Richard J. Evans

(In Defence of History. Granta Books, September 1997. ISBN 1-86207-068-7).

Introduction

This is a reply to my critics. It will be updated periodically, as more reviews appear. This version was written in August 1999. The amendments and additions incorporated into this article in addition to the text of the first version (March 1998) take account of additional reviews and discussions of the British edition of the book, published by Granta Books in paperback in September 1998; the American edition, In Defense of History, published by W. W. Norton & Co. in January, 1999; and the German edition, Fakten und Fiktionen: Über die Grundlagen historischer Erkenntnis, translated by Ulrich Speck and published by Campus Verlag, Frankfurt am Main, in October 1998. A Korean edition has also been published, and Japanese, Portuguese, Swedish and Turkish editions are in preparation; my linguistic limitations mean that I will regrettably be unable to reply to any reviews that might appear of them, as of the sole review that has so far appeared in Arabic.

The text of the book was revised and updated for the US edition, partly to rectify mistakes, partly to deal with criticisms and clarify passages that had given rise to misunderstanding, and partly to take account of significant work in the field that had appeared in the year or so since the publication of the first British edition. The book's central arguments, however, remain essentially the same. The US edition also contains some new examples drawn from American history. It serves as the basis for all translations.

Contrary, perhaps, to the impression given in the present reply to critics, the book has met with a very positive response from reviewers and others. But inevitably it has also aroused a good deal of controversy. This reply does not deal with reviews which made no criticism of substance. It deals only with the following critical, though not necessarily always hostile, reviews and reviewers, which are listed here in order of appearance:

1. Michael Burleigh, 'Making history or making-up history?', The Sunday Telegraph 14 September 1997, review section, p. 13.
2. Roy Porter, 'Defending History', BBC Radio 3 Book of the Week, 15 September 1997.
3. Bernard Crick, 'The truth, the whole truth - and nothing but', The Independent Saturday Magazine, 20 September 1997, p. 14.
4. Niall Ferguson, 'History is dead. Long live history!', The Sunday Times, 21 September 1997.
5. Ronald Hutton, 'What is history really about?', The Times Educational Supplement, 26 September 1997.
6. Samuel Brittan, 'The many failings of post-modernism', The Spectator, 27 September 1997.
7. John Charmley, 'Time to move past postmodernism?', The Daily Telegraph, 27 September 1997, p. A4.
8. Steven Kassem, 'In Defence of History', Epigram, 16 October 1997.
9. Keith Thomas, 'A good kicking', New Statesman and Society, 17 October 1997, p. 46.
10. A. C. Grayling, 'Historical truth put on the line', The Financial Times, Weekend Supplement, 25/26 October 1997, p. vi.
11. Joyce Appleby, 'Does it really need defending?'. The Times Literary Supplement, 31 October 1997, p. 10.
12. Steve Smith, 'Truth in an age of challenge', The Times Higher Educational Supplement, 28 November 1997, p. 26.
13. Stefan Collini, 'The truth-vandals', The Guardian, 18 December 1997, p. 15.
14. Daniel Johnson, 'The History Man', Prospect, November 1997, pp. 64-65.
15. Matthew Trinca, 'History's Impossible Dream', The Australian, 3 December 1997, p. 40.
16. Chris Harman, 'Subjects and Objects', Socialist Review 215 (January, 1918), p. 27.
17. Rudrangshu Mukherjee, ‘Clio versus pomo sapiens’. in The Telegraph, Calcutta, 12 June 1998.
18. Peter Schöttler, ‘Prologe im Himmel der Theorie’. in Die Zeit, 10 September 1998
19. Nils Minkmar, ‘Keine Plots zum Holocaust’, in Süddeutsche Zeiting, 7 October 1998.
20. Rebekka Habermas, ‘Wenn Klio (ein kleines bisschen) dichtet’, in Frankfurter Rundschau, 7 October 1998.
21. Peter Ghosh, ‘Laid Down by Ranke’, in London Review of Books, 19 October 1998 (and subsequent correspondence in the LRB).
22. Bernd Roeck, ‘Rächer der Verderbten’, in Frankfurter Allgemeine Zeitung, 3 November 1998.
23. Lynn Hunt, ‘Does History Need Defending?’. in History Workshop Journal, Issue 46, December 1998.
24. Anthony Easthope, in Textual Practice, Winter 1998, pp. 563-66.
25. Roy Porter, ‘The Untrustworthy’, in The New Republic, 14 December 1998.
26. Ernst Nolte, ‘Auschwitz als Argument in der Geschichtstheorie’, in Die Welt, 2 January 1999.
27. Marie Theres Fögen, ‘Geschichte vor dem Zusammenbruch: kann ein Faktenkult sie retten?’. in Neue Zürcher Zeitung, 16/17 January 1999
28. Steve Weinberg, ‘Not all ways of recalling the past are equal’, in Christian Science Monitor, Boston, 21 January 1999
29. David Gress, ‘The "End" of History?’, in Orbis: A Journal of World Affairs, Spring 1999, pp. 314-336.
30. Doug Munro, in Journal of Social History, Vol. 32, Summer 1999, pp. 941-42.
31. Diane Purkiss, 'Response to Professor Richard Evans', originally published in Reviews in History, Institute of Historical Research website (http://www.history.ac.uk/); now archived in the Discourse section of the site.

In addition, this reply also considers some of the points raised in letters published in The Times Higher Educational Supplement on 19 and 26 September 1997 in reply to my article 'Truth Lost in Vain Views', published in the same periodical on 12 September 1997 (p. 18).

The reply tries to deal with the various points made by critics under a number of headings. The criticisms that have been made both of the original hardback edition and of subsequent editions of the book can be summed up very roughly under the following propositions:

1. The book is unnecessary because history doesn't need defending
2. The book is unfairly critical of conservative historians.
3. The book fails to engage directly with the major postmodernist philosophers.
4. The book defends an outmoded empiricist concept of objectivity.
5. The book defends a conservative approach to history.
6. The book's concept of a fact is untenable.
7. The book misunderstands key arguments of the postmodernists.
8. The book is unfair to those it criticises.

I will deal with each of these basic criticisms in turn.

Criticism 1: the book is unnecessary because history doesn't need defending

Several reviewers thought the book was, in a sense, unnecessary. Writing in the Times Literary Supplement for example, Joyce Appleby suggested that the book is only 'defending history from attacks that must surely have vaporized in their own light-headedness'. Its 'destruction of postmodernism may be a little wearying', Samuel Brittan confesses 'with chapter after chapter and example after example'. It could easily be done 'in a few lines', he says.

Appleby's claim that the book 'certainly will please those who think that pre-1970 history needed no defending' is somewhat untypical in this respect. Leaving aside the mysterious question of what happened in 1970 to make history worth defending in some people's eyes afterwards but not before, it seems strange that a book explicitly devoted to defending history (both before and after 1970) should please those who think it needs no defending at all. And in fact this is what a number of reviewers have more or less said. The fact that over 40,000 British 'A' level students took history examinations in 1997 and over 15,000 students were reading the subject at university, history in the same year means, according to Niall Ferguson, that 'history does not need much defending.' Of course, these figures are a substantial decline on those of previous years, the numbers of students taking history at A level have been falling for over a decade, and in the USA there were only a third as many students reading history at university in 1990 as there were in 1970. But this does not seem to bother Ferguson. As an economic historian, he should know that it is trends that count, not single-year figures.

To be sure, the numbers of history students at British universities have been growing in recent years, but this is a reflection of the huge expansion of the British university system that has taken place since the late 1980s, an expansion which has seen the participation rate of 18-21-year-olds in higher education virtually double to its present third from previously less than a fifth. In his review, however, Ferguson does not regard this as a welcome development. Indeed, if there is a threat against which history needs defending, it is according to Ferguson not postmodernism but 'the expansion of the subject at the universities since the 1960s'. Most historians, one would imagine, would welcome such a development, but not Ferguson, who as an Oxford don presumably thinks that universities other than his own and a handful of others should either teach useful subjects like carpentry or not exist at all. What has been the result of this expansion? he asks. 'Cosy enclaves such as "gender studies" or "gay history" exist to protect the talentless from serious intellectual challenge.' It should not be necessary to rebut such a statement of crass prejudice in detail; nowadays, a huge amount of imaginative, pathbreaking and first-rate scholarship is being devoted to these areas.

A more subtle argument is put forward by Daniel Johnson, with implicit support from Michael Burleigh, Matthew Trinca and Steve Weinberg, and that is that 'it is not history that needs defending, but professional historians', and this is what the book sets out to do. History books, both Burleigh and Johnson point out, are more popular than ever; but on the whole those that are widely read, are written by non-academics, and the popular view of history is stamped not by the Oxbridge professoriate but by cinema and television, and by writers such as Antonia Fraser. Roy Porter, indeed, reviewing the book on Radio 3, wished that it had 'looked beyond the walls of academe', just as Daniel Johnson criticises its author for giving 'no credit to those who practise his craft outside the university'. In Johnson's view, therefore, the book is written in defence of the academic historical profession, what Trinca calls 'a capital H History that is professional, discretely packaged with neat sub-specialisms, and manageable', not in defence of history per se. 'In Defence of History remains the swansong of a golden age of academic supremacy.'

This is an interesting criticism, and perhaps the most difficult to answer of all those that have been made of the book from a conservative angle, not least because there is something in it. The book does defend academic and scholarly standards in history and attack those who violate them. I considered at one point including a chapter about popular representations of history but in the end concluded this was a subject too vast to treat briefly, and deserved a separate publication in itself. In Defence of History focuses on the issue of how we define and achieve such things as truth and objectivity, whatever kind of history we are practising. These matters are, to my mind, the same whether you are Antonia Fraser, writing fo r a broad general public outside the academy, or whether you are Joyce Appleby, writing predominantly for a readership of students and other professional historians. It was gratifying, therefore, to see this view confirmed when Antonia Fraser named In Defence of History as her book of the year in the Christmas books issue of The Observer in 1997. Clearly some popular historians at least do think that the book applies to them as well as to their colleagues in academia.

There are two further points worth making in reply to Johnson's criticism. The first is that the book does endorse the writing of popular history by academic historians, and mentions a number of books which provide good examples of this. The more that the gap can be bridged, the better. The second, also mentioned (briefly) in the book, is that it is essentially not true that professional, university-based historians are facing a new challenge from popular representations of history in the media. Popular representations of history have always been widespread, whether in folksong and ballad, saga and legend, or broadside and chapbook, and they have always structured the historical perceptions of the majority. Two and a half thousand years ago, the Greek historian Thucydides complained in the preface to his history of the Pelopponesian War that poets and others were purveying false and imaginary accounts of what had happened, and announced his intention of setting the record straight. In the past, only a tiny minority of the literate and the educated were exposed to professional history and historians.

Today, at the end of the twentieth century, with over a third of the entire population passing through higher education when they reach the end of their teens in most advanced industrial societies, and a growing proportion of mature and part-time students entering an ever-expanding process of lifelong learning, the number of those who have access to and are influenced by university-based history and historians is probably greater in absolute terms than it has ever been before.

This returns us to Ferguson's original point about the popularity of history as an academic subject today, even if he did overdo it a little. So while Johnson is correct to say the book defends history as an academic discipline, principally against its critics within the academy, he is wrong to say it represents in so doing the swansong of a declining elite, and wrong to say that it has no application to non-academic versions of history.

Criticism 2: the book is unfairly critical of conservative historians

As the above discussion reminds us, the review pages of the daily, weekly and weekend quality press in Great Britain are still dominated by a generation of young, conservative historians and journalists who came to such intellectual maturity as they have during the Thatcher years. With the sole exception of John Charmley, writing in the Daily Telegraph, they regretted the fact that, as one of them, Michael Burleigh, put it, 'unfortunately Evans keeps a foot in the "progressive" camp'. Why this should be unfortunate may be clear to the readers of the Sunday Telegraph, who, one imagines, would not take kindly to the pages of their favourite newspaper being filled with articles written by the apostles of Derrida and Baudr illard, but it is not at all clear to me. Nor does Burleigh do very much to make it clearer to those of us who do not automatically assume that everything that is progressive is necessarily wrong. All of Burleigh's criticisms on this point are based on tendentious misrepresentations of In Defence of History. At the risk of descending into triviality, it is necessary to deal briefly with each of these in turn.

Burleigh charges the book with inconsistency in saying on the one hand that postmodernists write jargon and on the other hand that they have encouraged more literary ways of writing history. But there is no inconsistency here, for it is not postmodernist historians (Burleigh's word) who are criticized in the book for jargon, but postmodernist literary theorists who adopt jargon in order to seem scientific. Paradoxically, the book argues, by treating history as literature although not writing it themselves, these theorists have encouraged practising historians to write better. This still seems to me to be an advance.

Burleigh's paranoia about the book's attitude towards well-known, mostly conservative historians leads him into misrepresentations even more crass than this. 'Andrew Roberts', he thunders, 'is ticked off for his personal sources of income; John Vincent for writing in a tabloid newspaper; Arthur Marwick for being slow with his inaugural lecture. Gossip involving Hugh Trevor-Roper figures more prominently than his major contributions to European history.' Let's look up the passages in question and see what they really say. On page 210 the book says the following of Andrew Roberts, after outlining John Vincent's argument that historians nowadays are all left-liberal because they work for the state:

If we really believed his somewhat reductionst view, we would only expect robust right-wing opinions from historians who lived off a private income, and while it is certainly true that there are some individuals, such as Andrew Roberts, author of the neo-Thatcherite text Eminent Churchillians, who fit into this pattern, there are many more who do not.

No hint of criticism here: Roberts is merely used as an example in an argument.

