" Encompassing the future "

“Encompassing the future” Jonathan Gorman Oulu, 6 October 2017 1 The heyday: Hempel versus Dray Jouni-Matti Kuukkanen’s Call for Papers is a valuable contribution to the analysis of an academic state of affairs and its possible advance. With regard to our conference subject, The Role of Philosophy of History, it suggests among other things “that the philosophy of history does not have much of a role in current academia having lost it since the heyday of the analytic philosophy of history”. “Perhaps now”, it continues, “could be the time to regain that position”. Ah, those were the days. Some comments are in order about when exactly the heyday of the analytical philosophy of history was, who was experiencing it and what made the period a “heyday”, if it was. We all know that the period in question was not the heyday of so-called speculative philosophy of history, for which we have to return even longer ago to Kant, Herder, Hegel and Marx, among others. Later, such writers as Toynbee and Spengler were already out of their proper time since the whole speculative approach had been successfully squashed by Karl Popper and Isaiah Berlin, so far as analytical philosophers were concerned; Patrick Gardiner had been able to write an introductory book in the area without referring to the speculative philosophy of history at all,1 although W.H. Walsh, who had 1 Patrick Gardiner, The Nature of Historical Explanation, London: Oxford University Press, 1952. 2 published his Introduction one year earlier than Gardiner, had felt it necessary to apologise in case it was thought to be speculative.2 However, non-“analytical” philosophers of history and those reflecting on historical method had had a bit of a heyday too, without being “speculative”. Thus Ranke – who, intriguingly from my point of view, married an Irish woman from Dublin – had made claims about history telling how it “actually was”, claims that exercised and encouraged in particular American historians, encouraged so much that they made him the first honorary member of the American Historical Association. Yet despite Ranke’s influence, American historians worried about the historical relativism described by Carl Becker in 19103 and its implications for the nature of historical facts, an issue that E.H. Carr picked up in 1961, contrasting the Lord Acton of 1896 with the Sir George Clark of 1957.4 Every intending historian in the UK used Carr as their source of historical theorising for decades, with many still doing so. Historians if not philosophers may recall the Bury-Trevelyan debate, Bury asserting that history was a science in 1902 and Trevelyan denying it in 1903.5 A number of non-analytical philosophers writing in English took their cue from Hegel, although not from his speculative philosophy of history but – like Ranke’s use of the zeitgeist – rather from his idealism. Thus F.H. Bradley had written The Presuppositions of Critical History in 1874,6 and Michael Oakeshott had written on historical causation in 1933. 2 W.H. Walsh, Introduction to Philosophy of History, London: Hutchinson, 1951, 11ff. E.H. Carr, What is History?, London: Macmillan, 1961, 21, n. 2. 4 E.H. Carr, What is History?, London: Macmillan, 1961, 7ff. 5 See “The Science of History” and “Clio, A Muse”, in Fritz Stern (ed.), The Varieties of History, Cleveland: Meridian, 1956, 209 and 227 respectively. See also Richard J. Evans, In Defence of History, London: Granta, 1997, chap. 1. 6 F.H. Bradley, The Presuppositions of Critical History, Oxford: James Parker, 1874. 3 3 R.G. Collingwood’s 1936 lectures on the philosophy of history became particularly influential among historians through his posthumously published The Idea of History, in 1946. Yet, however active and indeed influential these many discussions were, and for however long they continued, none of them did much to influence analytical philosophy of history. William Dray’s authoritative collection of 1966, Philosophical Analysis and History, did not include anything by Collingwood, despite the obvious debt to him in his 1957 work Laws and Explanation in History, which itself explicitly owed so much to his teacher Walsh,7 who himself explicitly owed so much to Collingwood.8 “Analytic” philosophy of history, obviously enough, was part of analytical philosophy. That philosophical approach continues, but the Call for Papers plausibly sees the “heyday” of analytical philosophy of history as ending when our interests transitioned towards “studying the narrative aspects of history writing in the 1970s and 1980s”. That is a fair assessment, given that many of us will think that the proper study of the “narrative aspects of history” began with Hayden White’s 1973 Metahistory, which is plausibly not a contribution to the analytical philosophy of history. There was a bit of a spat about who did it first, but, if it was not David Hume himself, then either Carl Hempel, or Karl Popper, or Ernest Nagel introduced into more general analytical philosophical discussion, from the empiricist tradition in the philosophy of science, a theory of causal explanation. But the “heyday” of analytical philosophy of history began with the stimulation provided, not by Hume, Popper or Nagel, but by Hempel’s 1942 7 8 W.H. Dray, Laws and Explanation in History, Oxford: Clarendon Press, 1957, Preface. W.H. Walsh, Introduction to Philosophy of History, London: Hutchinson, 1951, Preface to First Edition, 10. 4 “The function of general laws in history”,9 which by 1959 had “attained the status of a kind of classic in the field”, as Patrick Gardiner observed.10 Hempel applied that theory of causation explicitly to historical explanation. In itself, Hempel’s 1942 article was (to adapt Imre Lakatos’s terminology) just another confident application of the very fruitful, successful and influential analytical philosophical research programme of mainstream pre-Kuhn Humebased empiricist philosophy of science. As such Hempel’s might have remained a comparatively unimportant publication, being perhaps no more than a minor extension of existing successful philosophy to (for analytical philosophers) the extra-mural discipline of history, in which few analytical philosophers qua philosophers were interested. That lack of interest has lasted for most of my lifetime; indeed, even the history of analytical philosophy itself, with respect to which what I am now saying is a tiny contribution, has only in the last five years warranted its own academic journal. However, a heyday for the analytical philosophy of history initially developed with the responses to Hempel from the wider intellectual world of social theorists, for his article provided a focal point for those wondering about the possibility of a genuinely scientific social “science”. That issue, reinvigorating the traditional question of how far history was itself a science, began to attract the attention of a few analytical philosophers. The heyday arguably reached its highest point when William Dray responded to Hempel in 1957 with his book Laws and Explanation in History,11 9 Carl G. Hempel, “The function of general laws in history”, Journal of Philosophy 39, 1942, 35-48. “Recent views concerning historical knowledge and explanation: Introduction”, in Patrick Gardiner (ed.) Theories of History, New York: The Free Press, 1959, 265-274 at 269. 11 W.H. Dray, Laws and Explanation in History, Oxford: Clarendon Press, 1957. 