past has left us—such as documents, archaeological findings, works of art, and so on—but these are the sources of historical knowledge of the past and not the past itself. Naturally, the constructivist thesis (defended, for example, by L. J. Goldstein) owes its plausibility to this fact about the origin and nature of historical knowledge.
However, we need not be completely discomfited by the encounter of
this first obstacle. For, we may remember now that both German and Anglo-Saxon hermeneutics (one can think of the names of Droysen, Dilthey, Collingwood, or Dray) often have the pretension of offering us a theoretical account of the historian's experience of the past itself and not merely of material from the past that has survived into the present. As is well known, it was Dilthey's philosophical program to develop a quasi-Kantian theory of historical knowledge and to construct a theory of historical experience of physical reality. But once again, a closer inspection will undeceive us. For, if I am permitted a ruthless simplification, what we will discover in German and Anglo-Saxon hermeneutics is the following: Their argument ordinarily unfolds in two steps. First, an account is given of how the historical agent experienced the historical Umwelt in which he lived. The second step is a philosophical analysis of how the historian may actually copy the historical agent's experience of his past world. Though hermeneuticists often differed with regard to the extent that a reconstruction of the agent's experience is actually possible, the truth is that this copying of experience rather than experience (of the past) itself was what hermeneuticists were mainly interested in. And the conclusion seems to follow that hermeneutics aimed for the elimination of historical experience rather than for recognizing its significance and accounting for it.
One may object at this stage that this is an unfair appraisal of hermeneutics since copying historical experience is a reenactment of historical experience and therefore no different from actual experience itself. But this objection ignores the hidden agenda of hermeneutic theory. We can get a glimpse of this hidden agenda by recalling Ranke's well-known dictum that the historian "should wipe himself out" when representing the past—a dictum that was echoed in Fustel de Coulanges's equally well-known exclamation: "Gentlemen, it is not I, but history that is speaking to you!" and in Michelet's saying that the historical text (at least his text) is a "resurrection" of the past. Of course, I am not implying that Ranke, Fustel de Coulanges, and Michelet should be seen as precursors of hermeneutics (though Ranke certainly wrote and thought in a kindred intellectual atmosphere). But what their words eloquently express is telling and may give us the proper hint for how to interpret hermeneutic theory as just described. For, what is present in all these views and what permeates so much of historical theory, hermeneutic or not, that has been produced since the beginning of the previous century, is a deep and ineradicable distrust of historical experience. The omnipresent, tacit assumption always is that the historian's own experience of the past will unavoidably lead to subjectivity, to a distortion of the past and to the illegitimate interposition of the historian himself between the past and the reader of his text. The historian must completely disappear from the text; the text should be a kind of epiphany of the past that has miraculously come about without the inter-
vention of the historians. And this requires, above all, the radical elimination of the dimension of the historian's own experience of the past. In fact, this is precisely the main point of the hermeneuticist's hidden agenda. For, when he assigns to the historian the task of copying the historical agent's experience of his historical world, this undoubtedly is the most effective way of getting rid of the historian's own experience of the past and replacing it by a more palatable alternative. Hence, historical theory, as exemplified by hermeneutic theory, presents us with the amazing spectacle of a theory founding a purportedly scientific discipline while denying this science its experiential basis. Many of the strange extremisms that we may find in the history of historical theory (no theory of an academic discipline moves so easily from an extreme of realism to an extreme of idealism) originate in this denial of the dimension of historical experience.