What about Vincent himself? There are many references to him in the book a good number of them favourable, including, in the argument just quoted, a mention of the fact that his most original and unconventional historical work was written while he held a university appointment, as indeed he still does, and was not dependent on a private income, which to my knowledge he has never been. But the one to which Burleigh presumably objects is the following, in note 15 to Chapter 4, on page 268:

Vincent...cites with approval A.J.P.Taylor's belief that no new secrets were to be found in unpublished archival sources. Vincent himself advocates paying greater attention than is customary to the newspapers, and declares that newsprint is 'the Venetian archive of tomorrow'....Vincent's partiality to newspapers as unbiased sources of information may possibly be coloured by the fact that he himself wrote a column in the Sun for a number of years, until student demonstrations at his own university forced him to abandon it - an event which, however deplorable in itself, would scarcely have taken place had the column been a model of impartiality.

This passage also enraged Daniel Johnson, who devoted a whole paragraph to denouncing it and accused the book of unfairness towards Vincent and 'justifying the silencing of a scholar by the mob'. But the passage in question was intended to suggest that newspapers have their biases, not just in columns written by moonlighting dons, but also in their reporting and their general editorial line, whether it be that of the Sun, the Telegraph or the Guardian. The use of the word 'deplorable' gives the lie to Johnson's suggestion that I actually approved of the demonstrations in question. As for Burleigh's objection, nowhere in the passage is there any hint of disapproval for Vincent's work for the Sun in itself.

Similarly, on page 299 the reference to Arthur Marwick having delivered his 'inaugural lecture after decades of occupation of the Chair of History at the Open University' contains not so much as a hint of criticism for the delay, however critical the discussion of the lecture is in other respects. The criticism is a pure figment of Burleigh's imagination; I included the delay in the reference merely as a curious and possibly amusing fact.

As for Hugh Trevor-Roper (Lord Dacre), it is difficult to work out what Burleigh means by 'gossip', unless it is perhaps a reference on page 19 to his 'authentication' of the Hitler diaries, which is not gossip but fact, on the basis of a cursory examination of the 'diaries' in a Swiss bank vault - all of which is documented in Robert Harris's excellent book Selling Hitler. Why the book would have been improved by mentioning Trevor-Roper's other work on European history is a mystery. His important work on seventeenth-century British history and his acute review of Carr's What is History? are both mentioned favourably, and at length. This clearly isn't enough for Burleigh. But mentions of work such as this are all designed to illustrate arguments in the book; it's not about giving out pluses and minuses to practising historians on the basis of a complete assessment of all their work. Most of Trevor-Roper's (often important and influential) work on European history is simply not directly relevant to the arguments advanced in my book. Failure to mention it is not intended as a slight.

By contrast, Burleigh finds the book gives 'generous treatment' to 'the American neo-Marxist David Abrahams' (Burleigh is not the only reviewer to misspell Abraham's name) in its account of the controversy over the gross errors in his book which led to his losing his job as a historian. At least Stefan Collini finds my book's handling of the Abraham case 'exemplary'. Given the fact that the book's account convicts Abraham of subconsciously manipulating evidence to fit a thesis, inability to tell fact from fiction, failure to follow basic historical procedures, poor research skills and excessive haste in research, it is difficult to see how its treatment of him could be fairly be described as 'generous'. I cannot imagine that Abraham and his friends (including Peter Novick) are very pleased with the account. Burleigh is imagining things here again.

Michael Burleigh is not the only young fogey to be upset by the book's criticism of admired figures on the right. Niall Ferguson also convicts the book of 'rudeness' towards 'historians of broadly conservative inclinations' (Steven Kassem alleges that the book's treatment of those it disagrees with is abusive; Keith Thomas thinks it is brisk; Bernard Crick thinks the book deals with them 'clearly and calmly, treating fools with scholarly courtesy'; German reviewers have commented on what they see as its unpolemical character). Let us take in turn the examples Ferguson cites.

'John Kenyon, for example', he says in his review of In Defence of History, 'is (found) guilty of "mental insularity and prejudice"'. This remark, which Ferguson quotes correctly from page 179, is a comment on Kenyon's claim, advanced in 1993, that undergraduate courses on the history of West Africa and Indochina were 'hastily cobbled-up' by history departments acting on the dictates of fashion, but soon faded away because the difficulty of finding competent external examiners left students 'at the mercy of their tutor's whims and prejudices'. I have to say that (a) the book's description of this view as reflecting its author's mental insularity and prejudice is no ruder or more insulting than Kenyon's own remarks are to the many distinguished historians of Africa and Indochina teaching and researching in British universities, and (b) it seems to me to be entirely justified to call it insular and prejudiced.

Ferguson goes on to attack the book for describing Hugh Trevor-Roper as 'stubbornly parochial'. This phrase refers not just to Trevor-Roper but to 'a number of stubbornly parochial specialists in European history', among whom he is placed by his remark that Africa has no history, merely 'the unrewarding gyrations of barbarous tribes in picturesque but irrelevant corners of the globe'. Once more, the term used seems to me merely to be accurate rather than to be rude as a way of referring to this remark by Trevor-Roper, who in the course of his long career has proved himself no amateur at invective himself.

Finally, Ferguson takes exception to the book's remark that John Vincent's comment that 'History is deeply male...History is about the rich and famous, not the poor' shows its author's 'deep ignorance of other kinds of history than the history of British high politics in the nineteenth century which he himself writes'. By omitting the context of the remark and all the words in the quotation except the first too, Ferguson gives the impression that the book is claiming Vincent suffers from 'deep ignorance' in every respect, which is not true at all, of course. Apart from this, the book's statement seems to me once more to be entirely justified, and no less insulting to Vincent than he is to those many fine historians who have devoted their lives to writing about the female, the poor, and the unknown. In a similar way it would not seem unjustified to describe Ferguson's own remarks about gender history as deeply ignorant. It may be rude to say so, but it's also true.

All these points are in the end somewhat trivial ones. More serious by far is the one made, yet again, in the most intelligent of the critical reviews to appear immediately after the book's publication, namely Daniel Johnson's article in Prospect magazine. Johnson argues that the book is 'not so much a defence of history as of one school (left, populist, socially conscious) against another school, less fashionable but no less respectable (conservative, elitist, high political)'. He points out that intellectual history is elitist, economic history can be right-wing, and there are even socialist historians of high politics. But these points, while perfectly valid, do not alter the fact that there are a number of historians (including, again, Kenyon and Vincent, but also others who write prescriptively about how history should be researched and studied, such as Gertrude Himmelfarb) who dismiss all kinds of history beyond that of high politics within the framework of nation-states as trivial, irrelevant, or not history at all. It is against this narrow conception of history that the book is in part arguing, and it is an inescapable fact that the majority of those who believe in it are politically conservative, although the book goes to some lengths to point out that E. H. Carr, a Stalinist if ever there was one, dismissed the history of the great mass of ordinary and people and unpolitical subjects as irrelevant as well.

But while it argues that history needs defending against the crippling and stultifying influence of narrow and elitist conceptions of what it is, or should be, about, the book does not argue for any one particular conception of history in itself as an alternative. Nor does it say that intellectual history, the history of elites or the history of high politics are in any way wrong in themselves. Its purpose here, as a number of other reviewers, from Ronald Hutton to Roy Porter, have recognized, is to argue for a broad, inclusive conception of history, to celebrate its diversity while at the same time as trying in various ways to overcome its fragmentation, and nowhere does it in fact argue that left-wing versions of history are inherentl y superior as history to those practised on the right. Indeed it has been sharply attacked on the left, for example, by the convenor of the London Socialist Historians' Group, Keith Flett, who thinks that the book needed to be 'more hard-hitting still to repulse the postmodernist challenge', and the Trotskyite writer Chris Harman, who convicts it of 'backsliding towards postmodernism.'

Criticism 3: the book fails to engage directly with the major postmodernist philosophers.

In Defence of History is a book that emerged out of the experience of teaching. In the mid-1990s, I agreed to take on the history element in a series of lectures and classes on the epistemology of the social sciences in a joint BA degree in Politics, Philosophy and History at Birkbeck College, University of London (where I was Professor of History from 1989 to 1998). Birkbeck's undergraduates are all mature part-timers, and teaching takes place in the evening. Teaching and teaching materials have to be both accessible and exciting, and they also have to be capable of satisfying the demands of an unusually critical and sceptical student body.

Looking around for suitable textbooks on historical epistemology, I was dismayed to find that the most widely used works in this subject were more than thirty years old - E. H. Carr's What is History? and G. R. Elton's The Practice of History. Rereading these books for the first time in many years made me only too aware of their inadequacies. Worse still were the alternatives. On the one hand there were tedious history primers which talked down to the student in patronising and didactic tones. On the other there was a rash of new books, many of them explicitly designed for teaching, which disputed the legitimacy of history altogether and mobilised postmodernist theory in the service of the argument that it was no different from fiction or poetry. A glance at journals like History and Theory made it even clearer that this was now more or less the orthodoxy among those who specialized in the theory and philosophy of history, which indeed had become a separate branch of learning in its own right since the days of Carr and Elton (Steven Kassem, who finds it 'puzzling' that many of the writers discussed are not practising historians, has not been paying attention).

It seemed to me that there was a need for a new book that tackled these problems, as Carr and Elton had tackled the problems of their own day, from the point of view of the practising historian, and gave readers some examples of what historians were actually doing in the 1990s as well as putting forward some general arguments about it. The lectures I gave on the course duly became, after many revisions, In Defence of History. In the book, I tried to steer a middle course between the extremes of postmodernist hyper-relativism on the one hand, and traditional historicist empiricism on the other. But the weight of the argument was directed against the postmodernists. This was not because I thought, or think, that they are becoming d ominant in the historical profession as a whole. On the contrary, their presence is felt amongst working historians in only a relatively few areas, such as European intellectual history, or some kinds of feminist history, for example. The reason why I felt history needed defending was principally because of the dominance of hyper-relativism and scepticism about history's validity as an intellectual enterprise amongst those who write about historiography and history as a discipline in a general, theoretical sense.

In the 1990s increasing numbers of history degree courses, both undergraduate and postgraduate, have included a substantial element of historiography in them. It seemed to me that it was at this level that the intellectual validity of history really needed defending in the light of the kind of literature that was, and is still being published on the subject. Beyond this, too, growing numbers of historians, particularly on the left, have been voicing a growing degree of scepticism about the possibility of doing history at all. More broadly still, historians of all persuasions have been talking in increasingly alarmist tones in recent years about the state of crisis they see as overwhelming their profession as a whole. Here too, I wanted to persuade people that it was possible to defend history as an intellectual undertaking by genuinely confronting and arguing with the extreme sceptics rather than by simply ignoring them or covering them with abuse.

These aims, I thought, were clear enough from the book. But a number of critics mistook them and thought the book was trying to do something else, or ought to have done something else. Rebekka Habermas found fault with the book because it failed to refute the central philosophical theses of Derrida and Barthes. Bernd Roeck complained similarly that the book did not tackle Dilthey or Gadamer, Lyotard or Foucault, but concentrated its fire instead on second-rank figures. Marie Theres Fögen attacked me for ignoring Kant. Another reviewer thought I should have dealt with Hume. Anthony Easthope pilloried with the book's failure to provide a full-length discussion of Foucault or devote a Chapter to Hayden White.

All of these omissions I gladly concede, apart from Hayden White. Although there is no separate chapter devoted to White's work, his arguments are discussed or referred to on 26 pages of the book, that is, over ten per cent of its pages of text, and they are not evaded, as Easthope alleges, but fully dealt with, and indeed to a considerable extent, at least in their most recent version, endorsed. As for the rest, yes, it is true that I have not tried at length to deal with the epistemology of Derrida, Foucault, Gadamer, Lyotard, Hume, Kant and the rest. This complaint about the book's failure to deal at length with the most important postmodernist thinkers (not to mention major epistemologists down the ages) is in effect demanding a dif ferent book from the one I wrote.

But this does not prevent reviewers from asking for such a book. Steve Smith demands 'more philosophical spadework' to clear up some of the issues at hand, and Appleby also charges that 'what is lacking in A Defence of History' (it is symptomatic of her scholarly standards that she even manages to misquote the book's title) 'is a serious discussion of the way that postmodernists have examined the tools and strategies involved in the production of knowledge.' In similar vein, Stefan Collini is critical of the fact that the book tackles what he calls 'a blend of vulgarised post-structuralist theories of language with self-consciously radical notions about contesting hierarchy, hegemony, and so forth' instead of grappling with the French epistemology which is behind all this 'at the appropriate level of abstraction'. The result, he says, is a series of 'vulgarised rebuttals of vulgarised ideas'.

If by 'vulgarised' he means 'popularised', then I gladly plead guilty. The book is essentially not about Michel Foucault, Jacques Derrida, Ferdinand de Saussure, or Roland Barthes, though they do get a mention, but much more about Hayden White, Frank Ankersmit, Keith Jenkins, Patrick Joyce, Dominick LaCapra, Beverley Southgate and their likes; it's precisely not an attempt to tackle the complex epistemologies of modern French philosophy head-on.

What it precisely is, is an attempt to deal with the application of some of these epistemologies, often in a simplified and, yes, vulgarised form to the theory and practice of history by people whose intellectual home is half-way between the two disciplines of history and philosophy. Collini's dismissive argument that historians aren't qualified to write about French epistemology, coupled with his view that a vulgarised rebuttal of vulgarised versions of them isn't really worth writing, is to my mind excessively Mandarin. If the only kind of book that's worth writing about postmodernism is a philosophical dissection of the central theories of Foucault, Derrida and other French thinkers at a high ('appropriate') level of abstraction, the n the field really will be left open to the Jenkinses and the Ankersmits and the Southgates. Somebody's got to take them on at their level; that's what the book tries to do, and from the point of view of the health of the discipline of history, I remain convinced that this is certainly something worth doing.

Moreover, it is misguided of Collini in my view to tell historians, as he does, just to get on with writing history and leave the theorizing to the experts. We all have to make some fundamental decisions about what we're doing nowadays before we do it. History is becoming a more theoretically and epistemologically self-conscious discipline, and that's all to the good. To follow Collini's advice would not only be foolish, it would not really be possible either any more. One feels, somehow, that there's an element of 'get off my patch' in this review too. To reiterate: In Defence of History is not supposed to be a work of philosophy. It's about the influence of post-structuralist and other postmodernist epistemologies on historical theory and practice, not about those epistemologies themselves. It's a book that emerged from, and is intended to be used in, history teaching. It's not a monographic study in philosophy or the history of ideas.