10 5 and the Hempel-Dray philosophical debate stimulated many analytical contributions. With Hempel’s assertions about causal explanation in history gaining this wider traction, analytical reflection about the usual mainstream philosophical issues required that attention be given, not to history itself, but to historians’ modes of thought: to historical method or to how historians behaved or to how they wrote.12 Analytical philosophy of history thus had its heyday in virtue of being valued, not by historians (who took little notice until philosophers started saying things about narrative), but by those analytical philosophers who had not previously taken seriously the explanation of human action by historians. How historians actually explained then took centre stage for those analytical philosophers with an interest in understanding humanity, not least because historical explanations seemed to many to provide some kind of factual test for the universal application of their empiricist philosophical approaches. There was (at least in principle, for those who were interested) a role for historians in these philosophical debates, and a small number of historians contributed; the journal History and Theory was founded in that climate in 1960, three years after Dray’s book. Causes versus reasons in the explanation of human action was an ancient philosophical debate that was advanced by that discussion and continues today, but as the debate progressed it crystallised into what we analytical philosophers now call the philosophy of action,13 and over the years it left behind any new input from 12 Unless the historian were a historian of science like Thomas Kuhn, but no “proper” historian I knew would have accepted him as a member of their discipline. Reflection on Kuhn’s work required philosophy of science, not philosophy of history. 13 See, for example, Donald Davidson, “Actions, reasons and causes” [1963], reprinted in Alan R. White (ed.), The Philosophy of Action, London: Oxford University Press, 1968, 79-94. 6 historians or from historical examples. Once the main points of the Hempel-Dray debate had been absorbed, later analytical philosophy of the causation of human action made no further significant reference to historians’ explanations. As the analytical philosophers came to think, it was not as if historians explained human actions particularly well; although they were using partisan notions of what counted as “explaining human actions particularly well”. When philosophical outsiders were admitted to later discussions of the philosophy of action, they were from psychology, or economic theory, or anthropology, or robotics, but not from history. 2 The heyday ebbing: analysis of narratives However, by the 1960s those few remaining analytical philosophers still interested in historical understanding had also moved on to the analysis of narratives. Analytical philosophers’ attention to narratives was mainly a brief continuation of their interest in causation. Three main works in analytical philosophy of history addressed some of the issues involved in making sense of narratives: Bryce Gallie’s Philosophy and the Historical Understanding of 1964,14 Arthur Danto’s Analytical Philosophy of History of 196515 and Morton White’s Foundations of Historical Knowledge, also in 1965.16 Each of these philosophers made some important contributions to the theory of history: Gallie, for example, described the need for a historical story to be “followable” and introduced the idea of 14 15 16 W.B. Gallie, Philosophy and the Historical Understanding, London: Chatto & Windus, 1964. A.C. Danto, Analytical Philosophy of History, Cambridge: Cambridge University Press, 1965. Morton G. White, Foundations of Historical Knowledge, New York: Harper and Row, 1965. 7 “essentially contested concepts”; Danto introduced the notion of “narrative sentences” and with that made a contribution to making sense of temporal language; White described and analysed historical discourse in terms of fact, law and value and used Hempel and the then current standard philosophy of science more explicitly than the others. Between them they said much more that was of interest to other theorists of history, but, from the point of view of what was of interest to mainstream analytical philosophy, all three were contributing in some small way to the analytical discussion of the nature of causation, a continuing and central analytical issue as the advance of twentieth-century physics forced a reconsideration of some fundamental concepts of science, in particular space, time and causation. Gallie, who had followed closely the debates in the philosophy of science, argued for an interpretation of historical causation in terms of necessary conditions rather than in terms of the then standard Hempelian model accepted by Danto and Morton White. However, neither Danto nor White had been mainstream contributors to the philosophy of science, and rather than producers of the “standard” model they were consumers of it. They took for granted Hempel’s philosophy of causal explanation and tried to develop it as a way of structuring narrative explanation in history. Very few analytical philosophers, centred on the sciences as they mostly were, thought that it was worth pursuing Gallie’s view that causes might be necessary conditions, and they quickly gave up the suggestion as the debates about causation became flooded with more complex arguments relating time, space, logic and probability. If historians actually explained in terms of necessary 8 conditions, that in itself was of very little philosophical interest and merely showed that historians were missing most of the subtleties that physics or even the developing social sciences embraced. Again, very few analytical philosophers, centred on the sciences as they mostly were, thought any continuing philosophical interest arose from the detail of how causal explanation might structure historical understanding presented in narrative form. Causal explanation was best understood through the sciences and the philosophy of science was developing well in regard to that. That the analytical philosophy of causation might be shown to be applicable to historical narrative understanding was to be expected. How exactly it was to be done was a minor interest. As I have said, the philosophical analysis of the explanation of human action was in due course developed in other contexts than history. So the interest in history on the part of analytical philosophers was ebbing. By the late 1960s those few analytical philosophers who had taken a special interest in historical understanding seemed to have drawn all they could from their brief excursion into thinking about historiography. After their works on narrative, Gallie moved on to political science; Danto concentrated on art and aesthetics; Morton White moved on in intellectual history. That, indeed, ended the “heyday” of the analytical philosophy of history. Intriguingly, significant objections to the heyday of philosophy of science itself were already looming after Thomas Kuhn’s work, particularly his The Structure of Scientific Revolutions in 1962.17 Yet discussion of Kuhn’s problematic relations between the analytical philosophy of science and the history of science involved 17 Thomas S. Kuhn, The Structure of Scientific Revolutions, Chicago: The University of Chicago Press, 1962. 9 very little connection with any remaining discussion in the analytical philosophy of history.18 3 Historians’ objections From the point of view of analytical philosophers (who have for one hundred years now taken over the philosophical mainstream in the English-speaking world), there was for many decades and continues to be a small rump of philosophers who have a special interest in, for example, Hegel, Gentile or Collingwood. On the whole analytical philosophers have ignored them,19 much as they regarded the history of their own subject as a different kind of activity from their own subject itself. For their part, historians have mostly ignored philosophers of any stripe. As I have said, they have long made and continue to make their own contributions to theorising about history. “Some historians,” says Richard J. Evans, thinking particularly of Geoffrey Elton [Return to Essentials.., 1991], “have even disputed the right of non-historians to write about the nature of historical knowledge at all”.20 Those few historians who gave Hempel any attention mostly thought that analytical philosophy was not to be taken seriously as any kind of player in mainstream historical debates, and that Bill Dray, who had begun his academic life studying history, had as a philosopher in later life done little more than refine Collingwood’s position on empathetic understanding, a position most historians 18 See issue 1 of History and Theory, 1960. While there was some discussion of history and philosophy of science here, the arguments in history and philosophy of science did not much connect with the arguments then used within the analytical philosophy of history. 