However, the resources of hermeneutic theory may not yet be exhausted by the foregoing. For the following three reasons, it is especially the name of Gadamer that may now come to mind. First, Gadamer is quite explicit about the paramount importance of a suitable and well-informed theory of historical experience and of the urgency for a satisfactory philosophy of historical interpretation to develop such a theory. Second, Gadamer's own theory of historical experience is self-avowedly anti-Kantian. For, it is his most original and most serious objection—to both the historists and to hermeneuticists like Dilthey—that for all their rhetoric about the autonomy of the historical understanding within the whole of the sciences, they nevertheless unwittingly accepted Enlightenment's notions about the nature of knowledge and the scientistic ideals of the Enlightenment that went with these notions. Moreover, according to Gadamer, the historists and the hermeneuticists carried the Enlightenment project even further than the philosophers of the Enlightenment themselves had ever dared and hoped to do: if eighteenth-century epistemologists had been notoriously indifferent to the effort to conquer the historical world,[31] historists and hermeneuticists devised highly effective means for improving upon their Enlightenment predecessors. In short, historism is not an attack on the Enlightenment (as the historists themselves believed) but, in fact, the continuation of the Enlightenment, with far more effective historist methods.[32] By exchanging the epistemological effort, that had been shared by the Enlightenment and the historists, for a Heideggerian ontology of Verstehen , Gadamer hoped to emancipate the notion of historical ex-
[31] Of course, this is a caricature of the eighteenth-century historical consciousness that had already been exposed as such by Meinecke in his book on the origins of historism.
[32] H. G. Gadamer, Truth and Method , trans. G. Barden and J. Cumming, New York, 1986, 239-240.
perience from its Kantian heritage and to free historical theory of its transcendentalism. That is also why Rorty discovered in Gadamer his most useful ally in his attack on the epistemological tradition in the history of Western thought since Descartes. Third, Gadamer is quite clear about his Aristotelian inspiration. Since the epistemological tradition, just mentioned, was mainly a rejection of Aristotelian conceptions, indeed the most obvious thing to do for a philosopher questioning the epistemological tradition is to start with a reconsideration of Aristotelianism: if we discover that we are on the wrong route, then we had best retrace our route back to the last junction. It certainly is to be regretted that Rorty never felt the urge to do so.
With regard to this third point, Bernstein even goes so far as to say that "Gadamer's own understanding of philosophic hermeneutics can itself be interpreted as a series of footnotes in his decisive intellectual encounter with Aristotle,"[33] and he goes on to explain that Gadamer's rejection of transcendentalism in favor of a "peculiar interlacing of being and knowledge"[34] is profoundly indebted to Aristotle. For it is the Aristotelian concept of phronèsis in terms of which this fusion of being and knowledge is achieved. Phronèsis, the knowledge of how to act and do the ethically right thing, is described by Gadamer as follows: "for moral being, as Aristotle describes it, is clearly not objective knowledge, ie the knower is not standing over against a situation that he merely observes, but he is directly affected by what he sees. It is something that he has to do."[35] Ethical knowledge is not knowledge of an objective reality outside ourselves but can only be operative on the assumption of a fusion of knowledge and the world. Here we observe how (by means of the Aristotelian concept of phronèsis ) all the Kantian demarcations between epistemology and ontology, between knowledge and being, even between the true and the good, as well as between the transcendental self and what is seen from the perspective of the transcendental self, become blurred. The conclusion seems obvious: Gadamer may well be our best guide when we are looking to develop an alternative to a Kantian theory of historical experience.
Needless to say, it would be impossible to do justice in the framework of this introduction to all the subtlety, the richness, and the complexity of Gadamer's ontological hermeneutics, and I shall therefore have to restrict myself to merely one necessarily incomplete comment about why I believe that Gadamer will nevertheless not provide us with the theory of historical experience that we are looking for here. My comment has to do with the use that Gadamer makes of Aristotle. As may already have become clear
[33] R. Bernstein, Beyond Objectivity and Relativism , Oxford, 1983, 146.
[34] Ibid.
[35] Gadamer, Truth , 281.