That is why, finally, the book does not discuss 'important thinkers' such as Michael Oakeshott, whose omission is deplored by Niall Ferguson. I read Oakeshott and thought seriously about including a discussion of his writings on the philosophy of history, which are both extensive and important. But the fact is that he plays no part whatsoever in the current debate about history, which has entirely different theoretical and philosophical origins, and it would have been both confusing and unnecessary to have included him.

Sir Herbert Butterfield, whose omission Ferguson also deplores, is another matter. If he is remembered nowadays for anything, it is for The Whig Interpretation of History, an attack on teleological history written in the 1930s, the main argument of which is that what happens is often an unintended consequence of something else. This is such a truism that it hardly seemed worth mentioning, and in any case it was put much earlier, and much better, by Frederick Engels, with his notion of history operating through a 'parallelogram of forces' in which pressure at one point would result in movement, unintended by the author of the pressure, at some other point far away. Like Oakeshott, Butterfield doesn't really feature in the current debate; unlike Oakeshott, he made to my mind no really serious contribution to the theory and philosophy of history, so the case for including him is even weaker.

Criticism 4: the book defends an outmoded empiricist concept of objectivity

Joyce Appleby, like the German reviewer Peter Schöttler, thought that I was trying to discredit postmodernism as a whole by attacking its silliest and most extreme manifestations. But in doing so, she succumbed to precisely the kind of dichotomizing thinking of which her review argued postmodernism had taught scholars (though not apparently myself) to be wary. As Appleby surely knows, and as the book makes perfectly clear, 'postmodernism' is a label covering a wide variety of positions, and just because the book rejects some of these, it doesn't follow that it rejects them all; the same point also applies to Schöttler's criticism of the book as an out-and-out polemic against postmodernism in all its manifestations. Pursuing this Manichean misreading of the book, Appleby goes on to convict it of an outdated belief in the historian's 'impartial and omniscient voice of a reconstructor of events and analyser of developments.' In fact the book uses up a good deal of space trying to demolish such simplistic denials of the historian's subjectivity, and to deny the possibility of historical omniscience.

'Historians', she goes on, 'prefer to think that they reconstruct the past, rather than acknowledge that in the process of doing so they also create texts and construct knowledge.' Contrary to what readers of her review might suppose, the book defends the latter position and deploys a range of arguments against those who, like Elton, take the former. On pages 125-6 of the book, far from ignoring Hayden White's ideas on this subject, as Appleby claims, the book explicitly endorses White's most recent account of how historians go about their work of writing and research. Thus Appleby's patronising conclusion that British historians preserve simplistic ideas about truth and objectivity which have long been discarded in the more sophisticat ed intellectual atmosphere of the United States is based on a superficial misreading of the kind she claims the book is guilty of when it tackles the postmodernists itself.

The reason why the book refers so frequently to Elton and Carr is simply that, as many historians of a variety of different persuasions have conceded, their books still form the basis for much if not most teaching of historical epistemology today. They are still much the most readable and approachable accounts of the nature of historical knowledge written by practising historians. Lynn Hunt finds this a 'disappointing...flight...into "daddyism", the search for fatherly figures that warrant the legitimacy of an approach. Evans', she says, 'refers almost obsessively to Carr and Elton as if to reassure himself that he will be their appointed successor; his criticisms of them sound like those of a deferential grandson.' In fact, In Defence of History is intended, not to replace their books, but to be read alongside them; hence the chapter headings, borrowed from Carr, or the final paragraph, a parody of Carr's final paragraph. Thirty or more years ago there were hardly any books which introduced history students to the conceptual and methodological problems which they faced, so Carr and Elton virtually had the field to themselves. Now there are many such competing texts, and the very idea of any single book achieving the kind of status theirs did is ridiculous.

Matthew Trinca ignores the carefully circumscribed definition of objectivity given in the book's final Chapter, and, simply assuming that the word can only be used in the absolutist sense intended by traditionalists like Elton, places the book's central argument squarely in the Eltonian camp. While Trinca obviously disapproves of this, David Gress, advocating a return to military history and the 'rote learning of facts' in school history lessons, sees it as a confirmation of his conservative approach. Gress enlists In Defence of History in support of his call for a rejection of the 'present-mindedness and political correctness' through which he thinks social history, women's history, and other relatively recent new developments in historical scholarship and teaching have created a 'barrier to learning'. Thus Gress finds the book's criticisms of Elton to be an 'almost ritualistic' cloak for 'true respect and admiration' on the part of its author (myself). A similar position is taken in the review by Greg Munro, who thinks the book does indeed 'embrace wholeheartedly' the 'hard-line concept of historical objectivity espoused by Elton', for all its protestations to the contrary.

With friends like this, who needs enemies? Such a reading of the book, however complimentary it might be, is, of course, extremely selective. The book actually rejects Elton's simplistic view that the historian could shake off all present-day beliefs and ideas when studying the documents. It rejects his narrow-minded empiricism and his parochial insistence on the political history of the nation-state as the only real history. It rejects his belief that interpretations sprang unaided from the sources. It rejects his whole account of how historians go about the business of research. It explicitly welcomes developments which he found objectionable, such as women's history, social history and so on - all these are, after all, areas of history which have been the subject of much of my own work. On all these points, I side with Carr. On the other hand, this doesn't mean that I'm simply a follower of Carr. The book rejects Carr's account of objectivity and causation, for example, since both of them are tied to his belief in the defining quality of a human future locate inevitably in a Soviet-style planned economy.

Given the fact that the book strongly rejects both Carr and Elton's central ideas about objectivity and many other things, it is difficult to see how it can be understood as using them as ways of legitimating my own approach. More profoundly, as the book makes abundantly clear, obscurantist conservatism of the sort to which Elton too often fell prey, and Stalinist Marxism of the sort which ultimately lay at the root of Carr's work, are both objectionable as well as inadequate bases for historical scholarship and thought. On the other hand, I don't see why I should not be reasonably courteous in advancing such criticisms, severe though they are. A polite style of criticism doesn't in any way imply a deference to the views of those who are being criticised.

Hunt, in other words, is mistaking style for content. This is particularly clear when she comes to deal with the book's account of Carr's notions of causation, which she considers to be merely a 'rehash' of Carr's own views. The conclusion she cites from the book, that 'Carr did not really think his argument through' comes, contrary to what Hunt maintains, not at the end of a mere repetition of his views, but at the end of a lengthy and reasoned criticism of them. This criticism is based principally on my objection to Carr's view that historians should only be interested in understanding the cause of a past event if it could help us change the present and shape the future.

The implication of this position is that the historian should suppress any mention of any causes of a past event, whether accidental or not, if it does not fulfil this purpose. Carr did not mention this (extremely disturbing) implication of his method, however, and that is why I felt justifying in suggesting that he did not really think his argument through. I don't think this is the mere bald assertion Hunt claims it is; no more than is the conclusion of the paragraph-long criticism in the book which is directed at Hunt's own attack on Carr's notion of causation in her (generally interesting and often persuasive) book, written with Joyce Appleby and Margaret Jacob (Telling the Truth About History) that Hunt's own argument on this point is 'factually incorrect as well as irrelevant'. Perhaps her irritation at this criticism has prompted her irritation with my book as a whole.

At times Hunt's misreading of the book becomes almost comical, as when she objects to the sentence in the Chapter on causation where I remark that social and economic history have become the 'principal victims' of 'the new theories'. 'What does this have to do with causation?' she asks rhetorically. The answer is, not a lot, because it's the last sentence in the Chapter and is intended to point the way to the next Chapter, which is precisely about social and economic history. Let me quote the sentence in full, to make it clear how misleading Hunt's paraphrase of it is:

While cultural history, intellectual history and even the history of high politics have received a fillip from the new theories and aproaches of the late 1980s and 1990s, the principal victims have been social and eonomic history; precisely the areas, as we shall now see, which experienced the greatest growth and expansion in the 1960s and 1970s (my italics here).

And this final sentence of the Chapter comes at the end of a final paragraph, also transitional to the following Chapter, which deals with the problem of looking for a context through which to explain a past event or process, and makes it clear that the social and economic context favoured by historians like Carr has been challenged as an explanatory tool by the rise of cultural history.

Hunt's misrepresentations continue even when she is positive about the book, as in her view that it is at its best when it 'unravels controversies that have nothing to do with epistemological issues, such as the one about David Abraham's reading of sources in his analysis of business interests in Nazi Germany'. The whole point of recounting this controversy, however, as the book makes quite clear in the course of its analysis of the affair, is that it has a direct bearing on central issues of historical epistemology: namely the relationship between fact and interpretation, the possibility of falsifying interpretations by referring to evidence, and the argument of Hayden White that a Marxist historical interpretation such as Abraham's ca nnot be disproved because what Marxists view as relevant evidence is not so viewed by non-Marxist historians.

Anthony Easthope's too thinks that that In Defence of History endorses a 'resoundingly empiricist' notion of objectivity. Let us examine the quotations from the book he uses to support this interpretation. First: 'the past does impose its reality through the sources in a basic way'. The key word here is 'basic', as the following sentence on page 115 of the book makes clear: 'At the most elementary level, one cannot simply read into documents words that are not there.' Second: 'the past does speak through the sources' (p. 126); well, yes, but the whole of the preceding discussion, which includes a lengthy account of the 'Abraham affair', makes it clear that this is not a blanket claim about how historians do their work (merely lis tening to what the sources say), it is a limited and partial claim; it does not end there; in fact it is only half the argument. What the book actually argues is that the past only speaks through the sources when the historian interrogates them. It's a two-way process, as the book makes clear time and again; so Easthope is misleading his readers when he gives the impression by selective quotation that it argues that the past speaks through the sources unaided and that in so speaking it says all there is to be said.

Easthope concludes this part of his argument by citing the book's 'resoundingly empiricist conclusion that, despite it all, "it really happened", we can "find out how" and know "what it all meant".' What this passage in the book actually says, following on a series of references to the theses of various postmodernist writers on history, is as follows: 'I will look humbly at the past and say despite them all: it really happened, and we really can, if we are very scrupulous and careful and self-critical, find out how it happened and reach some tenable though always less than final conclusions about what it all meant.' By removing all the qualifications from this statement, Easthope has manipulated it to give the appearance of a dogmatic c onfidence in the possibility of absolute knowledge that was a very long way from what was intended. Thus he has only managed to support the charge he levels against the book, of mindless empiricism, by giving a deliberately false and distorted version of what it actually says.

Because there seems to be a quite widespread misunderstanding of what I mean by historical objectivity, with numerous reviewers assuming that because I actually dare to use the word it can only have one meaning, that is, a strong and traditional one, rather than, as I intended, a weak and qualified one, I have added a sentence to the penultimate paragraph in the American and German editions: 'Objective history in the last analysis is history that is researched and written within the limits placed on the historical imagination by the facts of history and the sources which reveal them, and bound by the historian's desire to produce a true, fair, and adequate account of the subject under consideration.' Any questions this might seem to beg are, I hope, dealt with in the preceding text of the book.

The need for such an addition is further illustrated by the misunderstandings present in the review by the Soviet history specialist Steve Smith, who supposes that In Defence of History argues that 'there is a singular truth to be told about the past' and that it can be 'discovered' from the evidence. The book goes to some length to argue that what historians write is the result of a dialogue between their own purposes and ideas and what they find in the sources. Unwilling to recognise this, Smith proceeds to entangle himself in a web of contradictions, as he argues on the one hand that historians' interpretations of past events cannot stand or fall by the extent to which they conform to the historical evidence, and on the other accuses the Harvard historian Richard Pipes of providing 'a deeply distorted representation' of the Russian revolution.

How can Smith consider Pipes's account to be distorted if not by an appeal to the evidence? Not just the evidence which Pipes himself cites, that is, but more importantly, perhaps, the massive amount of evidence which Pipes has omitted from his account. Saying that he, Smith, brings a different narrative perspective to bear on the Revolution doesn't amount to any criticism of Pipes at all, since in this way of thinking there is no criterion by which anyone can decide whether it is Smith's narrative or Pipes's which is distorted; the inescapable conclusion of Smith's general argument is that both are equally valid, just as the inescapable conclusion of Smith's particular remarks about Pipes is that this general argument is invalid.

A somewhat different confusion appears in Ernst Nolte's review. Nolte agrees that '"Auschwitz"...is probably in reality the most effective argument that can be directed against the "postmodernists" and their concepts of "text", "invention" and "fictionality"'. However, he goes on, this does not mean that any challenge to the accepted historical representation of Auschwitz is tantamount to Holocaust denial. Is it, he asks, a form of Holocaust denial to dispute the once-widespread claim that four million people were gassed in Auschwitz, or to note, as Daniel Jonah Goldhagen does, that the efficiency of the gas chambers has been exaggerated, or to concede that Binjamin Wilkomirsky, author of an acclaimed eyewitness account of Auschwitz, was never actually in the camp at all?

Auschwitz is not an invention, Nolte says, but it has constantly to be rediscovered and reinterpreted. No generation has the right to close off research for the future by declaring we know all we can ever know. Auschwitz must be studied with the same historiographical tools of source-criticism and so on as any other subject. And it must constantly be compared with other genocides. Who knows whether, at some future time, its singularity will be compromised by some other case of mass murder similar in scope and method? Historical relativization, Nolte concludes, is not the same as moral relativization. Moreover, incorrect arguments are often beneficial to scholarship because they sharpen and improve better arguments.

Nolte suggests that if I follow my own principles, I should agree with these points. He suspects I do not. But he is wrong. I do agree with them, at least, in the way that he puts them in his review. Unfortunately, what he is really talking about here is not In Defence of History at all, but another book which I published ten years ago, In Hitler's Shadow, in which I dealt with (West) German views of Nazism and strongly criticized Nolte's own writings on the subject, particularly his claim that hard-line Holocaust deniers, people who allege that there was no Nazi policy of exterminating the Jews, that there were no gas chambers, that the number of Jews who were killed was far smaller than six million, were honourable people whose views should be taken seriously. In his book Der europâische Bürgerkrieg, indeed, Nolte even took over standard theses from the Holocaust deniers, including the (demonstrably false) allegation that the Jews declared war on Germany in 1939 and therefore Hitler was justified in 'interning' them in concentration camps.