19 As contributors to a failed programme in traditional idealism. 20 Richard J. Evans, In Defence of History, London: Granta, 1997, 11. Evans characterises in this book a wide range of attitudes of historians to the theory of history, particularly postmodernism. See also his “Afterword” in the 2000 edition. As a philosophy Ph.D. student trying to make sense of history, I found Elton very kind; but he certainly thought I needed to do history in order to understand it philosophically. 10 were happy to live with if they thought about it at all. Unlike Dray they took little interest in Hempel’s theory of causation. But when Morton White and Arthur Danto extended what Dray had called “covering law theory” as a way of analysing historical narratives, for a short period around those works’ publication in 1965 some historians found the approach less easy to ignore. White’s and Danto’s science-sourced causal structures were in general not accepted by historians or by non-analytical theorists of history, and historians’ responses were in the main loud and, to begin with, simple. J.H. Hexter,21 probably the loudest and most effective of historian objectors in the hearing of analytical philosophers, spoke for many historians in characterising the Hempelian solution to the nature of historical understanding as one which mistakenly assimilated history to science. Hexter saw the analytical philosophers’ model as being imposed on history, so that Hempel, White and Danto should be seen as making a prescriptive power-grab where the foundations of knowledge were concerned. The philosophers imposing the model, White and Danto particularly, had unpleasant traits of character: reviewing Morton White’s Foundations of Historical Knowledge, Hexter ascribed to White “an intellectual imperialism generated by the sin of intellectual pride”.22 White, he said, was “peremptory, preemptive, and prescriptive about ‘meaning’, ‘knowing’, ‘understanding’, ‘explanation’, and ‘truth’”.23 This objection looked for a moment like simple name-calling, and philosophers had long taught each other not to engage in 21 J. H. Hexter reviewed White’s Foundations of Historical Knowledge in “The one that got away”, New York Review of Books, 9 February 1967, 24. See also A.C. Danto, letter, New York Review of Books, 18 May 1967, 41. 22 J.H. Hexter, “The one that got away”, The New York Review of Books, 9 February 1967, 24-28 at 28. 23 J.H. Hexter, reply to letter (28-29) from Morton White responding to Hexter’s review, The New York Review of Books, 23 March 1967, 29-31 at 31. 11 arguments ad hominem. However, what I have quoted is only half of Hexter’s point, which was in its way more philosophical. In effect he attacked the status of Hempel-based so-called “models” of historical explanation by posing a philosophical dilemma: these models from analytical philosophy were either prescriptive or descriptive.24 If prescriptive, then they were indeed a power-grab by a disciplinary rival that failed to take history seriously as a knowledge producing discipline. Well aware that the fact/value distinction (to which Humean empiricist philosophers such as Hempel in 1942 were committed) applied as much to philosophy itself as to anything else, Hexter knew very well that prescriptive, otherwise normative, claims – all matters of valuation – were not seen as ultimately rationally supportable within that empiricist philosophy and ultimately depended on taste and temper. So such empiricist philosophers of science had no business setting standards for what ought to be done in history and were contradicting their own position in attempting it. On the other hand, Hexter’s argument continued, if an empiricist model of explanation such as Hempel’s was descriptive of history rather than prescriptive for history, it was flatly false. Historians knew better than philosophers did what was going on in history. History was a major epistemic discipline that involved successful methodological practices in which they were experts. It was clear to all that analytical philosophers’ reading of historical works was at best cursory and their experience of historical practice largely non-existent. When it came to describing what historians did, historians used evidence grounded in their experience, while the philosophers were offering abstract intellectual speculative 24 I will not address the question how far covering law theory is a “model”. 12 constructs that failed to fit the facts of the epistemic practices of historians. Arthur Danto complained about this review by Jack Hexter of his and Morton White’s books, describing it as “vagrant and irrelevant”;25 “What Hexter evidently wants…is some sort of exact composite portrait of the working historian”.26 Responding to Hexter in The New York Review of Books, Danto knew he would have historian readers and presumably thought that this supposedly authoritative rhetorical riposte would be effective for them. It wasn’t. Danto possibly thought that few analytical philosophical readers read The New York Review of Books for, if they read outside philosophy, they surely read science rather than literature, but in any event he didn’t provide any suitable summary argument in objection to the review, merely taking for granted that any analytical philosophers would understand what was so obviously wrong with Hexter’s demand. But whatever was wrong with it, it wasn’t obvious. Analytical philosophers generally were aware of objections like Hexter’s even though they didn’t get the problem from him, for it arose in relation to their work in analytical philosophy of science. Some of those philosophers brazenly accepted one choice posed by the dilemma and thought that, power grab or not, philosophers did indeed know better than historians what counted as knowledge. They imposed on history models derived from science just because science rather than history really was an epistemically licit discipline for reasons that they thought obvious. 25 26 A.C. Danto, letter responding to Hexter’s review, The New York Review of Books, 18 May 1967, 41-42 at 42. Op. cit., p. 41. 