from the preceding discussion, Gadamer is mainly interested in Aristotle's ethics. Thus, in the chapter of Truth and Method that is explicitly devoted to "the hermeneutic relevance of Aristotle," Gadamer begins with the statement:
Understanding is, then, a particular case of the application of something universal to a particular situation. This makes Aristotelian ethics of special importance for us—we considered it briefly in the introductory remarks on the theory of the human sciences.[36]
What interests Gadamer in Aristotle's ethics is what Aristotle says about the application of ethical rules to the context within which we have to act. Gadamer thus emphasizes, as does Aristotle, that ethical action always involves epieikeia, that is to say, a completion or perfection of the—in themselves, incomplete—ethical rules ("correction of the law")[37] by applying them to a given context. And the same emphasis on application can be found in Gadamer's account of how we experience the past when we are interpreting a text that is handed down to us by the past. Here Gadamer wishes to draw our attention to the fact that how we experience the text and its meaning cannot be dissociated from the question of what the text means to us in our present situation, that is, how the text applies to us and to our own world. And if this is true for us, this is no less so for previous and future interpreters of the text. Our interpretation of the text is, therefore, part of what Gadamer refers to as a Wirkungsgeschichte, that is, part of a history of the interpretation of a text, and that history has its coherence in how the text was applied to different historical situations in the course of time. I have no argument with these views of Gadamer and am convinced that the notion of Wirkungsgeschichte is decisive in undermining the pretensions of the Kantian, epistemological hermeneutics that preceded Gadamer.
Nevertheless, Gadamer's emphasis on Aristotelian ethics, on application, and on Wirkungsgeschichte unfortunately disqualifies his hermeneutics as a guide for the kind of theory on historical experience that we are looking for. The main datum here is that Gadamer's hermeneutics present historical experience—and that is, for Gadamer, the way in which we experience, read, and interpret a text—primarily as a phase in an interpretation history, in a Wirkungsgeschichte, and precisely because of this, it cannot count as a historical experience, as an experience of the past. To put it succinctly, Gadamer is interested in the historicity of experience (die Geschichtlichkeit des Verstehens ) and not in the experience of historicity (die Erfahrung der Geschichtlichkeit ).[38] The past gives way here to the interpretative texts
[36] Gadamer, Truth , 278.
[37] Gadamer, Truth , 284.
[38] Gadamer, Truth , 235.
that have been written about it, and the locus of historical experience is transposed from the text itself to its interpretation. Much as this may present us with a plausible description of what actually happens in intellectual history or in the history of philosophy, the authenticity that we naturally associate with the notion of historical experience will be lost and unaccounted for if we follow Gadamer.
Two lessons can be derived from our discussion of Gadamer. First, the foregoing will have made sufficiently clear that Aristotle is not only an obvious but also a useful guide in our search for an anti-Kantian theory of historical experience. Second, when we are looking to Aristotle for help, we should not concentrate on his ethics, since ethics is by its nature present-centered and therefore an unsuitable point of departure for our enterprise; because of the role application plays in both Verstehen and in ethical practice, Gadamer was tempted to model the former after the latter and thus lost his grip on historical experience. I therefore propose to consider what Aristotle said in De Anima about sensation and knowledge: application has no role to play there. Yet, perhaps the most appropriate thing to say about De Anima is that it does not develop a theory of knowledge and experience in the proper sense of the word. Indeed, the kind of concepts epistemologists currently use, like truth, knowledge or justified, true belief, reference or meaning, are largely absent from Aristotle's writings, and even when he does use them he does so in a way that is quite different from what we are accustomed to in epistemology. Paradoxically, this is a good omen. For, it suggests that in turning to the intellectual world of Aristotle we necessarily pass from one philosophical paradigm to another, and nothing short of a paradigm-change will be required if we want to outline an anti-Kantian theory of historical experience.
Indeed, if we consider Aristotle's view of sensation, we will observe at once that the kind of questions we have learned to ask with regard to contemporary theories of knowledge and experience are irrelevant here. When describing sensation, Aristotle uses the following suggestive metaphor: "we must understand as true generally of every sense that sense is which is receptive of the form of sensible objects without the matter just as the wax receives the impression of the signet ring without the iron or the gold."[39] What becomes clear from this metaphor is that the whole drift of Aristotle's conception of experience and knowledge is diametrically opposed to the modernist one we have been accustomed to since Descartes and Kant. For, the purpose of the metaphor is to suggest a maximum of continuity (both epistemically and ontologically, as we would nowadays say) between the object of perception and sensation or the act of perception;
[39] Aristotle, De Anima II , XII, 424 a 16ff.
the identity qua form of the signet ring and the impression it makes in the wax produces this suggestion of continuity. This suggestion is further reinforced by Aristotle's conception of sensation as a continuous chain of causal processes whose principal links are the perceived object, an intermediate sphere, and the perceiver.