The point here is that there is a difference between legitimate reinterpretation and deliberate invention and falsification; between debates over the significance of historical events and attempts to deny their existence altogether; between the proper methods of historical scholarship such as source-criticism, and the distortion, manipulation and suppression of historical evidence by Holocaust deniers such as Arthur Butz and Paul Rassinier, whom Nolte defends in his book. Of course it isn't a form of Holocaust denial to question the authenticity of Wilkomirski's memoirs. Of course Goldhagen was not flirting with Holocaust denial when he made his remark about the efficiency of the gas chambers, because what he meant by this was that only a portion of Europe's Jews were killed by gassing; millions were killed by mass shootings, a fact he claims is too often forgotten in the literature on the Holocaust. And of course it is not Holocaust denial to point out, as has been known at least since the postwar publication of the memoirs of Rudolf Höss, the Commandant of Auschwitz, that the best estimate for the number of victims of gassing there was slightly in excess of one million, not the four million that has sometimes been claimed.

All this has nothing to do with Holocaust denial, and I am not aware of anybody apart from Nolte who has ever claimed a connection. What Nolte seems to be suggesting is that my concept of historical objectivity rules out any dispute of any kind about Auschwitz. But it does not. Of course there is room for argument and debate. But only within the limits set by the evidence. A claim such as the familiar Holocaust denial assertion that nobody was gassed at all clearly steps beyond these bounds. Nor do I think that Holocaust denial has performed any real service to genuine scholarship, because it has nothing to do with reinterpretation. On the other hand, the work, say, of Goldhagen, though in my view demonstrably wrong on a number of accounts, has stimulated historians to think afresh about such issues as the reasons why ordinary men, or ordinary Germans, participated in the extermination of the Jews, the extent to which antisemitism was increasing during the Weimar Republic, and so on. In other words, on the evidence of his review, Nolte and I are in broad agreement on all the issues he raises. On the evidence of my In Hitler's Shadow and his Der europäische Bürgerkrieg, however, we are still a long way apart.

Criticism 5: the book defends a conservative approach to history

If Nolte misunderstands my notion of objectivity in one way, then Peter Ghosh misunderstands and misrepresents it in a whole variety of ways. In his review, he claims that the book is engaged in a 'polemic against history since 1960', that it defends an 'exaggerated empiricism' based on the 'fetishising of documents' and that I believe that facts and documents 'speak for themselves'. It is hard to believe that Ghosh has read the book properly, or that he is familiar with my other work. In Defence of History actually argues that history has undergone a welcome renaissance since the 1960s, and defends explicitly and at length in Chapter 6 the broadening of the discipline which has taken place in this period against conservative historians who would like to see it return to its old concentration on the politics of the nation-state. My own work on the social history of medicine and disease, the history of ritual, the history of feminism, the history of death, and other topics, would not have been possible without all the changes that have taken place in historical studies over the last forty years.

In a similar way, Peter Schöttler claims that the book avoids discussing, indeed basically dismisses, the new cultural history which has (in his view) been one of the principal and most welcome consequences of postmodernism.

Someone who says goodbye to optimism about progress and industry is not by a long way a rural romantic or an enemy of the Enlightenment; someone who does not derive people's statements and behaviour a priori from rational interests and intention but asks about gender and sexuality, desires and anxieties, fantasies of power, and mistakes, is no irrationalist, but takes account of the fact that subjects have a body and an unconscious.

I couldn't agree more; indeed I've tried to do all these things in my own work, particularly in Rituals of Retribution and Tales from the German Underworld. Schöttler's remarks here miss their target as much as do those of Rudrangshu Mukherjee, writing in the Calcutta Telegraph, when he declares that history should take from post-modernism 'those aspects of it which add value to the work of historians' and accuses the book of trying to close the window of opportunity which postmodernism offers. But on the contrary, the book welcomes the challenge which postmodernism issues to historians to reflect more carefully on their own work and their own position, and to take language more seriously. Both Schöttler and Mukherjee fail to note the distinctions it makes and so criticise it for things it doesn't say or try to do.

Ghosh's criticisms go beyond his claim that the book defends an old-fashioned concept of history. Ghosh takes the historicist point of view that 'theory comes from within history', and excoriates the historian's use of theories taken from other disciplines. This is sheer obscurantism. Most advances in historical scholarship since the 1960s and long before have taken place through the use of theories and methods borrowed from elsewhere, whether philology, economics, sociology, anthropology, or linguistics. In Defence of History argues that facts and documents do not speak for themselves, but only speak to when they are spoken to by the historian. Historians need to use, indeed cannot avoid using, theories and concepts developed in their own time. Ghosh also claims that if any aesthetic impulses go into the structuring of a work of history, that work of history must necessarily be conceptually vacuous. He refers to my book Death in Hamburg as an example (because I cite it in In Defence of History as an instance of how I used aesthetic criteria to organise historical material, much in the way described by Hayden White for historical work in general). If Ghosh had actually read Death in Hamburg, however, he would have discovered that it is structured by a set of Marxist concepts. The point is that conceptual and aesthetic aspects of writing a history book are not mutually exclusive; if they were, all history books would be completely unreadable.

The confusions and contradictions in Ghosh's review are compounded in his subsequent response to the points made above in a letter to the London Review of Books, where he goes on to declare that a 'Rankean' approach to history as described in my book is an invention of the last thirty years, and asserts roundly that I am a Rankean. 'Rankean' in this sense seems to involve the kind of documentary fetishism that Ghosh accuses me of in his review. But what could fetishise the documents more than Ghosh's own apparent belief that interpretations and theories in history can only be legitimate if they spring from history itself, or in other words, from the documents?

Moreover, since Ghosh also argues that it is illegitimate to describe anyone as postmodernist unless they describe themselves as such, he will presumably have to accept my insistence that I am not a Rankean even though he calls me one. Insofar as In Defence of History endorses 'Rankeanism', it is only in the sense of insisting that the methods of source-criticism introduced, not in the last few decades, but in the mid-nineteenth century by Leopold von Ranke himself, are still valid for historians today when they are analysing documents in the archive. This does not imply endorsement of Ranke's whole theory of history, however, still less of the way his name has been invoked by methodologically conservative historians wishing to reinstate political and constitutional history at the centre of the discipline, as the book makes abundantly clear in its critical discussion of both these topics.

Where Ghosh is coming from is indicated by the aggressive 'get off my patch' tone of his letter to the London Review of Books, in which he states that only professional historians of ideas such as himself are qualified to write about historiography, and mere historians of politics, culture and society such as myself should steer well clear. He justifies this by claiming that my portrait of the history of ideas as the continual reinterpretation of a limited number of classic texts 'bears no relation to the modern discipline of the history of ideas'. He goes on to note, however, that the 'old version' which I equate with the discipline as a whole 'actually resembles Post-Modernism in the style of Hayden White or Ankersmit'. This is precisely the point made in my book. White is surely for good or ill the most influential historian of historical ideas and practices writing today. All his numerous disciples and followers argue that all texts are capable of infinite reinterpretation. White's approach is not old-fashioned; quite the contrary, it is all too fashionable.

What Ghosh seems to be advocating is the method pioneered by Quentin Skinner, of fixing the meaning of a past text by comparing the language it uses with the language used in other, contemporaneous texts: thus we will be able to determine once and for all the meaning of Hobbes's Leviathan, for example, by fixing the meaning of the language and concepts it uses through locating them in their contemporary linguistic context. This is a very fruitful and influential approach, but it is not synonymous with the entire 'modern discipline of the history of ideas', and indeed postmodernist writers on historography regard it (without much justification, it has to be said) as an outdated form of historicism.

It seems to be this approach which leads Ghosh to argue that modern concepts should not be imported into the analysis of past texts. This may be an admirable precept when analysing the meaning of the writings, say, of John Locke, but it will not do when studying, for example, the social structure of a medieval village, or the origins of the First World War. Moreover, if Ghosh believes, as he says in his review, that the 'Great Texts' notion of the history of ideas is now 'defunct', then why is he 'working on a new translation of Max Weber's Protestant Ethic', as his biographical notes in the LRB reveal? Ghosh alleges that 'Evans supposes that the modern history of ideas, as practised by Quentin Skinner and Gareth Stedman Jones, for example, and "Post-Modernism" of the White and Ankersmit variety are essentially one and the same.' Nowhere in anything I have written is there the slightest warrant for such an absurd accusation.

Criticism 6: the book's concept of a fact is untenable

Perhaps the most extreme depiction of the book as a defence of a mindless empiricism is provided by Marie Theres Fögen's review in the Neue Zurcher Zeitung. Fögen's piece is often obscure in its meaning, partly because it is full of rhetorical questions rather than reasoned arguments, and partly because it is framed in terms of an extended metaphor - Little Red Riding Hood and the Big Bad Postmodernist Wolf - which further confuses the issues at stake. However, her review is by a long way the most hostile and personally offensive to have appeared so far, so it requires an extended reply.

Fögen begins her review with a crassly polemical misrepresentation of its central arguments. In her view the book takes exception to the idea 'that the past emerges from something else than facts and nothing but facts.' Well, no, that's not what the book says. Still less does it follow up this claim by a statement that if the 'cult of facts', to quote the subtitle of Fögen's review, is abandoned, history as hitherto practised will be at an end; a claim - quoted by me from others, including postmodernists, but pinned by Fögen to myself, in order to give the appearance of a paranoid alarmism which I don't share in.

The basic technique of Fögen's review is to rip little phrases out of their context in the book and string them together to make it look as if the book is an assemblage of crude and arrogant prejudice. Here we come to Auschwitz once more. 'Discourse theorists, constructivists of all kinds, all those who hold "real facts" in contempt, are generally dangerous (gemeingefährlich - the word has different resonances in German, being drawn from the vocabulary the police used about revolutionaries in the nineteenth century) and no better than the revisionists' is how she sums up the argument of the book. By this means she concludes that the book seeks to 'misuse the endless suffering in Auschwitz to rescue an endangered cult of facts of the historians. Evans instrumentalizes it for the purpose of defaming his enemies - and a discussion of their premisses becomes unnecessary.' But who is defaming whom here? Nowhere do I say or imply that discourse theorists, constructivists and so on are no better than Holocaust deniers. Nowhere do I say that they are generally dangerous - this is pure invention on the reviewer's part. Nowhere do I say that a discussion of their premisses is unnecessary, indeed I do my best to provide one in the book. Finally on this point, there is in the end a difference between those who argue Auschwitz definitely didn't exist, and those who argue that we can't know whether it did or not, and the book makes this difference perfectly clear.

In the end, therefore, one also has to ask: who is instrumentalizing the suffering in Auschwitz here? Presumably by referring to it in this way, Fögen is actually conceding that the endless suffering in Auschwitz was a fact and that we can know about it. Or is this a fact made, like all others in her view, by historians, television editors and journalists, 'a necessary lie in itself', as she puts it, 'because one can't talk about everything and therefore with every utterance one suppresses, silences, annihilates something else'? Fögen's use of the word 'annihilate' here surely can't be coincidental: vernichten is the term commonly used to describe the Nazi extermination of the Jews. This is almost as distasteful as her description of Auschwitz as a 'necessary lie'.

And Fögen follows it with a grotesque accusation that I am committing an immoral evasion of my personal and political responsibility for deciding which facts will be 'produced', which 'experiences not doubted but defended against evil revisionists', by my alleged failure to recognise that what historians do is to select and present facts. Once more: I don't think the suffering of people in Auschwitz is a fact created by historians. I think it really happened. Of course, historians choose to represent it in one way or another, are led to discover, reconstruct, narrate and interpret it by their own concerns and by those of the society they live in. The book repeatedly asserts the moral responsibility of the historian to people in the past - all people in the past - the reality of their lives and sufferings, and the immorality of distorting or abusing or manipulating the lives of those people in the service of some present-day cause.

As a good deal of evidence from other historians and writers presented in the book demonstrates, the issue of the factuality of Auschwitz and how we discover it has become a central issue in the debates about postmodernist scepticism. This is not instrumentalization. There is a strong moral charge to the issue which is leading people to raise it in the face of extreme scepticism about the possibility of knowing anything about the past at all. As many others have argued, Auschwitz was not a discourse, it was reality, a view to which Fögen would appear to subscribe, as already noted, when she refers to the 'endless suffering' that took place there. It follows therefore (a) that we can find out about it and establish what happened the re, and (b) that the same applies to other historical topics as well.

Instead of taking this point and reflecting on the way she approaches the subject, however, Fögen immediately goes on to try and make this argument seem absurd by linking it to the critique in the book of a recent book on witchcraft which refuses to make any distinction between the portrayal of witches in literature, folksong and so on, and 'the witches themselves'. 'Anyone', Fögen concludes, summarising what she wants the reader to think is the argument of In Defence of History, 'who considers witches as a construct, a story, a text, will also consider Hitler an invention - even the most stupid person can see that!'

This is a connection that the book doesn't make, nor would I want to make it in practice, since we are dealing with two different categories here - a named and identifiable individual on the one hand, and a group of people subjected to a contested cultural and legal categorization on the other. I suppose to cater for the literal-minded, or to prevent the deliberate misinterpretation of the phrase for polemical purposes, I should have written not 'the witches themselves' but some rather more cumbersome phrase such as 'the real people in history who suffered and were often tortured and killed because other people thought they were witches.' But proceeding on such a basis would have quickly rendered the book unreadable. There are many hist ory books on witchcraft which use the term 'witch' without such elaborate qualifiers, and I confess it hadn't occurred to me that someone would actually misrepresent my use of the phrase 'the witches themselves' to imply that I didn't think the categorization of these people as witches was 'a construct, a story, a text' or for that matter even 'an invention'. That doesn't alter the fact, however, that real people were involved, people who suffered real pain and death as a result of the discourse on witchcraft, however fantastic it might have been.