13 Others accepted the other horn of Hexter’s dilemma: as late as 1976 Leon J. Goldstein affirmed as an assumption for his own philosophy of history, in Historical Knowing, that “history is an epistemically licit discipline which deserves to be taken seriously on its own terms”,27 which meant that philosophers should not impose on historians alien criteria of knowledge. However, agreeing that history was epistemically licit or being otherwise “history-friendly” left wholly unclear what kind of “description” rather than “prescription” Goldstein and others similarly placed were themselves engaged in. Should analytical philosophers provide an “exact portrait”? Should they be writing the history of historiography? Few thought that, but nevertheless felt that in some way they had to do better, although, whether merely describing or rationally redescribing either historical methods or writings, the philosophical nature of what they were doing was not wholly clear to them. But then, in the 1960s it wasn’t wholly clear in philosophy of science either. What had driven the problem was the nature of philosophy itself and the common characterisation of it as finding fundamental, and so universal or exhaustive, categorisations or truths which – being universal or exhaustive – necessarily applied to philosophy itself. Just as medieval theology had held the immanence/transcendence distinction to be categorically foundational and exhaustive, and just as Descartes had held the mind/body distinction to be categorically foundational and exhaustive, and just as Kant had held the a priori/a posteriori distinction to be categorically foundational and exhaustive, so the Humean empiricism that underlay the so-called standard analytical 27 Leon J. Goldstein, Historical Knowing, Austin and London: University of Texas Press, 1976, xi. 14 philosophy of science expressed by Hempel and used by White and Danto carried on that ancient tradition and involved its own supposedly fundamental categorisations: fact/value; analytic/synthetic; prescriptive/descriptive; for example. Such distinctions were and are problematic. It is easy to invent categorical and exhaustive distinctions for, given elementary laws of logic, one can sort the world in terms of either asserting or denying any arbitrarily selected assertion whatsoever: either that assertion or its denial are bound to be true, which makes the distinction empty in its application. To make such distinctions work one needs to provide separate contentful criteria for each side of the distinction, and then at least semi-metaphysical (rather than merely logical) grounds that support the claims to the absoluteness of the distinction and the exhaustiveness of its application. On the whole such grounds have not been found. Maybe the elementary laws of logic are themselves at risk. Analytical philosophers had long worried about their own methods, and the journal Metaphilosophy was founded by 1970. Categorical distinctions could not be taken for granted at a metatheoretical level, so Hexter’s descriptive/prescriptive dilemma, while a proper one to pose in objection to philosophers like White and Danto committed to that very distinction, was ad hominem in its effect and not in general sound. Analytical philosophy was itself supposed to be in some way “scientific”, but what counted as “scientific” would be the upshot of philosophy of science, and the way to construe it would depend on the development of that subject. Falsification by “facts” was not involved except superficially. 15 By 1962 Kuhn had begun to revolutionise philosophy of science by historicising it, using a broadly pragmatic rather than simple empiricist approach. Hexter’s presupposition, that philosophers of history being “scientific” about history required that their models be falsifiable by the facts of historical practice, involved an understanding of the philosophy of science and its application in metaphilosophy that was already being overturned. After Kuhn’s 1962 publication, and apart from Goldstein, a dilemma like Hexter’s found no traction among analytical philosophers of science and hence no traction among analytical philosophers of history either. 4 Hexter’s empathy with Morton White While there were in the context of analysing narratives significant differences between them – mainly between Danto and Morton White on the one hand and Gallie on the other – all three analytical philosophers of narrative were nevertheless in some central sense, like Thomas Kuhn, philosophical pragmatists. Unlike Kuhn and by 1965 somewhat out of touch with the latest developments in philosophy of science, they thought that probably committed them to the epistemic priority of science. They were well informed in the American tradition of pragmatism, particularly its then outcome in Quine’s work, which dated in relevant aspects from 1951. Quine had argued by then for the continuity between science and philosophy without any principled distinction between those subjects. It was no doubt their pragmatism that led Danto and White to an interest in history anyway; pragmatism’s interest in 16 history had gone back to Peirce, on whom Gallie wrote a book.28 Like Quine, they did not distinguish “analytical” from “pragmatic” philosophical approaches, as some others wished to do, and they saw their pragmatism as being part of the analytical tradition. Those analytical philosophers who did read Hexter’s review of Danto and White would have seen it as intemperate, and I recognise from my own observation of historians what masters of invective they can be; Hexter was expert at that. Analytical philosophers, while familiarly bullying each other in seminars, nevertheless saw and continue to see themselves as operating according to standards of logic rather than rhetoric and as going wherever the arguments led. Hexter’s describing Morton White as assuming “an intellectual imperialism generated by the sin of intellectual pride”29 was a clearly illegitimate comment, from the point of view of analytical philosophy, since it addressed the person and not the issue. Nevertheless, let us take Hexter’s point seriously as a historical assertion about Morton White’s character or thought processes. Is it true that White was a proud intellectual imperialist? It is appropriate to use “empathy” to assess this, being a method widely viewed by historians, not least Hexter himself, as at least one acceptable and characteristic historiographical way of making sense of individual actions. Moreover, it is plausible to believe that the analytical philosophy of history would do well to revisit the analysis of empathy. We can leave aside for a moment whether making sense of White’s thinking by empathising with it can be properly modelled by covering-law theory. We may then attempt 28 29 W.B. Gallie, Peirce and Pragmatism [1952], New York, Dover, 1966. J.H. Hexter, “The one that got away”, The New York Review of Books, 9 February 1967, 24-28 at 28. 17 to empathise, at least in summary, with White’s intellectual attitudes at the time he was writing the book that Hexter reviewed with such venom, Foundations of Historical Knowledge. Empathising with individuals involves ascribing thoughts and attitudes to them, by doing our best to put ourselves in their position and attempting (at least in a metaphorical sense) to see things from their point of view. When we do this we are imagining being in their situation, but we are imagining what we hold to be real and not merely imaginary, although we must allow that it takes evidence to support a claim that our empathising is successful rather than merely the construction of something fictional. In so far as we succeed in empathising with others, we are sharing with those others a world of “thoughts” or “ideas”, but we do not need to share with those others any particular philosophical position about the status of those ideas. Nevertheless so central is the need for empathy in the shared construction of the world we share with others that the centrality of “ideas” may well drive the sense that some idealist philosophy is required. There are various characteristics to be picked out from within their point of view so understood. We are trying to get into their world as they are aware of it, but also into their world as they are unaware of it. Collingwood famously analysed this historiographical method as the “re-enactment of past thought” and it is common to understand him as referring to another individual’s conscious thought. However, as I have argued elsewhere,30 it is plausible to hold that that interpretation would make impossible the recovery of another of Collingwood’s notions, “absolute 30 Jonathan Gorman, Historical Judgement: The Limits of Historiographical Choice, Montréal: McGill-Queen’s University Press, 2008, passim. 