Contrast this to how the Cartesian methodological doubt and Cartesian metaphysics created an almost unbridgeable gap between mind and knowledge on the one hand and the world on the other, or to how epistemology was given the serf-contradictory task of building an epistemological bridge over an essentially ontological cleft between subject and object, between language and the world, or to how sense data are for the epistemologist mere signs for the existence of states of affairs in reality, or to how all this effected a radical separation between the world and the mind, which remains firmly locked up within its forum internum, and we will come to see that no gap could be deeper than the one between the Aristotelian and the Kantian transcendentalist paradigms of experience. It is the essence of Aristotle's argument that we should not conceive of sense-experience as essentially problematic because of what I have just referred to as the continuity between perceived object and perception. And yet, the main consequence of epistemological thought has always been precisely to problematize the certainties that sense-experience seems to offer. One might go even one step further and argue with Lear that according to Aristotle there is not only a continuity but even an identity of perception and the object of perception. I would like to refer here to Lear's comments on Aristotle's account of the perception of sound. In his recent book on Aristotle's theory of knowledge and experience, Lear reminds us that the Greek word for sound (psofos ) can refer both to sound in the world itself, as when, for example, a tree crashes to the ground and to the perceptive activity of the listener when hearing the sound. And, indeed, for Aristotle perception and what is perceived are the same "as we can call the very same activity 'either the builder building' or 'the house being built.'"[40] Since Aristotle believed that the mind worked in much the same way as the sense faculties, a roughly similar story can be told for knowledge. Mind is a faculty that has the potentiality (in the technical sense Aristotle attributes to that word) of taking on the forms of what is known and of what the mind understands: "contemplating consists in the mind actually becoming the object of thought." Lear hastens to add, though, that these objects of thought are the "objects which involve no matter or, literally, things without matter."[41] Hence, forms, experience, and knowledge
[40] J. Lear, Aristotle: The Desire to Understand , Cambridge, 1988, 106-107.
[41] Lear, Aristotle , 124.
do not separate us from the world—in the sense of giving rise to the epistemological question as to how the two are related—but unite us with the world.
Aristotle's relevant views can be summed up as follows: The dissociation of subject and object that is so characteristic of all transcendentalist and epistemological thought since Descartes is absent in Aristotle. Whereas the transcendentalist subject familiarizes the world outside us, in the sense that it transforms the world after its own image, perception as conceived by Aristotle shows the opposite tendency since the mind, in this case, takes on the form of the objects of the outside world. But this is not a matter of the mind being overwhelmed by the (forms of the) objects in reality; the mind has to activate its potentiality in order to take on these forms effectively if these objects are to be perceived. The image comes to mind of a string that has the potentiality to actually resonate with the sound of a certain pitch. And to pursue this metaphor: the mind could be compared to a string that can change its length in order to ensure resonance. One might now object that Aristotle had only sensory perception in mind, and, furthermore, that the transition from sensory perception to the experience of the past that is at stake in the present discussion is far from obvious. It is therefore a most happy coincidence that Freud, in one of his writings, worked with much the same metaphor as Aristotle; combining the use Aristotle and Freud made of the metaphor in question will enable us to continue Aristotle's argument in the desired direction.
I am thinking here of Freud's note in which he compares the perceptual apparatus of our mind ("unser seelisches Wahrnehmungsapparat")[42] to a mystic writing pad. The issue here is how our life experiences are taken up in our psyche and thus form our psychological history. Like the wax tablet of Aristotle's metaphor that receives the form of the signet ring, the wax of the mystic writing pad receives the imprint of the stylus that has been used for writing on it. And as with Aristotle's theory of experience, we can observe in the case of Freud's writing pad an identity of form between the movement that was made with the stylus and its impression on the wax of the block. And this identity of form is certainly not accidental to Freud's metaphor. For, as Freudian psychoanalysis amply demonstrates, there ordinarily is a striking similarity between the behavior of the neurotic, insofar as this behavior is expressive of the remembrance of a certain traumatic experience in the neurotic's past, and that traumatic experience itself. The experience itself and the way this experience imprints itself in the psyche of the neurotic are structurally alike, or, at least, closely related; the analogy with Aristotle's theory of sensation is obvious.