Moreover, in another entirely characteristic and obviously deliberate misrepresentation, Fögen tells her readers that I advise the author in question to study 'the witches themselves' instead of 'the representation of witches in poetry, drama, historical texts'. What the sentence in question actually says is that the important feature of the approach to witchcraft under discussion 'is its refusal to make any distinction between historical, fictional and poetical accounts of witchcraft, and its concentration on the representation of witches in poetry, drama, historical texts and other forms of historical literature, rather than on the witches themselves.' There is no 'advice' to the author anywhere here or anywhere else, for that matter. The approach under discussion is outlined simply as an example of what a postmodernist history might look like. The key point here is the refusal of this approach to distinguish between factual historical accounts on the one hand, and fictional poet ical accounts on the other, its treatment of all these kinds of texts on an equal basis, and its rejection of the idea that one can find out anything meaningful about the women who were persecuted as witches in the past, or discover why they were persecuted. There is nothing wrong with studying fictional accounts of witchcraft in itself, and I would never want to dissuade anyone from undertaking such a task.

If this accusation by Fögen is based on a manipulation of the text of In Defence of History, so too is the allegation that it claims that facts can be drilled into students by forcing them to write analytical rather than narrative essays and that this (as Fögen says) is what I do in Cambridge. Leaving aside the fact that I wasn't in Cambridge when I wrote the book, what it actually says is that history tutors (everywhere - no particular university is mentioned) tell their students to write analysis rather than narrative. Nowhere in the book is this suggested as a recipe for rescuing the primacy of facts; on the contrary, the insistence on analysis dethrones the mere recital of facts and quite properly puts argument and interpretation at the centre of the student's efforts. Finally, this example was given as an illustration of the fact that, contrary to what some postmodernist critics of historians have suggested, narrative is often not the central technique used by historians in ordering their material; and the sentence in question came after a list of major historical works not written in the narrative mode.

Both Anthony Easthope and Marie Theres Fögen object to my notion of 'fact' as something that happened or existed in the past irrespective of whether we know about it or not, and dismiss the argument that we discover facts as a kind of mindless empiricism. Both misrepresent the book, in different ways, as a defence of a crude and ignorant empiricism, or to use Fögen's term, a 'cult of facts'. To reiterate: I use the term 'facts' in this way because I do not want to fall into the trap of supposing, as Hayden White does in a passage quoted in Easthope's review, that what the historian studies are 'events'. Reality of whatever kind does not consist exclusively of events, though Easthope apparently thinks it does. A historical fact may be the nature of a field-system under feudalism, the equipment of an army, the structure of a Roman villa, the death-rate from tuberculosis, or any one of a huge range of things which could not be described as events. Thus when White and Easthope and presumably Fögen use the term 'facts' they are not using it in the way I use it.

Discovering facts in this sense is an important part of the historian's business, but the book argues that the crucial step is when they are used in the service of an argument, and thereby become evidence. Changing his terminology to mine, that is, his 'events' become my 'facts' and his facts' become my 'evidence', I would have nothing to quarrel with in White's statement, endorsed by Easthope, that 'The events have to be taken as given; they are certainly not constructed by the historian'. I suspect that when White goes on, in the passage in question, to say that historians are interested in giving a true account of what really happened in the past, he is really saying that historians are interested in giving a convincing interpretatio n of what happened in the past. Sticking to political history for the moment, for example, White (and Easthope) might regard an account of the origins of the First World War as a historian's attempt to represent what really happened, whereas my view would be that any account of such a complex and tangled web of facts must of necessity embody an interpretation (distributing responsibility, linking cause and effect, and so on).

We already have, therefore, a real gap between Easthope's view of what I am saying, and my own, and more profoundly, a serious gulf between what he thinks history is (the representation of events) and what I think it is (a subject vastly and rightly far more diverse than this). More seriously, however, Easthope's continual attacks on my qualifiers and modifiers seems to demand a hard-edged piece of logic where statements are made of a bald, extreme and universal kind. If I say, for instance, that historians in general do not work within rigid and constricting paradigms, what I mean by that is that of course a few of the most rigidly dogmatic have done so, and many have worked within a loose and flexible set of theories and assumptions s uch as Marxism in all its varieties, but that the Kuhnian notion of a strict paradigm which shapes all research in a particular field (e.g. optics) isn't really applicable to what historians do. Easthope slips the term 'paradigm' in to describe empiricist epistemology, but this isn't what Kuhn meant by a paradigm, as he ought to know: Kuhn didn't deny the reality of, say, a chemical reaction any more than White now denies the 'given' nature of 'events'.

But Easthope, apparently, does. Of course I am aware of the fact that there is a huge mass of philosophy arguing about whether or not we can prove that we experience reality through our senses, whether we can prove the world exists, what causation is, and so on. In the end, perhaps, no-one has proved conclusively and logically that we do exist; but no-one has proved that we don't, either. Here again, Easthope is caught in the classic self-contradictions of postmodernism. If he maintains, as he seems to, that there is a 'radical and irremediable' gap 'between reality and representation', then how can he dare to represent the reality of my text and assume he is being fair or accurate in doing so? The fact that in the particular case of hi s review of my book this gap seems to be a real one doesn't in the end affect the general principle. Easthope complains that I am lampooning postmodernism. He demands that the critic must give a 'fair, accurate and detailed account of the arguments he or she intends to attack'? What is the difference between this and demanding that historians must give a fair, accurate and detailed account of the past?

Easthope finds fault with what he alludes to as my statement that the limits which the language of a text imposes on the possibilities of interpretation are set to a large extent by the original author. There are several points to make here. First, by bringing this sentence on page 106 into immediate proximity to a phrase two pages earlier in the book, Easthope makes it look as if I am arguing that the limits referred to are set solely by the intentions of the text's author. This is not so: I go on to argue that the author may be including unintentional things in his or her text. And I add what he calls the 'dodgy qualifier' ('to a large extent') because it is obvious to me that in the case of diplomatic documents (which are the subject under discussion on the pages in question) misinterpretation and ambiguity, whether intended or not, whether the result of an author's intention or the result of a diplomatic compromise, can open the way to further possibilities of interpretation (though these too are limited by the language of the text). Least of all do I mean to argue, as Easthope surely implies, that the intentions of the author alone limit the possibilities of interpretation. There are many other factors involved as well, and the intentions of the author only constitute one of them, though they are far from being as irrelevant as many postmodernists would have us believe. Moreover, if Easthope demands that an interpretations of a text should be fair, accurate and detailed, then how can he say that such accounts have nothing to do with the language in which the author has written the text?

Finally, when I first read his conclusion ('Though his name is on the cover, Richard J. Evans did not really write In Defence of History - rather, the dominant paradigm of the English empiricist tradition wrote it for him, because he made no critical attempt to interfere with its passage through him on to the page'), I thought for a moment that Easthope was indulging in a rare postmodernist moment of self-parody. But no, he really means it (his total lack of a sense of humour is indicated a couple of sentences later on by his characterization of my little joke about Patrick Joyce meaning his own ideas when he referred to 'the intellectual history of our own times' as an example of the 'blunt, Hobbesian, man-of-the-world aggressive tone which in many circles of history writing seems to pass for machismo').

The obvious answer to his point is to say something like:

Though his name is on the title-line, Anthony Easthope did not really review In Defence of History - rather, the dominant paradigm of the English cultural studies industry wrote it for him, because he made no critical attempt to interfere with its passage through him on to the page. Such an uncritical stance in no way prevents the review from adopting that pompous, po-faced, superior, pseudo-scientific tone which in many circles of critical theory writing seems to pass for machismo.

But that would be unfair, wouldn't it? Easthope tries to defend his tone by claiming that while he is critical, at least he is not arguing ad hominem, and I suppose strictly speaking this is true, since how can he be ad hominem if he doesn't think I wrote the review at all? However, this is more than made up for by his distortion and misrepresentation of the book's contents, his concentration on minor slips and secondary issues in order to avoid confronting the central arguments, and his complete lack of any realization of the self-contradictory nature of his own position.

The same can be said of Marie Theres Fögen's review, except that it is nothing if not ad hominem. In order, for example, to convey the impression of vanity and self-importance in the author, she quotes my description of my earlier book Death in Hamburg in which I say that 'the book's structure was the result of a series of deliberate aesthetic and intellectual choices (US edition, p. 126). In the German edition 'deliberate' becomes 'reiflich durchdachter' - roughly, 'maturely thought-through', and I have to say that when I checked over the translation it didn't occur to me that this could be ripped out of context by a hostile reviewer and used to make it look as if I was praising my own book as the result of a mature process of reflection which arrived at an 'aesthetic' result. But no trick of the reviewer's blackest arts is too low and mean for Fögen to stoop to. She knows as well as anyone that this remark was made to emphasise the fact that historians, as the example of my own work showed, made aesthetic as well as intellectual choices about structuring their material and their arguments, an illustration, in other words, of Hayden White's point that aesthetic impulses do play a role in historical writing.

Indeed, by couching her review in the form of the story of Little Red Riding Hood and the Big Bad Postmodernist Wolf, Fögen herself makes a stab at aestheticizing the issues at hand. The trouble is that she gets the story wrong. Fögen concludes by noting triumphantly that the wolf ate up little Red Riding Hood, swallowing her whole. But as anyone who has read the Brothers Grimm should know, this wasn't the end of the story at all. The woodcutter came along, killed the wolf, ripped open its belly and restored Little Red Riding Hood to the world in perfect health.

Criticism 7. the book misunderstands key arguments of the postmodernists.

While some of the reviewers sympathetic to postmodernism, such as Joyce Appleby, or who are philosophically trained, like Anthony Grayling, accept that In Defence of History gives an accurate representation of the postmodernist theories it is attacking, others do not. The dispute begins with the definition of postmodernism itself. The book makes it clear that potsmodernism is a convenient label covering a wide variety of positions, not a unitary body of theory, and recognizes that some of these positions are mutually contradictory or antagonistic (pp. 254-5, note 5). It also distinguishes between moderate and radical variants, following the distinction made in Rosenau's Post-Modernism and the Social Sciences. Broadly s peaking, it recommends the former for attention but rejects the hyper-relativist theories of the latter. It is quite true to say that the book does not hazard a definition of this protean body of ideas. It does its best to avoid ascribing arguments to 'the postmodernists' in general. All it argues is that it is convenient to attach the label of 'postmodernist' to the 'linguistic turn', the denial of the possibility of objective knowledge about the past, the rejection of socioeconomic models of causation, the insistence that 'grand narratives' in history are no longer possible.

Lynn Hunt complains that in the book, 'postmodernismis neither clearly defined nor seriously engaged'. She justifies this criticism by listing a variety of different positions described as postmodernist in the book and countering them with a single definition of her own: 'For historians', she writes, 'postmodernism generally means the view that the historian cannot penetrate the veil of language to historical reality, that is, the historian can only write rhetoric, not truth.' More widely, she says, it involves an attack on modernism and modernity. Even these two very broad definitions do not, however, encompass the variousness of the phenomenon, and the problem of definition remains.

Hunt complains that the arguments of postmodernism are not seriously engaged in the book. But the problem with Hunt's review is that all she offers in support of this accusation is mere assertion; nowhere does she herself seriously engage with the arguments of the book. Thus she convicts it of 'logical inconsistencies, weakness of philosophical analysis, and failure to grapple with the main tenets of postmodernism', while at the same time failing to identify a single instance of a logical inconsistency or philosophical weakness in the entire book. The single example she provides, namely the book's account of Carr's theory of causation, is based on misrepresentation and is again a tissue of assertions rather than arguments. Thus she desc ribes for example its consideration of postmodernist positions on time as inconclusive, without saying precisely in what way they are inconclusive, and without mentioning the fact that the analysis of these positions ends by arguing that time does, contrary to some postmodernist positions, have a forward direction, but that historical time moves at different levels and at different speeds in different modes or fields of historical change.

In the end, therefore, Hunt's review is itself 'a mish-mash in which bald assertion substitutes for analysis', to quote her own dismissive verdict on the book. At least it manages to avoid open self-contradiction, however. That distinction belongs to the critique by Peter Ghosh in the London Review of Books. Ghosh's comprehensive misreading of In Defence of History culminates in the claim that it construes 'postmodernism' exclusively as the denial of the possibility of truth and objectivity. If he had read the book with any care, he would have discovered that - to repeat - it distinguishes at the very least between extreme postmodernist hyperrelativism, which it argues against, and moderate versions of postmodernism, which it defends.

Peter Ghosh declares that it is illegitimate to call anyone a postmodernist unless they describe themselves as such or work on behalf of 'an intellectual cause known as Post-Modernism'. He asserts that there was 'no conscious tradition of Modernism' in history against which people could react, so there can by implication be no such thing as 'Post-Modernism'. If we stuck to such a view, we would never be able to use concepts about people in the past which they did not use themselves. This would make the historian's job completely impossible.

While attempting to argue on the one hand that there is no such thing as postmodernism, Ghosh himself refers in one paragraph of his review to 'the various strains of historical thought grouped under the Post-Modernist label'. In another paragraph he declares: 'Post-Modernism in the historical sphere is a sphinx without a riddle'. While he criticises me for using the label without defining it, he does exactly the same thing himself. It is extremely difficult to see any kind of consistency or direction in Ghosh's criticisms in the light of these kinds of contradiction.

In a somewhat different vein, Nils Minkmar, writing in the Süddeutsche Zeitung, suggests that the concept of postmodern hardly applies to such historians, for 'the theoretical discussion of postmodernity as it has been carried on in literary scholarship and philosophy has hardly inspired concrete works of historical scholarship'. This latter remark has some truth in it, for the only logical consequence of accepting the extreme relativist position taken up by some theorists is that we should stop writing history altogether, since there is no way of getting at historical truth. Minkmar can only say this, however, because he ignores the distinction made in the book between extreme and moderate versions or aspects of postmodernism.