18 presuppositions”, just because they are understood as assumptions which are so firmly taken for granted that they are not a part of the past individual’s conscious deliberation. In the history of philosophy we often have to recover unconscious assumptions by the analysis of explicit thoughts, an analysis that might rely on predicate logic, conversational implicatures or other time-extended notions of dialogue or argumentation. It may be that few of us recognise ourselves well, as the Scots poet Robert Burns remarked: “O wad some Pow’r the giftie gie us to see oursels as ithers see us!”.31 Morton White’s conscious thought surely did not include some explicit recognition on his part that his intellectual attitudes were proudly imperialist. But they may possibly have been so anyway, consciously unexamined as we might take them to have been.32 Yet any historian would need some evidence for this, and it is difficult to see what logical analysis of his writings would yield such a result. However, there is some important evidence against this ascription as a particular feature of White’s character, namely that White shared his attitudes with other analytical philosophers. As I remarked earlier, he was himself, like Danto, a consumer of covering-law theory rather than a producer of it, intent on applying this idea in the apparently alien intellectual land of history. That does, indeed, have an imperialist look. However, neither White nor Danto questioned the empiricist foundations of covering-law theory with its commitment to certain positions on knowledge and truth. Pragmatically, these were for 31 Robert Burns, “To a Louse, On seeing One on a Lady’s Bonnet at Church”, 1786. I recall Herbert Butterfield’s observation that there are “things that the men of 1600 shall we say – but the men of 1900 similarly – do not have to explain to one another, and the result is that they do not always get into the historian’s evidence”. Herbert Butterfield, “The discontinuities between the generations in history: their effect on the transmission of political experience”, The Rede Lecture, London: Cambridge University Press, 1971, 6. 32 19 them the best answer. I stress, “for them”, for pragmatism is wary of universalising. Those foundations operated just as Collingwood described in introducing his idea of an “absolute presupposition”: the empiricist assumptions were for White and Danto psychological rather than Cartesian certainties, unexamined presuppositions underlying the beliefs and attitudes involved in their ordinary analytical philosophical ways of life, contingently uncriticised and uncriticisable by them at the time.33 While they probably gave the matter no conscious attention whatever, a Kuhnian revolution in their analytical philosophy was not in their view required by their pragmatism, and as pragmatists they were engaged in what Kuhn would have called “normal” rather than “revolutionary” analytical philosophy. The empiricist foundations of covering-law theory were uncriticised and uncriticisable by White and Danto at the time because they were not entertained as conscious thoughts for examination, they were at no point doubted, and neither White nor Danto perceived any serious alternatives to them with which they might be actively contrasted. Problems not frameable in the empiricist terms familiar to them were not recognised. Hexter complained that it was explicit in White’s approach that he did not allow the elements of historical writing outside the causal skeleton of explanatory and descriptive statements to be expressive of knowledge of the historical past. White responded that he had made plain the literary or evaluative elements in historical writing. But it was not the claim that the literary or evaluative elements were there which offended Hexter, but rather the implication that such elements had 33 See W.H. Walsh, Metaphysics, London: Hutchinson, 1963, 160ff.; R.G. Collingwood, Essay on Metaphysics, Oxford: Clarendon Press, 1940; and R.G. Collingwood, Autobiography, Oxford: Clarendon Press, 1939, chap. 8. 20 nothing to do with knowledge, the implication that any feature of historical writing outside the “scientific” skeleton had no explanatory or truth-telling use. Hexter did, it may seem to us today given our familiarity with Hayden White’s position, offer Morton White and Danto what we might now judge to be, and to have been at the time, a serious alternative to their approach: he thought that certain literary elements, which he called the “rhetoric” of history,34 were themselves an essential way of expressing historical knowledge. The rhetorical side of historical writing, which may be seen as the ordering or “emplotment”35 of sentences in a narrative, has an explanatory and truth-telling use. This is sure to remind us of Hayden White’s position. But Hayden White’s Metahistory in 1973 fared no better than Hexter in 1967. Neither White nor Danto nor analytical philosophers generally recognised any possible alternative here. They ignored later theories of narrative with their frequent use of or references to literary approaches. They found these literary approaches, with their implicit denial of compositionality, unacceptable, holding as these analytical philosophers characteristically did to atomistic, empiricist and Fregean theories of meaning, reference and truth. Only a “fancy view of historical truth,” Morton White claimed in due course without argument,36 would permit such an approach. Yet this was not imperialist stonewalling but simple blindness born of normal philosophy, well described by Kuhn albeit in different contexts. 34 J. H. Hexter, “The rhetoric of history” [1968], in Doing History, London: George Allen & Unwin, 1971, 15-76. This is Hayden White’s word: see Metahistory, Baltimore: Johns Hopkins University Press, passim. 36 M. G. White, letter, New York Review of Books, 23 March 1967, 28, attempting to clarify his position in Foundations of Historical Knowledge, New York: Harper and Row, 1965. 35 21 5 Holistic meaning pragmatically understood I have argued elsewhere that the main characteristic of narratives, which I prefer to call “accounts as-a-whole” because there are various ways of structuring them, is their time-extended nature, both in the time it takes to read, write or understand them and the temporal extension of the period covered and to which they give meaning. Such accounts, I have said, are to be understood as shareable with others in a shared historical consciousness in both reading and writing. To be shared, communicable thought – and therefore language – is essential. Much analysis is still required of meaning, reference and truth at this holistic level, but, as I have suggested, they have features that may well require to be grasped on idealist assumptions. However, as a whole, each account sorts the shared world it describes in an analogous manner to that in which atomistically understood concepts may be, in a neo-Kantian way, thought of as sorting the world in their less, or non-, temporally-extended ways. Pragmatic choices of reference and description are in principle available to us.37 As I have argued elsewhere on pragmatic grounds, it is often the case that modes of reference and description are established and there is no practical room for revising them. Nevertheless there are continuing opportunities for choice in many situations, particularly when it comes to sorting our understanding of the world and making our future. The theoretical principle that “we can choose” how to sort the world needs to be understood as directing theoretical attention to the myriad contingencies of when choices are available, what those choices 37 See Jonathan Gorman, Historical Judgement: The Limits of Historiographical Choice, Montréal: McGill-Queen’s University Press, 2008, passim. 