[42] S. Freud, "The Mystic Writing Pad," in Freud, The Standard Edition of the Complete Psychological Works of Sigmund Freud , vol. 19, London, 1961, 232.
No less interesting is another parallel between what one might call the psychoanalytical experience of the past and the writing pad. The writing pad consists of three layers. On the bottom we have the wax itself. The wax is covered, first, by a thin plastic sheet that makes contact with the wax at those places where one has been writing and it is in this way that the message that is written on the pad articulates itself. On top of this sheet there is another, thicker plastic sheet, which protects the one underneath and on which one actually writes. What struck Freud is that when the two plastic sheets are removed, the message that had been inscribed on the pad will become invisible. Yet, Freud adds, "But it is easy to discover that the permanent trace of what was written [die Dauerspure des Geschriebenen ] is retained upon the wax slab itself and is legible in suitable lights."[43] This fact about the writing pad invites Freud to see the wax as analogous to the unconscious and the two plastic sheets as analogous to what he referred to as conscious systems. What becomes clear from this is that when we look at the writing pad, under certain circumstances, with the appropriate lighting, we see the message that has been inscribed on it, which later became invisible and which we may therefore have forgotten about. Similarly, there must be an experience of our psychological past that has the characteristics of a discovery, without our being aware of it, of what has always been a true part of our psychological constitution. And of such a discovery we may say that it is not a matter of an appropriation or a domestication of an intrinsically alien reality. For, the movement is quite the reverse: we are confronted here with a part of ourselves that appears to have acquired an uncanny independence from us in the course of time. This is not a familiarization of the unfamiliar, but a defamiliarization of the familiar: at our very heart, we have become strangers to ourselves. Here, then, we may discover the formal difference between the experience of the past along Aristotelian-Freudian lines and the transcendentalist, metaphorical conception of experience.
Few philosophers nowadays will be prepared to defend the Aristotelian-Freudian theory of experience and knowledge. Yet I am convinced that it is only a theory such as this that will enable us to make sense of how we experience our personal and cultural past; the kind of theories of experience that are en vogue since the Enlightenment are, in my view, utterly incapable of accounting for how the historical past is experienced both by individuals or individual historians and by the relevant intellectual or cultural communities. What is given us in and by historical experience is not in need of decoding, but should be understood as having formed us in the Aristotelian sense of the word. Perhaps the most convincing and most decisive proof of this incapacity is that philosophers of history have not even
[43] Freud, "Mystic Writing Pad," 230.
thought it necessary to develop a theory of historical experience and—as we have seen—have always been strangely acquiescent in (tacitly) conceiving of history as a discipline in which the experience of the objects studied in that discipline counts for nothing.
I must concede, however, that a similar reproach could be addressed to some of the essays that are included in this collection. In fact, it is here that one might situate the plot that underlies this collection as a whole. If the collection could be said to tell a story, it would be the story of how to move from a metaphorical, transcendentalist conception of history to the Aristotelian-Freudian conception of historical writing. The first chapter gives a short enumeration of all the doctrines that I hold to be of central importance to traditional, Kantian narrativism and to the transcendentalist role played in it by metaphor and tropology in general. Chapters 2 and
3 add some further details to the picture and suggest how a narrativist philosophy of history can emphasize its independence from scientistic patterns of argument while, at the same time, remaining safely within the circumspections of "Kantian" argument. From chapter 4 the argument tends to move outside the Kantian framework. Thus in chapters 4 and 5 an aestheticist view of historical writing is set in opposition to the transcendentalist approach. That aestheticism is a challenge to transcendentalism is not difficult to explain. If we compare reality itself to its aesthetic representation by the painter—and obviously this is the point of departure of all aesthetics—exactly the same epistemological problems we encounter with regard to reality itself will repeat themselves with regard to its aesthetic representation. Epistemology is indifferent to the question of whether we are dealing with reality or with its aesthetic representation: both belong unproblematically to the inventory of phenomenal reality. Hence, if we ask for the relation of reality and its pictorial representation, transcendentalism is sui generis incapable of yielding helpful insights.