Clearly, he and other critics are right to say that some of the historical works I describe as postmodern would probably not be accepted as such by their authors. And yet, something important has happened to history in the last twenty years or so. The great overarching narratives such as Marxism and modernization theory have collapsed. The idea of history as progress has been abandoned. Innovation has come above all from historians writing about the marginal, the bizarre, the individual, the small-scale. It seems reasonable to call these now-defunct metanarratives 'Modernist', as indeed many postmodernist writers on history do; and equally reasonable to call the new development 'postmodern', even if those who have pioneered it and participated in it would not regard themselves as postmodernists in any sense at all.

To put it another way, I know that Orlando Figes would never describe himself as a postmodernist because he has told me so himself. But I also know that his book A People's Tragedy, an almost self-consciously literary narrative of the Russian Revolution and Civil War, weaving in the stories of individuals, some of them very obscure, to the larger picture, and eschewing the kind of socioeconomic and statistical analysis commonplace in histories of the Russian Revolution twenty or thirty years ago, could not have been written without the theoretical and methodological impact of postmodernism and the decline and fall of the grand metanarratives, even if this has only had an indirect effect on the author rather than a direct or personal impact.

The question of the definition of postmodernism certainly isn't one which my book has solved, or indeed set out to try and solve. Perhaps in the end the charge that my book fails to deliver such a definition is of rather secondary importance. A far more serious criticism is provided, by contrast, in Anthony Easthope's review of the book in Textual Practice., which charges that the book misrepresents major thinkers such as Saussure. Unlike, say, Hunt, Easthope actually provides examples of this alleged misrepresentation. Thus for instance where the book says that Saussure's term 'the signified' meant the thing denoted by a word, what he actually meant by 'the signified' was the concept or meaning of a thing. Easthope says roundly: 'This is a howler', and I'm afraid he's quite right. I will try to see that it is changed in future editions. Easthope scores another hit when he complains that no reason is given for the book's claim on pages 159-60 that Derrida 'rejected the search for origins and causes as futile'. He is quite right here too; I should at least have given a reference, or supported this assertion with a brief analysis.

Other points made by Easthope, however, begin to slip from the identification of genuine errors or omisisons to the partial or in the end complete misrepresentation of what the book actually says. This begins with its compleint that Lyotard is 'dismissed in a single sentence, and a bizarre one at that, to the effect that "master-narratives are the hegemonic stories told by those in power".' There are in fact nine sentences about Lyotard on the page in question (p. 150), not one, and they refer to Lyotard's view that Soviet historiography was a master-narrative which had to be countered by local narratives (or in other words can't be countered by rival metanarratives). This is the context within which the sentence which Easthope thinks so bizarre has to be seen. His misrepresentation is compounded by his omission of the qualifying statement that Lyotard 'tended to' say that 'master-narratives are the hegemonic stories told by those in power' (my italics here). Now perhaps the disc ussion at this point should have made it clear that Lyotard's definition of metanarrative goes way beyond this, including 'the emancipation of the rational or working subject, or the creation of wealth', and so on. Nevertheless, it does not say, as Easthope claims it does, that all Lyotard said was that master-narratives were hegemonic. As becomes clear later in his review, Easthope is unwilling to accept qualifiers and has a strong need to regard all statements as absolute, so his claim here is all of a piece with the other claims he makes in his review about the book.

He goes on to assert that the book claims that Derrida wrote that nothing existed outside language, whereas what he in fact said was that there was nothing extra-textual. However, the previous sentence, not quoted by Easthope, makes it clear that Derrida argues that 'everything was "discourse" or "text".' So the book does after all say that Derrida argued there was nothing extra-textual. It seems a reasonable conclusion to draw from this that Derrida argues that nothing existed outside language. Similarly, Easthope alleges that the book misunderstands what Derrida meant by 'logocentrism'. But in fact it does not attempt a complete account of Derrida's concept, only to one implication or aspect of it, namely his criticism of historians a s 'logocentric, that is, they imagined they were rational beings engaged in a process of discovery'. Easthope does not say whether this was part of what Derrida meant by logocentrism or not.

In all these critical points, I am perhaps paying the price of merely mentioning these thinkers tangentially without actually engaging in a full-scale confrontation with their work, something that other critics such as Sokal and Tallis have done far more effectively and knowledgeably than I could ever hope to do. So I concede that Easthope is at least partially justified in advancing these criticisms.

However, we have to go on to ask precisely how all this actually affects the central arguments of my book. Here the answer is clear: it doesn't affect them at all. Like a number of other critics, Easthope is in the end demanding a different book to the one that I wrote, namely a critical account of major postmodernist thinkers, dealing with all the central points of their philosophy. But - to repeat a point already made in an earlier section of this response - that is not what the book is intended for; it addresses the use, however much it may be based on oversimplification or misunderstanding, of some of the arguments of Saussure, Derrida, Foucault and others by writers on the theory of history: it is interested only in the major theor ists insofar as it seeks to explain as briefly as possible where the arguments of postmodernist theorists of history are coming from. It could just as easily have been written without any reference to these theorists at all. Perhaps it should have been.

What Easthope does in his review is to employ a familiar tactic of hostile reviewers: he is seeking to discredit the book by distracting the reader's attention from its central arguments by pointing to inaccuracies of detail on peripheral points (its brief and marginal allusions to Saussure, Derrida and Lyotard). Rather than engage directly with the book's arguments, he consistently evades and misrepresents what it has to say. This quickly lands him in a hopeless quagmire of self-contradiction.

Easthope describes In Defence of History as a 'lampoon' of postmodernist positions, and condemns it for ignoring the basic precept that 'The first obligation of a critic is to give a fair, accurate and detailed account of the arguments he or she intends to attack.' If we read Easthope's review in the light of his own precept, however, what do we find? Take his very first couple of sentences:

In Defence of History aims to defend a mainstream notion of history writing against 'intellectual barbarians' (p. 8), namely 'the invading hordes of semioticians, post-structuralists, New Historicists, Foucauldians, Lacanians and the rest' (p.9). That statement is typical of the tone of the book..

Easthope achieves this impression of blinkered and aggressive prejudice by the simple means of removing the first three and the last three words of the sentence in question. The full version reads as follows, with the following sentence added:

Historians should approach the invading hordes of semioticians, post-structuralists, New Historicists, Foucauldians, Lacanians and the rest with more discrimination. Some of them might prove more friendly, or more useful, than they seem at first sight.

A postmodernist might find nothing wrong in doctoring a sentence in a text to make it support the argument; I do. It also stands in crass contradiction to everything Easthope says about the way one should treat texts.

Moreover, the term 'intellectual barbarians' is clearly meant ironically and was certainly intended to be taken in this way. It is part of an extended military metaphor actually started by Patrick Joyce in a reference to the 'commanding heights' of the academy's walls having 'fallen' to 'skirmishing bands' of postmodernists in the United States. I must confess that when the book was going through the copy-editing stage, my editor at Granta wanted me to remove this metaphor, since he said some people might take offence at the reference to postmodernists as intellectual barbarians. I told him I didn't think anyone would be stupid enough to think I genuinely believed that postmodernists really were intellectual barbarians, least of all pos tmodernists themselves, who were used to reading texts in an ironic mode. He accepted my point and we left it in. How wrong I was.

A somewhat milder foretaste of this kind of response was provided by some of the letters which appeared on 19 and 26 September 1997 criticising the preview of the book printed in the Times Higher Education Supplement. John Arnold, for example, claimed that it was wrong of the book to argue that postmodernism holds that 'there is a multiplicity of equally valid truths'. Rather, he went on, 'it is that truth (as a claim to one dominant and essential position) is untenable...all truth (with a small "t") is situational, political and engaged. To claim that something is true is to attempt to place it beyond discussion.' This argument seems to me to be a tissue of contradictions. Is Arnold saying there is no such thing as truth at all? Or is he saying that there is truth with a big 't' and truth with a small 't'? If the latter, then he is surely saying that there is indeed a multiplicity of equally valid truths. If the former, then how can he believe that what he is saying himself is true?

Secondly, Arnold charges that 'Evans misunderstands the notions of "text" and "discourse"'. He seems to think that I argue that discourse is a false image of reality, whereas in fact in his view it is reality 'or at the very least, as close to reality as we are ever going to get'. Already here he has introduced a confusion in his argument. I suspect that what he is trying to say is that we can never know reality at all; everything is discourse because everything is mediated through language. I thought that this was the argument I was summarising in my brief discussion of the view that texts are not a window through which we can perceive the reality of the past, but that they in some sense are the reality of the past, because the past as such no longer exists. I do not think I gave the impression that dicourse is merely an image of reality; if Arnold took this impression away from what I wrote, then he was under a misapprehension of what I meant to say.

Among a variety of arguments the book deploys against this view, is the admittedly rather bald statement that Auschwitz is not a discourse, and it trivialises mass murder to see it as a text. To counter this, Arnold points out that 'the gas chambers were a physical expression and outcome of a particular discourse that presented a 'reality" wherein Jews, homosexuals, communists and others were sub-human.' This brings yet more confusions into Arnold's argument. For here he is reintroducing a distinction between Auschwitz and discourse in which the former is the product of the latter; an entirely unexceptionable point made by almost every historian who has ever written about Nazi antisemitism in theory and practice, even if they have used words like 'ideology' rather than 'discourse'. But it is a point that totally undermines Arnold's earlier argument that discourse is reality in the sense that Auschwitz itself is discourse because we only know about it through language; a position that is far less easy to defend, in my view. We may agree that Jews were beaten, starved and sent to the gas chambers because of the Nazi discourse of antisemitism, but it is quite a different matter when someone tries to tell us that being beaten, starved to death or gassed is also a 'discourse'.

A related argument to that put so confusingly by Arnold is made by David Andress, who suggests that In Defence of History fails to understand a significant point about the relationship between truth and power. Andress asserts that truth is a moral category rather than something defined by its relation to reality. Thus, he says, E. P Thompson's influence over later historians was greater than that of G. R. Elton not because Thompson had a better control of fact but because he had a more morally appealing vision of the British past than Elton did. He triumphed, as it were, because his truth was morally superior to Elton's. Andress makes this point in response to the argument of In Defence of History that the influence of a particular historical work or view of history was not a function of the institutional power of its author, contrary to claims made by writers such as Keith Jenkins. Thus Thompson's The Making of the English Working Class had a huge influence despite the fact that its author was never a member of a university history department, while Elton's Tudor Revolution in Government was widely disputed and never fully accepted by historians despite the highly influential positions he occupied in academia. Andress argues that it is moral rather than institutional power which makes historical arguments influential and causes them to attain the status of 'truth', so that truth is in the end indeed a function of power, contrary to what In Defence of History tries to claim.

Andress's point is an interesting and perceptive one, and prompts me to think I should have put my argument a little differently. For if one thinks about it, the central historical thesis in Thompson's book - that the English working class was 'made' by 1832 - was even less widely accepted than the central historical thesis in Elton's - that there was a revolution in government and administration under Henry VIII's minister Thomas Cromwell. An examination of subsequent research monographs, articles and textbooks will soon show that very few historians of any kind incorporated Thompson's central thesis into their work, while some at least gave credence to Elton's. The difference in influence lay far more in the method and approach to history of the two historians, where Thompson was immeasurably more effective. Thompson's book had many imitators, Elton's had few. Elton did not start a wave of studies of administration and government in early modern history. Thompson inspired a whole generation of social historians to study a whole range of topics in working-class history. When the question is seen from this point of view, Andress is right: Thompson was influential because his vision was more in tune with the spirit o f the younger generation of social historians in the 1960s and 1970s.

Yet this has nothing much to do with the truth or otherwise of the interpretations each of the two historians advanced about the historical subjects and events they were investigating. Andress is wrong therefore to say that Thompson's influence was greater because he told a story that was morally appealing enough to be regarded as true. Thompson was influential because even if historians thought he was not giving a true picture of English plebeian and working-class society and politics in the late eighteenth and early nineteenth centuries, they none the less thought that the approach he adopted would enable them to give a true picture of the (usually much smaller) historical topics to which they chose to devote their own researches.

The argument which Andress puts forward thus in no way supports the hyperrelativist version of the truth-concept which he is advocating ('history...is an exercise in language because that is all it can be'). Andress takes the alarmingly anti-intellectual position that 'the marshalling of facts behind an argument is rendered irrelevant by the reply "I don't believe you, or "I don't care"'. The kind of antisemitism that led to Auschwitz, he says, is dead not because it has been shown to be untrue, but because 'we live in a truth-regime where the values behind those "facts" have been rejected.' If he actually believes that rational argument is so irrelevant, what is he doing teaching in a university? If somebody tells him he or she doesn't accept an argument, why doesn't he challenge him to disprove it rationally? Does he really want to live in a society where the evidence for an argument counts for nothing and the moral (or immoral) force behind its advocacy for everything? Surely the duty of an intellectual or an academic is to fight blind prejudice of the kind Andress so depressingly thinks unanswerable, not to give in to it. Contrary to what he argues, there is plenty of evidence to show that the truth prevails in the end because it is true; that antisemitism is no longer widely accepted because among other things it has been seen to be based on lies. Having reformulated the question of the relationship between power and kowledge in a stimulating way, Andress then delivers an answer that reduces the whole problem to the most basic level of 'might is right'.

While Arnold and Andress attack particular aspects of my arguments as allegedly based on misunderstandings of postmodernist concepts, William Keenan, also writing in the Times Higher Educational Supplement, finds my depiction of postmodernism to be a 'caricature'. In a rather polemical letter, Keenan criticises me for my 'apparent unawareness' of the fact that everything I am saying is an assertion or an opinion and not an objective truth. How can we arbitrate between arguments in such a situation? he asks, admitting that his own views are only opinions too. To this he gives no answer, taking refuge in the paranoid assertion that any criticism of postmodernist relativism is evidence of negative stereotyping and a determination on the part of people like me 'to cleanse academic life of views contrary to their own'.