22 are, and who the “we” might be that has a realistic opportunity to choose in such contexts and the power and ability to carry through any decision, in effect, the power and ability to create a future, whether that involves an advance in theorising or a political programme. There is no “essence” of history; there are a number of ways of giving meaning to time-extended experience. The role of “narrative theories” in any meaningful structure is contingent. We do not face a binary choice between Frege and narrative theory. It is appropriate here to present examples of different ways in which the meaning of a time-extended experience can be given by quoting from the classical Latin author Plutarch: “They sent a dispatch scroll to Lysander to recall him. These scrolls are made up in the following way. When the ephors send out a general or admiral, they prepare two cylindrical pieces of wood of exactly the same length and thickness, each corresponding to the other in its dimensions. One of these they keep themselves, the other being given to the departing officer, and these pieces of wood are known as scytalae. Then whenever they want to send some important message secretly, they make a long narrow strip of parchment, like a leather strap, and wind it round the cylinder with the edges touching, so that there is no space between the folds and the entire surface of the scytale is covered. Having done this, they write their message on the parchment in the position in which it was wrapped round the cylinder, and then they unwind the parchment and send it without the cylinder to the commander. When it reaches him he has no means of deciphering it, as the letters have no connexion and 23 appear to be all broken up, and he has to take his own cylinder and [306] wind the strip of parchment round it. The spiral is then arranged in the correct sequence, the letters fall into their proper order, and he can read round the cylinder and understand the message as a continuous whole. The parchment, like the cylinder, is called a scytale, just as the thing that is measured often has the same name as the measure”.38 Notice that it is an accident of language, a mere contingency of symbolic presentation, whether the “letters” that have no connection when not in correct spiral form are rather whole words or even whole sentences. The message has to be understood as a continuous whole regardless. Logically, the “whole” in this case is not a narrative, but probably has the pragmatic status of a command. A second example from Plutarch: the King of Persia39 gave Themistocles “leave to speak with complete frankness about the affairs of Greece. Themistocles replied that human speech may be compared to an embroidered tapestry, which shows its various patterns when it is spread out, but conceals and distorts them when it is rolled up, and for this reason he needed time. The king was pleased with this simile and told him to take as much time as he chose. Themistocles asked for a year, and in that time he mastered the 38 Plutarch, “Life of Lysander”, translated Ian Scott-Kilvert, The Rise and Fall of Athens: Nine Greek Lives by Plutarch, London: Penguin, 1960, 305-6. 39 Plutarch did not clarify which King. 24 Persian tongue sufficiently well to converse with the king without an interpreter”.40 But even a carpet entirely in the Persian tongue still needs to be unrolled, of course. Classically trained as they undoubtedly were, Morton White and Danto were nevertheless not open to conceptions of meaning beyond the atomistic approach of their version of empiricism. That version was an absolute presupposition for them. Absolute presuppositions are recognisable with historical hindsight when it might be found that alternatives had been in principle available, but which were not in fact noticed or practicably noticeable at the time. Only later philosophers – perhaps even White and Danto themselves – might be able to bring these to mind in an examinable manner because, with hindsight, they could perhaps then recognise that there were alternatives to them that marked the limitations of the presuppositions they had once unthinkingly had. More likely, they would have tried to avoid what C.E.M. Joad called the “acute mental discomfort” involved in the required “rearrangement of mental furniture”. “This last”, Joad continued, “is a task from which we increasingly shrink as we grow older and after middle-age are usually unable to perform at all”.41 However, Hexter’s response did not amount to presenting a recognisable alternative. For White and Danto, looking forward along what they thought of as the path of further analytical philosophical development, there was no way forward in the future without the empiricist and Fregean assumptions they were taking 40 Plutarch, “Life of Themistocles”, translated Ian Scott-Kilvert, The Rise and Fall of Athens: Nine Greek Lives by Plutarch, London: Penguin, 1960, 104. 41 C.E.M. Joad, Philosophy, London: The English Universities Press (Teach Yourself Books), 1944, 13. 25 for granted. There was no self-doubt. There was no ground for examining the presuppositions. It was not that White and Danto were imperialistic in their attitudes, but rather they were responsible, middle-of-the-road in their unexamined acceptance of what was for them established theory. Hexter’s denial of their attitude was intellectually irresponsible, indeed, revolutionary from their point of view. Normal philosophy would collapse if Hexter’s unFregean claims about truth, knowledge and explanation were followed through. That Danto and White thought in such terms was understandable in terms of proper historiographical empathising with them. 6 Shared imagining When we empathise with some past individual such as we are imagining doing now, we should recognise that this is in principle no different from empathising with somebody in our present world. Can we “really” get into the mind of another? It is a familiar problem in philosophy, derived from Descartes’ position, whether we can get into the mind of another person and indeed we cannot, according to him, be certain that that other person has a mind at all. It is a philosophical advance on Descartes, in the world of analytical philosophy most explicitly derived from Wittgenstein, to recognise that much if not all of “mind” and the correct application of mental predicates lies in the public realm. We share much with other people and we are in fact able to understand them, often to the point of being able to ascribe a range of beliefs and past experiences to them and also to trust them to act appropriately in a wide range of circumstances. We also know that some people 26 are untrustworthy, or have mental problems we are not familiar with, or experiences we have never had, or speak in languages difficult to interpret, or are secretive and unwilling to share. When I moved from Edinburgh University to Cambridge to work with Bryce Gallie, I was warned by one of my tutors, George Davie (the expert on the Scottish Enlightenment), that, in his experience of Gallie, I would not be able to find out what he thought philosophically by talking to him but would have to read his publications. That turned out to be and still is true, and it makes no difference to my understanding of him that Gallie is now long dead. Facial cues or turns of phrase are indeed no longer available, he is not able to confirm or disconfirm an attempt to grasp his thinking, but since he would not have done that anyway such face-to-face evidence would not have been of much help if he were here now. Such as these are familiar practical hurdles that we all have to jump in understanding other people. We need not import the alleged Cartesian impossibility of entering another’s mind as a ground for making impossible empathising with a past individual, as if it were an additional ground for scepticism with respect to knowledge of the past. There is no difference in principle from empathising with a present individual, merely