But if the aestheticist approach to historical writing merely suggests the irrelevance of transcendentalism rather than an outright rejection of it, in the last two chapters transcendentalism is attacked on its own ground, that is to say, insofar as transcendentalism offers no account of (historical) experience. This attack is not inspired by theoretical considerations but by what is suggested by the actual practice of contemporary history of mentalities. (It is, incidentally, part of the argument in chapter 5 that historical theory should not look for the epistemological foundations of historical writing but reflect on the unexpected and sometimes fascinating philosophical problems that are occasioned by the results of contemporary historical writing.) In chapters 4 through 7 I try to demonstrate that the contemporary practice of the history of mentalities—or at least some of its variants—actually amounts to an overthrow of what previous forms of historical writing, and their transcendentalist legitimation by historist or posi-
tivist historical theory, accepted as the serf-evident hierarchy of the important and the unimportant in sociohistorical reality.[44] Metaphor has always required the historian and his audience to attribute importance to what easily lends itself to metaphorical organization (see above) from the point of view suggested by a specific metaphor—whether embodied in the idea of the nation state, of an intellectual movement, of a social class, et cetera. And the knowledge gained by historical research that resisted this organization by the metaphorical point of view had to be regarded as irrelevant and unimportant. In this way the metaphorical dimension that has always been present in transcendentalist historical theory presented the historian with fairly reliable criteria of the important and the unimportant. Since these criteria could be applied in a more or less general way and were often used by historians to that effect, we may find here the origin of historical writing as an academic discipline uniting many historians in a common effort to describe and explain the past.
But one of the most striking features of the history of mentalities is that it is largely indifferent to this hierarchy of the important and the unimportant; it no longer has the pretension of presenting us with those elements or aspects of the past in terms of which an entire part of the past ought to be understood (as invariably is the message of all historical writing informed by any of the tropes). Hence, what the history of mentalities achieved was not merely the exchange of one set of important historical topics for another (this has, for example, been the accomplishment of Marxist historical writing) but the disruption of the very idea of such a hierarchy of the important and the unimportant. Whether we think of the microstorie, of Alltagsgeschichte, or of deconstructivist intellectual and cultural history, in all cases we witness a revolt (the political connotations of this word are quite appropriate with regard to the victory of the "democratic" detail over "aristocratic" essence that is at stake here) of the marginal against the important, without, however, the marginal ever aspiring to take the place of the historist's or the positivist's categories of the important. The development of historist or positivist historical writing into contemporary history of mentalities may thus show us how the rise and fall of metaphor were realized in actual historical practice.
In the last two chapters several of the threads spun in the previous ones are brought together. For several reasons my argument in these last chapters can best be seen from the point of view of hermeneutic theory as previously discussed. First, my argument shares with hermeneutic theory an interest in the category of historical experience. Second, the postmod-
[44] For an extended analysis of this disruption of the hierarchy of the important and the unimportant, see my "Twee vormen van narrativisme," in F. R. Ankersmit, De navel van de geschiedenis: Over interpretatie, representatie en historische realiteit , Groningen, 1990, 44-78.
ernist theory of history expounded in the last chapter is a self-professed radicalization of classical historism and is developed by carrying Gadamer's hermeneutics a step further. But more important is a third consideration. One might say that hermeneutics has always been an uncomfortable compromise between transcendentalism and Aristotelianism. Hermeneutics is strongly reminiscent of Aristotelianism in that all its many (German and Anglo-Saxon) variants have always strived for a certain identification of the historian with his object (Collingwood's reenactment theory is, of course, exemplary); this reminds one of Aristotle's argument that contemplation should aim for the mind's actually "becoming" the object that is contemplated. But this aim would necessarily remain an unattainable goal since, in accordance with transcendentalism, the object was, in the end, always conceived of as an intellectual construction by the historian's mind. I am not imputing idealism to transcendentalism here—though the affinity between the two will need no clarification, either by way of argument or history—but have in mind transcendentalism's tendency to erect an epistemological and ontological barrier between language or knowledge and the world and transcendentalism's permanent effort to firmly encapsulate the mind in the former. Because of this ultimate unattainability of the historical object, the relationship between subject and object in hermeneutic theory necessarily becomes a matter of copying (see above) rather than of identification. The persistence of the transcendentalist temptation is demonstrated by the fact that even such a staunch antitranscendentalist as Gadamer at times yielded to it. As becomes clear from his emphasis on application and Wirkungsgeschichte and from his preference of Aristotle's ethics to the theory of sensory perception expounded in De Anima, even Gadamer has a tendency to favor the subject over the object and its historical context, and, because of that, his hermeneutics still bears the unmistakable traces of transcendentalism and of how the transcendental self organizes the world after its own image.