Unfortunately Keenan fails to provide any evidence demonstrating that my account of postmodernist theories is the kind of caricature he says it is. And while he urges 'open discussion' on the topic, he is unable to say how this can take place at all if there is no objective, over-arching principle of evidence or rationality to which we can appeal in order to assess the opposing points of view. In fact the language Keenan uses suggests he does believe it is possible to provide a full, objective and fair description of an argument (the opposite, presumably of caricature), on the basis of which a reasoned debate can actually take place. Otherwise, why is he complaining of an attempt to suppress postmodernism, if he believes it's all about power, and polemicises so violently himself in order to suppress the modernist views to which he is so vehemently opposed? This position deprives his complaint of any moral force whatsoever.

Criticism 8: the book is unfair to those it criticises

As well as misunderstanding or misrepresenting postmodernist theory, some critics of In Defence of History have attacked the book for what they see as its unfairness to other individuals whose work it criticises. For example, a number of reviewers, such as John Charmley and Ronald Hutton, not necessarily defenders of postmodernism, have suggested that Keith Jenkins's hostility to conventional history as practised in the universities is ascribed by In Defence of History to the fact that he has never had a job in a university himself. But this is not so. What the book actually says, after summarising Jenkins's argument that professional historians believe in an 'objective' and on the whole liberal-conservative interpretation of the past because they teach in 'multi-million pound' universities engaged in the 'ideological control' of society, is as follows:

That Keith Jenkins sees university history departments in this jaundiced way may, if one wishes for a moment to borrow his own mode of argument, have something to do with the fact that he is only a lecturer in an institute of higher education, as so feels excluded from the multi-million pound institutions he is criticising so aggressively. Doubtless this would be an unfair charge, but no more unfair than the charges he is levelling at the university historical profession as a whole.

In other words, the book was not saying that Jenkins takes the views he does because he is not teaching in a university, only that if he applied his own arguments to himself, that is the kind of argument he would end up with.

Indeed this is actually put forward as an example of how unfair, how reductionist, postmodernist arguments can be. As Hutton observes, one of the main techniques of In Defence of History is to apply the postmodernists' theories to their own writings, and no more should be read into this passage than precisely that. Elsewhere, indeed, the book points out that a number of postmodernists have major posts at senior universities, so the argument as applied to Jenkins (ultimately by Jenkins) falls down on this count too, since I explicitly argue that, contrary to what some postmodernists appear to believe, we are not dealing here with conservative empiricist orthodoxies entrenched in major citadels of academe and confronting progressive thinkers who have been deliberately excluded from the centres of academic power.

This argument, like some of the others dealt with in the previous section of this 'Reply to Critics', comes down again to the relationship between knowledge and power. I confess that when I wrote the book, I did not realise quite how dangerous the cruder attempts to link the two could be. If you think knowledge attains the status of truth solely through the exercise of power, then power is what you will try to get, and there might indeed be some justification in the Marxist writer Chris Harman's claim in his review of In Defence of History that 'popular postmodernist arguments have been used to terrorise students in innumerable university departments into submission to their lecturers' dogmas'. If this were true, it would be a be trayal of all the values of rationality and the free exchange of ideas which the concept of a university is supposed to be about. In this sense, it is not just history that needs defending, but the spirit of rational argument and the unfettered discussion of differing views, indeed, the whole spirit in which academic and intellectual life is lived.

That this spirit has been violated by my book is the complaint of Diane Purkiss, who writes to defend her own book The Witch in History against the brief remarks made about it in In Defence of History. Purkiss accuses me of misquoting from her book and claims I have discussed it without having read it, or alternatively that I have read it but have not read it accurately (that these two accusations are mutually contradictory does not seem to occur to her - or is this mere rhetorical hyperbole not intended to be taken seriously?). A merely inaccurate reading as distinct from no reading at all would seem to be a relatively venial sin, for Purkiss takes my allegedly inaccurate reading as evidence for her view that 'reading all the words that are there, just those words and no others, not adding, not taking away, just reading' is a 'nearly impossible task' which 'none of us domost of the time.' Purkiss's 'Response' would certainly provide startling evidence of this point, supposing one was inclined to agree with it.

Let us start with a rather trivial example. On page 4 of In Defence of History, she charges, I attribute to her opinions which have been put forward not by her but by unnamed 'conservative defenders of history'. When we look up page 69 of The Witch in History, however, which is where the passage in question occurs, what do we find? At the end of a paragraph attacking male historians for their lack of interest in feminist and gender theory, Purkiss concludes that they evade the truth-claims of such theory by dismissing it as of no use to them in their work. There follows the sentence: 'The fact that poststructuralism threatens to throw historians out of work does not make it untrue.' This opinion is not attributed to anyone else and is neither set in quotation marks nor supplied with a footnote referring the reader to someone other than Purkiss herself as its author.

It is, to be sure, followed by another sentence, which reads as follows: 'Though some historinas (sic) do offer other refutations, these are usually based on fairly comprehensive misreadings of what they refute, again suggesting a certain interpretive carelessness.' The footnote to this sentence, like the previous footnote in Purkiss's text, refers to Joyce Appleby, Lynn Hunt and Margaret Jacob's Telling the Truth About History (a book by three historians, or perhaps we should now call them 'historinas', who can hardly be accused of a conservative approach to the past, or of a lack of interest in feminist and gender theory, or indeed of being male). But the view that poststructuralism threatens to throw historians out of work is not put forward in their book either, nor as far as I know in any other book except The Witch in History, apart from Keith Windschuttle's exaggeratedly alarmist The Killing of History, which is not cited anywhere in The Witch in History and which, presumably therefore, Purkiss had not read at the time when she wrote her book. Purkiss does not say in her book that conservative defenders of history often sound as if they believe that poststructuralism will make them unemployed. The sentence in question is unambiguously her own. Indeed, Purkiss's claim that it would be 'hilarious' if she had actually said that poststructuralism threatened to do historians out of a job does not prevent her from repeating this view in her 'Response', where she remarks: 'Any resemblance between my work and claims made about it by Evans is purely coincidental, but of course one simply must have grist for the conservative argufying machine, otherwise, I suppose, redundancy would truly beckon.'

This leads on to a broader assumption in Purkiss's 'Response', reflecting a more general prejudice in her book, which is that In Defence of History is part of a 'conservative argufying machine', and, by implication, that I am part of a wider group of 'conservative defenders of history'. Since Purkiss never actually explains what she means by 'conservative', this is a charge that it is difficult to refute. However, if she rereads the book, she will find that it defends, as noted earlier in this 'Reply', a broad approach to history, including social history, cultural history, feminist history, and all the other recent innovations condemned by the true conservative defenders of history like Gertrude Himmelfarb, Geoffrey Elton, John Kenyon, and indeed all the historians to whom the young-fogey critics of my book think I have been so unfair. The book argues that socialist, feminist, black and other radical histories need a concept of truth and objectivity just as much as any other his tories do, but it says nothing against them in principle, quite the contrary. It also points out that once you open the relativist door, not only conservatives but also fascists and neo-Nazis can come in through it too. The U.S. edition of the book provides fresh evidence of the conservative tendencies of at least some postmodernist writers on history (in particular, David Harlan in his book The Degradation of American History). It is a travesty to assume that postmodernism or poststructuralism are radical, and anyone who criticises them is conservative.

Purkiss also assumes, to read out the implications of her rhetoric, that I regard feminist and postmodernist historians as a 'rabble'. Where have I ever said such a thing? This is pure invention. Perhaps Purkiss would care to read the books, articles and reviews I wrote on feminist history in the 1970s and early 1980s, or to note the close and repeated attention paid to gender as a social construct in my recent books Rituals of Retribution and Tales from the German Underworld. Or, indeed, the criticisms advanced in In Defence of History of a historian like John Vincent who has declared recently that history is about the rich and powerful, 'history is deeply male'.

What I object to is not feminist history - the more of it there is, the better - but crude and blatant sexism of the sort Purkiss displays in her book and her 'Response'. In the latter, she is insistent on referring to me as a 'man', goes out of her way to describe historians of witchcraft as 'gentlemen', says that I think historians should argue in a 'manly fashion', accuses 'academic historians' of 'duffing up a woman historian (i.e. Margaret Murray) for being somehow too close to the figure of the witch', and argues that 'male historians tend to ground rationality in a repudiation of the feminine as problematically inchoate'. She goes on to allege that Keith Thomas and Alan Macfarlane unconsciously try 'to use the figure of the witch to define history in masculine terms'.

To realise how sexist such remarks are, one only has to reverse the gender designations. If Purkiss thinks it is legitimate to write about male historians in such terms, then what objection could she possibly have to a male historian who insisted on pointing out that someone he disagreed with was a woman and as such part of a school of 'ladies', who accused her of wanting to argue in a 'womanly' fashion, whatever that might be, who objected to her 'duffing up a man historian for being somehow too distant from the figure of the witch', and who raised similar objections about female historians trying to use the figure of the witch to define history in feminine terms? She would rightly find such views offensive.

Or, to refer again to Purkiss's book, if it is wrong for male historians to 'ground' male identity in a historical myth of women's powerlessness in the past, then why is it not also wrong for female historians to 'ground' female identity in a historical myth of women's articulacy in the past? The point once more is that such arguments cut both ways. If an antifeminist wants to attack Purkiss's work with sexist political arguments of this sort, if the persuasiveness of historical scholarship depends not on the extent to which its arguments fit the evidence but on the political and moral assumptions behind it, then she has no means of defending herself against her critics. If for example Purkiss objects to the fact that Keith Thomas has used the figure of the witch as the ground for his own academic identity, then what reason has she for objecting to the view that she herself has used the figure of the witch as the ground for her own academic identity?

Purkiss alludes to my description of her as a postmodernist as 'hysterical'. How would she feel if I described her arguments as 'hysterical', or imagined her, as she does Thomas and Macfarlane, writing 'tearfully' and 'with a maniacal cackle'? Purkiss complains about the 'discourtesy' and 'disrespect' with which she alleges I have approached her work. These epithets would seem to be more properly applied to the way she deals with people (men) with whom she disagrees in her own book and in her 'Response'. Leaving this aside: once again, I am not using the label 'postmodernist' as a term of condemnation, simply as a useful description, and one to some of whose various manifestations I attribute positive and fruitful intellectual developme nts; I would be equally happy to call Purkiss a poststructuralist, a designation she appears to accept, if that made things easier. My footnote on the term 'postmodernist' tries to justify its usefulness while conceding its problematic nature.

In terms of Purkiss's own work this means that I place it in the context of an approach which makes no distinction between fact and fiction, or to put it another way, different forms of representation. Here we come once more to 'the witches themselves', a phrase which has obviously provided something of a hostage to fortune, and here again I have to explain that what I meant was the real people in the past - men as well as women - who were accused of witchcraft and who appear in trial and other historical records. I continue to think that there is an important distinction to be made between these people, who really lived, and the figures who appear in literature, poetry and other fictional representations, who did not. I do not say, however, and never have said, that it is wrong to study the latter or even to study both real people and fictional people side by side (indeed I do this in my own book Tales from the German Underworld).

This brings us to Keith Thomas and Religion and the Decline of Magic. It may be true, as Purkiss remarks on page 44 of her book, that 'Keith Thomas and Alan Macfarlane, the dominant voices in English witchcraft studies for over a decade, are rarely read by modern witches, and none of those I met had even heard of them.' However, Purkiss has more serious criticisms to make than this. She denies that she simply dislikes their work because they are men, or because they are creating myths of male dominance rather than serious historical scholarship. She says they base their work on a 'misreading of historical materials' and that they 'get it wrong'. Thomas uses only printed sources and 'neglects to read these sources scrupulously or carefully' and 'often uses literary sources without distinguishing them carefully from trial materials'.

To some extent we are arguing at cross-purposes here. I do not accept Thomas's interpretation of early modern English witchcraft in my book, nor do I say anywhere that Religion and the Decline of Magic is the 'last word' on the topic (in fact there is now a superb example of an explicitly postmodernist approach to the subject in Stuart Clark's Thinking with Demons, a book which does not manage to avoid the inherent contradictions of poststructuralist theory, but nevertheless provides a stunning illustration of how they can be used to breathe new life into an old subject). Of course I know that Thomas's book was based on pamphlet and printed literature, and of course I haven't read this literature myself - I'm not an expert on seventeenth-century English witchcraft. To demand that a critic or commentator on a book has to examine all the evidence on which it rests before being capable of saying anything about it would disqualify anyone from saying anything about any historical work apart from their own.

One might indeed legitimately turn the tables and ask how much of the evidence used by Thomas has also been studied by Purkiss. This is not easy to discover since she has not troubled to provide her book with a bibliography. However, a check through Chapter 3 ('The witch in the hands of the historians - a tale of prejudice and fear') reveals references to primary printed sources in seven notes out of eighty, with a total of seven works listed. There are a handful, and no more than a tiny handful, of manuscript sources cited in the footnotes to other Chapters, but Purkiss's book itself is based overwhelmingly on printed material. It does not cite more than a tiny fraction of the source material cited by Thomas. Purkiss says in her book t hat she has read more than she cites. Be that as it may, her book's criticism of Thomas in Chapter 3 is not based on a comprehensive comparison of what he says with the materials on which it draws, so that her attack on me for not doing this in my discussion of Religion and the Decline of Magic must appear disingenuous, to say the least.

Of course Thomas and Macfarlane's work has been rightly criticized on a number of grounds, some of them mentioned by Purkiss. But this doesn't justify the crude and sweeping dismissal of it as 'wrong'. I have reread the relevant parts of her book, moreover, and I still can't find any evidence presented by her that convinces me that they were out to empower men in the present and build their own careers in the process by repudiating the feminine, using the silent figure of the witch to disempower women, and so on. Women do appear in their work as prepared to use weapons like cursing in order to get their own way; they are not merely silent and submissive objects of elite persecution. Moreover, there is a whole genre of feminist historical writing which portrays women as victims in order to highlight the oppression and injustice under which they have had to suffer through the ages. In any case, the motives for which a historian writes, while they may be relevant to understanding the interpretations advanced, are altogether irrelevant to the validity or otherwise of the interpretations in question, a point that Purkiss does not seem to understand. Her own account of the historiography of witchcraft has depressingly few detailed or convincingly supported criticisms of the interpretations this historiography has advanced.