the absence of some cues we might otherwise have. On the other hand, in such practical terms, once a person is dead we often have evidence about what they thought which was not available to us when they were alive. I do not need, in my use of empathy in what I am saying, to carry some extra burden of proof with regard to understanding others in the past beyond what is required in the 27 present. In summary, if you can empathise with me, then I can empathise with Morton White. Imagining empathising with the more general case, I imagine that I, like them, am trying to give some meaning to the imagined space and time they are in, some meaning to the world that they are conscious of being in, some meaning to the time-extended experience they are living through. “Meaning” here is not some merely metaphorical expression. Rather, that “whole” of spatially and temporally local experience which empathy gives us expresses the world so experienced in a full-blooded way, so that I can draw on a non-Fregean conception of empirical meaning and hold that whole to be what Quine called a “unit of empirical significance”. Helpfully, Quine’s view in “Two dogmas” was that the term and the individual statement are not “units of empirical significance”, and that the “whole of science” is, although we should not accept his view that these are incompatible alternatives. Taking pragmatism seriously – indeed, more seriously than Quine did – I hold that it is a pragmatic or contingent matter what a “unit of empirical significance” can be, depending on the different purposes we may have. It is not just Quine’s “whole of science”, “terms” and “individual sentences” that can be “units of empirical significance”, but, centrally for historical understanding, that whole time extended world with which we empathise, and that whole time extended world expressed in a typical historical account. Nevertheless we need some of the atomistic empiricist elements and pragmatism enables us to keep them. As Frege said “It must be laid down that a letter retains in a given context the meaning 28 once given to it”,42 and it is a characteristic feature of language that it be reliable in shared contexts. This does not, however, require that symbolic units be universally applicable in the sense of having eternal reidentifiable meanings regardless of context. We do not need to think of our temporally extended worlds as quasimathematical and requiring to be sorted in terms of fixed meanings, although there are areas of discourse where that is indeed our aim. In the small areas of life – operating with “moderate-sized specimens of dry goods”, in J.L. Austin’s words,43 it is helpful in all manner of ways to have established meanings. For example, it seems to us that objects persist through time and persistence of word meaning can reflect that. Establishing each word or symbol or other signifier to mean the same is similar to making, in monetary systems, each relevant coin worth the same. Mutual trust and our faculty of memory enable us to hold the meaning of certain words constant in localised situations. But such as these are not the only units of empirical significance. It is also characteristic of a shared world that its temporal extension means that it can be structured in terms of concepts and modes of presentation that can cover change and can themselves change and that includes the stories we write about others or tell ourselves. I imagine I am engaging with a being that is alive, that has memories, beliefs, hopes, expectations, although I accept that their memories may be excellent or poor, their beliefs well-grounded or perhaps self-contradictory or weakly held, their situation hopeless. Others are hopeful, with plans that they propose to carry out and 42 Peter Geach and Max Black, Translations from the Philosophical Writings of Gottlob Frege, Oxford: Basil Blackwell, 1966, 1. 43 J.L. Austin, Sense and Sensibilia, London: Oxford University Press, 8. 29 with sufficient confidence to succeed, seeing themselves as engaged in ongoing actions in states of affairs in which they play a part. The world in which they see themselves as living may be perceived by them as fixed or as changeable. I imagine that they have inherited, learnt or developed a complex range of attitudes. I imagine they are rational much of the time, but I would be very careful what sense to give to that and do not need to commit myself to a particular position on the matter. Any of these characteristics might be consciously or unconsciously held. I imagine that I share a “human nature” with them, but I am very wary of all the philosophical models that have been offered for that. For example, Plato’s Socrates thought, in simple summary, that some people were motivated by reason, other by so-called spirit, others again by desire. This three-fold suggestion has over the millennia morphed into some assumptions widely used today in some contexts of social theorising, that humans are motivated by desires and means-ends beliefs. There are theories which make desire-maximisation essential to understanding human choices. Some have attempted to model such reasons as causes of actions, in ways that might allow covering-law theory to get a grip on understanding human action. Kant gave morality a central place. Dray rightly brought to our attention the calculations under constraints of appropriateness that individuals might make. Yet all these philosophical models seem to me flatly false if they make any claim to being exhaustive as characterisations of that human nature which we empathise with in the general case. Where they are true, they are contingently so, applying to some and not others. 30 Supposedly universal models of human nature such as these miss out on the most obvious feature of the general case of empathy, which is that we are attempting to put ourselves in the position of another and looking all around their world for ourselves, spying out the spatial and social surroundings and looking forwards and backwards in time, as far back and as far forward as we can, reconstructing relevant memories and imagining their future as they imagined it. In so far as such looking requires conceptualisation rather than what might once have been called raw experience, we have to conceptualise that world in which we find ourselves when empathising with another. It is an abstract Descartes-led possibility that our conceptualisation of another’s world may be radically different from theirs, but we should hesitate before thinking that to be a practical problem. It is structurally similar to that problem on which first year philosophy students cut their teeth, whether you “really” see as green what I “really” see as red, but where our use of those words has been such that we always agree on what colour something is. Rather, the evidence by which we are able to empathise with another person, either past or present, drives the conceptualisation to be used in making sense of their world. There may not be any choices to be made here, either for us or for those we empathise with. It is a factual question, determinable by evidence, whether or how far any difference in fundamental modes of conceptualisation exists between us and those we empathise with. It remains the case that they knew things we don’t, and we know things they didn’t. It is contingent what those things are. 31 7 Making the world There is a sense in which we share consciousness with those with whom we successfully empathise just in so far as we share with them the meaning they make of their world. It might be thought that merely sharing language with them would ensure this, but, apart from possible differences in language, there are various ways in which the meaning we make of their worlds will not necessarily be the meaning they make. We have hindsight, and can use what Danto called “narrative