In order to circumvent these pitfalls of hermeneutic theory, a new theory of historical experience is developed in the last chapter. This theory of historical experience is required, first of all, to fill in the ditches that transcendentalism dug in Aristotle's notion of experience and knowledge. This theory of historical experience must, furthermore, recognize the authenticity of historical experience as a token of its willingness to abandon the pretensions of the transcendental self to familiarize the (historical) world in the manner that has always been peculiar to the transcendental self. The notion of historical sensation as described by Goethe, Meinecke, and Huizinga enables us to get a clearer view of what is involved in historical experience as described here. Next, the nostalgic experience of the past is proposed as the matrix for a satisfactory analysis of historical expe-
rience. Nostalgia can serve as such a matrix because it is not the experience of a reified, objective reality out there, but of a difference (between the present and the past): since difference demands the presence of both present and past, it allows for this flowing together of subject and object that is so essential in Aristotle's theory of perception. A curious corollary is that the kind of historical consciousness that is exemplified by this form of historical experience is strongly suggestive of a movement of withdrawal: what is experienced historically is a former part of ourselves that in the course of time has acquired a certain independence with regard to ourselves. Part of ourselves was permitted to develop an autonomous existence and, apparently, we have withdrawn from it at some stage. So, once again, we are surrounded by nature but, this time, by a nature that was once part of ourselves and from which we have now become estranged. This, then, is what we see when we look in the mirror of the past: we look at ourselves and see a stranger. Contrary to Vico's intuitions, we may therefore write: "Verum et factum non convertuntur": the historical world is an Other precisely because it is a human artifact.
As will be clear from the foregoing, the greatest debt I owe is to Hayden White. The essays in this collection are mainly ramblings through the intellectual fields surrounding the route he had already mapped out in his writings of the last two decades. His capacity for identifying what really demands our attention at each phase of the intellectual debate on historical writing is, in my opinion, the most formidable asset in the possession of contemporary philosophy of history. There are other American colleagues of mine—especially Hans Kellner and Allan Megill—whose ideas and suggestions are also present throughout these essays. They have functioned for me as a kind of reality principle: the conversations and the correspondence that I had with them taught me to distinguish between mere theoretical fantasy and what I might say with at least a semblance of plausibility. Returning to this side of the Atlantic, I am no less grateful for the advice that I received from my closest colleagues here in Groningen: Josine Blok, Jaap den Hollander, Ernst Kossmann, and Wessel Krul, and from Ann Rigney in Utrecht. For the enrichment of my intellectual background, I owe my gratitude to the members of the Amsterdam group, led by Theo de Boer, for the Foundations of the Humanities.
Anthony Runia has protected the text from the dangers occasioned by my uncertain and limited grasp of the English language. His accuracy, his sensitivity for semantic nuance and, above all, his readiness to go beyond the mere words and to penetrate into and think over the argument itself, are a more solid guarantee for a correct and comprehensible text than I had ever dared to hope for. A very special debt I owe to Machteld
Strabbing-Brinkman. Her tempo, her passion for a perfectly produced text, her knowledge of the most recent programs for word processing, and her indefatigable good humor are beyond all praise. However antifoundationalist we may become in theory, we know that in practice reliable foundations are indispensable. In the more than twelve years that we now have been working together, Machteld Strabbing has been such a reliable foundation for me.