The real quarrel Purkiss has with my discussion of her work, however, does not relate to the historiography of witchcraft, but, once more, to postmodernism and 'Holocaust denial'. Purkiss points out quite rightly that Holocaust deniers claim they are telling the truth, and do not on the whole use the conceptual apparatus of poststructuralist relativism. The key passage at issue in her book is on page 70:

Historians are fond of pointing out that deconstruction's leading American exponent was a Nazi sympathiser and anti-Semite during the war, indicating the dire political consequences of deconstruction's challenge to conventional history. But in that case, traditional empirical history must also have dire implications, since most neo-Nazi historians adopt the most conservative possible protocols of discovery, revelation and truth-telling. It might be equally valid to argue that notions of the indeterminacy of truth can be used against the self-proclaimed certitude of deniers. Deniers' assumption that the "truth" about the Holocaust lies around waiting to be "discovered" and encapsulated in a historical text forms the greatest possible contrast with the discourse of Holocaust survivors' testimonies, which affirm the impossibility of representing the ultimate "truth" of the event.

In her 'Response', Purkiss quotes selectively from this passage in her own book to make it look, once more, as if what she puts forward there as her own views are in fact her summary of other people's. She achieves this by omitting the second sentence in the passage above. Purkiss just didn't quote what she wrote with even the meanest level of care. Or rather, to put it less charitably, she took very good care to ensure that the incriminating sentence in her text mysteriously vanished when she quoted it in her 'Response'.

If readers want to look up the passage in my book which discusses this argument, they will find it, not on pages 140-1, where Purkiss directs them (they contain a discussion of postmodernist concepts of time), but on pages 241-2. She does, however, quote the sentence in question correctly, and I accept that I should have put 'dire implications' instead of 'dire consequences'. The meaning and effect are the same, however. The problem with this passage, as I pointed out in the book, is that it assumes that - at the very least - one cannot distinguish between the methods of real historians and the methods of Holocaust deniers because both present themselves as seekers after historical truth and both use the conventional apparatus of histor ical scholarship to achieve this effect. Either they both really are seeking after the truth, or they are both putting forward political arguments dressed up as such, but in both cases the effect is the same.

What this argument ignores is the fact that Holocaust deniers are not engaged in a search for truth, but manipulate and falsify the historical record for their own political purposes. Whether or not they really believe they are searching for truth is a question which cannot be answered without recourse to speculation about their own psychology, but the way they treat historical evidence strongly suggests that their falsifications are conscious and deliberate. What I argue in the book is that it is possible to draw a line between this fundamentally dishonest abuse of the historical record and the honest attempts of real historians to approach some kind of truth about the past. The book cites examples of people who have quarried the remains left by the past in order to support a present-day thesis, and contrasts this with the true historian's willingness to jettison dearly-held theories and interpretations if they are not supported by the evidence. But the falsifications of Holocaust deniers go a lot further than this, and anyone who has read their work (as I suspect Purkiss has not) will not find it difficult to tell the difference. It isn't that they are bad historians, they are not historians at all.

Purkiss goes on to say on pages 70-71 of her book that

careful reading of texts and especially stories is germane to Holocaust studies. Reading in detail and with due attention to textuality need not imply a fixed or cynical scepticism; such care in reading can signify proper respect for the unique and moving humanity of stories.

Why not proper respect for their truth? Why does it matter that stories are uniquely and movingly humane, but whether or not they are true is a matter of such subordinate importance that it doesn't even warrant a mention? Why should care in reading merely 'signify' respect instead of helping us determine the truth or otherwise of the representations the text contains? Coming after a plea for the validity of 'notions of the indeterminacy of truth' in writing about the Holocaust, these two sentences do indeed warrant the statement in my book (described as 'not even a reasonable inference' in Purkiss's 'Response') that Purkiss is saying that we should tell stories about the Holocaust and not worry too much about whether they are true or not.

What in any case is the truth in such matters, asks Purkiss? 'We can never know the whole or absolute truth' about the Holocaust, she says, 'because it would be more than any of us could bear'. If we cannot know it, how do we know we could not bear it? In pursuing this argument, Purkiss not only gets herself tangled up in contradictions, but also attributes to me a whole series of propositions that I have never advanced. How my book, for example, 'conflates the Holocaust with the witch-trials, ignoring the historical specificity of each', as Purkiss charges, is beyond me; I cannot see any evidence for this absurd charge anywhere in my book.

More importantly, however, to come to the point that really seems to annoy Purkiss, I have never equated her or any postmodernist or poststructuralist either implicitly or explicitly with Holocaust deniers. What I said was - to repeat a point made by Deborah Lipstadt in her book Denying the Holocaust - that postmodernist hyperrelativism encourages tolerance of Holocaust denial because it rejects the idea that one can tell truth from fiction and interpretation from falsification. Purkiss's own arguments are themselves evidence of this. But of course they are not responsible for Holocaust denial themselves, just as postmodernism is not responsible for Holocaust denial. Nowhere have I said this; such a claim would be absurd. I am not 'upset' because Purkiss is threatening to merge me and my 'party' (I wasn't aware that I had one - perhaps Purkiss would care to supply me with a few names?) with the deniers. Rather, I am arguing that extreme historical relativism, the idea that the truth about history can never be discovered, makes it impossible to refute Holocaust deniers. Notions of the indeterminacy of historical truth encourage Holocaust deniers to gain a hearing because their point of view is thought to be as valid in principle as anyone else's (Lipstadt cites a number of disturbing instances of this in her book). This does not mean that they would have any truck with such notions themselves, something I have never claimed anywhere. The notion of the indeterminacy of truth is no asset to Holocaust deniers in framing their pseudo-arguments, nor have I claimed that it could be. But it is an asset to them in getting their claims heard.

This does not mean that they are in league with 'sinister' postmodernists. In the first place, I have never described postmodernists as 'sinister'; this is pure invention on Purkiss's part. Secondly, in summarising Lipstadt's view that 'the increase in scope and intensity of the Holocaust deniers' activities since the mid-1970s has among other things reflected the postmodernist intellectual climate, above all in the USA', I did not mean to imply that Holocaust deniers were themselves affected by this climate, though I can see that it would be easy to misconstrue my statement in this way, and perhaps I should have phrased it more carefully. As the rest of the passage on pages 240-41 of my book makes clear, I agree with Lipstadt's view that the climate of total relativism in American intellectual life has encouraged Holocaust deniers to expand their activities on American campuses by lowering resistance to their falsifications and manipulations of the truth. But of course neither she nor I would claim for a moment that postmodernism was responsible for the rise of Holocaust denial or that postmodernists endorsed the Holocaust deniers' views.

I continue to think, however, that the 'notion of the indeterminacy of truth' espoused by Purkiss robs us of our ability to say to the Holocaust deniers that what they are claiming is not true but false. In this sense I do indeed think that her work demonstrates, as I say in my book (p. 241) that the threat posed by Holocaust denial to all who believe in the power of reason is a real one, and it seems to me that the utterly inadequate way in which Purkiss responds to the threat is an example of how postmodernist views undermine the power of rational argument to refute it. I did not say, however, and do not think, that Purkiss herself is, as she claims I said, 'part of the threat of Holocaust denial', which is a different thing altogether.

In this context, it is deeply troubling that Purkiss should choose to end her 'Response' by criticising the reliability of the testimony of Holocaust survivors, something which of course is grist to the Holocaust deniers' mill, though she may not realise it herself. It is not true, to begin with, that, as she maintains, 'the only testimony, with a few rare exceptions, is from survivors'. It depends of course on what the testimony refers to, but even as regards the actual operation of gas chambers, mass shootings and other forms of the deliberate mass murder of millions of Jews and others by the Nazis, there is a substantial quantity of perpetrator testimony as well, though of course Holocaust deniers have attempted to discredit this, as they have attempted to discredit the testimony of surviving victims. Secondly, whatever Primo Levi said, the survivors are witnesses to the mass murders carried out by the Nazis, and are regarded as such in courts of law. Of course, as Purkiss says, 'the very process of shaping memory into story alters and reshapes memory'. But this does not mean, as Holocaust deniers maintain, that all survivor testimony is invention after the fact. The whole discipline of history from the very beginning has been geared to trying to sift out truth from falsehood in sources, and although memoirs and memories are rather more problematical that contemporaneous sources, they cannot simply be dismissed in this fashion; they have to be sifted for the truth just like any other source, and are just as capable of leading us towards it.

This brings us, finally, to the question of truth itself. Of course Purkiss is right in saying that 'we can never know the whole or absolute truth' about the Holocaust or indeed about anything else. I tried to express this belief in the much-misunderstood concluding sentence of my book. But just because we can never attain the whole or absolute truth, just because we make mistakes in our search for the truth about the past, just because there will always be something new to say about any historical subject, it does not follow that there is no such thing as the truth at all. In a similar way, as Susan Haack has pointed out (TLS, 9 July 1999), just because what is accepted as true isn't necessarily so, does not mean that the concept of the truth itself is merely ideological. Purkiss argues 'against "the truth"' and in favour of 'truths' that are 'local and specific', but it is impossible on the evidence of her 'Response' to see what the difference is. Truth, as Haack points out, is not relative to perspective, though what is accepted as true is; 'a statement is true if and only if things are as it represents them to be.' So there cannot be incompatible truths - and although some postmodernists argue that there can (despite the fact that, as Haack notes, 'incompatible' means 'cannot be jointly true'), I don't see Purkiss taking this position in her 'Response'. So if her 'local and specific truths' fit together with other truths, then they are true, they are aspects of the truth, and Purkiss believes in the truth.

Indeed, as Haack says, 'everyone who believes anythingimplicitly acknowledgesthat there is such a thing as truth', and it is encouraging to see Purkiss, along with other critics of my book, acknowledging this by attacking what they see as errors and falsehoods in it, even though for the most part they miss their target. When Purkiss sweepingly condemns a variety of representations of witchcraft as 'wrong because all allow the play of fantasy to replace rigorous reading' she is indeed correct to say that this suggests she believes in truth, even though earlier in the same paragraph she argues 'against "the truth"'. The 'fantasies' she describes, therefore, are not 'local and specific' truths but falsehoods. 'Rigorous reading', one presumes, will deliver the truth rather than fantasies and falsehoods. So far, so good, then; we are in agreement.

And yet Purkiss also describes 'attempts to find out "what really happened" as 'doomed'. Questions such as why witchcraft persecution in England rose and fell are, she says on page 3 of her book 'unanswerable', and she goes on: 'I do not try to answer the empirical questions that preoccupy many of my contemporaries' (p. 4). The obvious question to ask is, if questions about why witchfraft persecutions rose and fell are unanswerable, then why aren't questions about the meaning of texts unanswerable? If attempts to find out what really happened are doomed, then why aren't attempts to find out what is really there in texts like Religion and the Decline of Magic doomed as well? Purkiss indeed concedes this argument in principle, as we have seen, but I doubt whether she would accept it in practice when applied to the questions asked - and answered - in her own book.

Purkiss argues against 'attempts to squeeze disparate materials into a large and general hypothesis', but her feminist approach itself is based on such a hypothesis about early modern women empowering themselves by fantasising that they were witches (an interesting idea, let me make clear, and one that deserves further debate and research), and more generally on an even larger and more general set of hypotheses about the power-driven nature of gender relations in past and present. All of this is accompanied by repeated attacks on historians and the historical profession for not reading sources rigorously, failing to note the rhetorical content of documents, and naively 'treating each text as a transparent window onto the mind of its wri ter or the mentalité of an entire society' (p. 71). I think these attacks are largely misplaced, but no-one could dispute the wisdom of the advice they imply. Once more, it comes down to urging historians to pay more attention to the style and content of the documents they use, the motives, if they can be discovered, of their authors, the audience to which they were addressed, and so on. Who could quarrel with that?

All of these multiple and varied misreadings, misinterpretations and misrepresentations of my book, from Burleigh to Purkiss and beyond, might indeed suggest, as Purkiss notes in her 'Response', that 'all of us have less control over meaning than we might like'. But in accusing me of being unable to read her own work accurately, Purkiss is conceding the point that we can in fact exercise some control over the meaning of the texts we generate, by putting something there that is capable of being read accurately. If this statement does not mean that we mean what we say when we write a sentence, and that readers can discover this if they read accurately what we have written, then it does not mean anything at all.

Purkiss says that 'meanings are generated quite independently of intentions', but this is putting it in too bald and extreme a way; meanings are not generated entirely independently of intentions, otherwise Purkiss would have no reason to criticise my reading of the meaning of her work; they do bear some relation to them, though it is not necessarily either exclusive or direct (all points made in In Defence of History). The truth of these observations is to some extent borne out by this debate. In some cases, I've obviously laid myself open to what I think is misinterpretation, perhaps by failing to make myself entirely clear (e.g. in the definition of objectivity) or by expressing myself poorly (for example, in some instances which I tried to correct in the U.S. edition of the book, and others noted above). In other cases, I think I have been the victim of a careless reading of my text in which conservative or postmodernist critics have seen what they wanted to see and ignored all the nuances and qualifications I advanced in respect of the argument they have been attacking. In still further cases I have suffered from deliberate distortion or manipulation of what I have said. And in a few instances critics have read into my book things that simply weren't there.

Because the debate involves large theoretical and methodological issues, it is important to have the space this website allows to reply in detail and at length. At nearly 24,000 words, indeed, this Reply is now about a third as long as the text of the book itself. I confess I have been taken aback by the sheer variety and by the utterly contradictory nature of the responses which the book has elicited. I would not have thought it possible for a single book to be read, or misread, in so many different ways. But stimulating debate was one of the book's principal aims; it was never intended to close down discussion (nor would that have been a feasible aim in any case). As those who have attacked the book for failing to confront the major postmodernist thinkers have rightly if somewhat superfluously pointed out, the issues at stake cannot be settled in a couple of hundred pages. It is important that the debate continues, therefore. So, as Bernd Roeck remarks in his review of the German edition of the book, es muß also weitergedacht werden - we must carry on thinking.

Discourse index | back to the top