sentences” to redescribe their world in ways that they might not have available to them. Even in our present we may be rivals with some people alive now as to how to give meaning to the world in which we live. “Sharing” that world with them will have its contingent limitations. Nevertheless, we may instead agree with them, and may also agree with those already dead as to what meaning is to be made. We may, for example, share a vision that they once articulated and be carrying on with its implementation. We can give meaning to our world by making it through our actions. On the other hand, in the case of individuals sufficiently distant from us in the past or with respect to our heritage we will not have been for them a part of their imagined future, not a part of the world they were or are trying to make. Their ability or inability to grasp and control their future is essential to our understanding of them, the more so the more they differ from us. The world sorted in the holistic temporal way that I have described thus includes (among other things) the past, present and pragmatically foreseeable future in which a person actually lives. Despite Morton White’s confidence, he was not able to make his 32 future academic philosophical world beyond the very immediate. Framing the future of philosophy, in particular the philosophy of history, in terms of the atomistic empiricist assumptions he brought to it was a failure. As a matter of historical fact, the future of philosophy of history in our intellectual inheritance was made by Hayden White, not Morton White. Similarly, it was Kuhn, not Hempel, who framed the then future of philosophy of science; I heard Hempel admit defeat in Cambridge in 1971.44 It is a contingency whether and how far anyone gets to make or control their future, but in general I share the view of Vico that we make the world in which we live. We can see this clearly using one of Danto’s examples of a narrative sentence: the Thirty Years War began in 1618. Nobody in 1618 could have known that the Thirty Years War was beginning, because they would have had to wait thirty years to find out how long it had been. But that is not a point of principle, as if it was impossible to know things thirty years ahead; it depends on how much power someone had in 1618. Imagine some Roman Emperor in charge. With overwhelming power for hundreds of years on the European stage, the imperial machine the Emperor controlled might well have continued to waged a planned war for thirty years and then called it off. But we do not need speculative examples. Historians are very aware of facts such as these. How politicians in power can make the future world is exemplified by historian Maurice Cowling describing high politics in his 1971 44 Hempel acknowledged the merits of Kuhn’s history-based philosophy of science when he gave his paper “Problems in the empiricist construal of theories” [“not to be quoted in print”] at a History and Philosophy of Science Seminar, Cambridge, 4 November 1971. 33 book The Impact of Labour 1920-1924: The Beginning of Modern British Politics:45 “…the first context in which high politics was played was the context in which politicians reacted to one another. The political system consisted of fifty or sixty politicians in conscious tension with one another whose accepted authority constituted political leadership. In this context significance arose from mutual recognition; not from office, but from a distinction between politicians, inside parliament and outside, whose actions were thought reciprocally important and those whose actions were not. It was from these politicians that almost all initiative came. The language they used, the images they formed, the myths they left had a profound effect on the objectives other politicians assumed could be achieved through the political system. …High politics was primarily a matter of rhetoric and manoeuvre. …Political rhetoric was an attempt to provide new landmarks for the electorate. Political manoeuvre was designed to ensure that the right people provided them. …Rhetoric…was an attempt at constructive teaching, an effort to persuade the new electorate to enter the thought-world inhabited by existing politicians. …Party politics issued in … a rhetorical persuasion to adopt the language and expectations of the politician who used it, a succession of affirmations designed to sound resonantly in the ears of whatever audience was being addressed. 45 Maurice Cowling, The Impact of Labour 1920-1924: The Beginning of Modern British Politics, Cambridge: Cambridge University Press, 1971, Introduction, 3-10, passim. I discussed Quine’s pragmatism with Cowling in 1970, wholly ignorant of the forthcoming book. He displayed what I thought at the time to be a surprisingly close interest. I am grateful to Michael Bentley and the Liberty Fund for copies of Cowling’s text of this work. 34 …the central political achievement was not representation but rhetoric…” It was that rhetoric, once successfully taken up and acted upon by the relevant audience adopting the thought world in question, that made the future world. Verum factum, as Vico said. Generalising from this and with an eye on theorising the reality of history, it is not representation but rhetoric which makes our world. This is not the world that science describes, but the actual world in which we live, a world that, in contingent fact, does support the scientific endeavour (at least for a while, with its frequent denigration of so-called “experts”). Philosophically foundational here is empathising with others, with seeing things from their points of view, which is now to be understood in terms of sharing with them a thought-world that is temporally extended. It is contingent what that temporal extension is or where power may lie in controlling it. Making that world meaningful cannot be done with theories of meaning that, like Frege’s and drawing on atomistic empiricism, allow as conceptions of time only the pointpresent or the quasi-mathematically eternal. Similarly, the timeextended advantage, that narratives and similar structures have in giving meaning to our world, only on occasion, in favourable circumstances, permits a reduction to either causal analysis or rational analysis as those were understood in the heyday of analytical philosophy of history. If we are to have a “heyday” for analytical philosophy of history again then we will need to find something in the work of historians, something that analytical philosophers will find new and valuable with respect to their mainstream interests. Depending on the 35 approach used, that may or may not be something that will permit historians themselves to engage with a degree of authority in such theoretical debates. However, such engagement is not necessary. What I have described in this minor contribution to the philosophy of history need not be seen as a philosophy of history at all. It suggests among other things that analytical philosophers attend closely to the notion of temporally extended imagination as being foundational to the construction of that shared reality of which we are conscious. The empathy required for that enables understanding of the world in which we live now. In principle this need not be a philosophy of historiography at all. It is a philosophy of temporal understanding. Nevertheless, historians empathise particularly well with the temporally extended individual and shared structures that I have described, and as a matter of contingent fact analytical philosophers have much to learn from historians in this regard. Analytical philosophers need to recognise and value such historical understanding, and if they do it may well yield another heyday for the analytical philosophy of history. We here can decide now to do this; thereby we will encompass the future of the analytical philosophy of history.