History In Focus logo and homepage link

History in Focus

the guide to historical resources • Issue 2: What is history? •

What is history?

jenkins (14K)appleby (3K)

Author's response (Prof. Alun Munslow)


A series of books published by Routledge about postmodernism and history

(see text for examples)

Professor Patrick Karl O'Brien

Professor Emeritus of the University of London

History is not what it was. Professor Patrick O'Brien's engagement with postmodern history certainly signals that particular fact. Professor O'Brien has written a lucid, cogent and skillful defence of history conceived as a particular kind of discipline. In the process he has struck a generous and open-minded note - something which is all too often missing in these kinds of exchanges. It is my intention also to be fair minded and, I hope, equally polite. But a brief prefatory comment. It serves no particularly useful purpose to argue about Professor O'Brien's description of Jenkins, Southgate and myself as postmodernists. I shall not presume to answer for Southgate and Jenkins as to whether they would accept the description. For myself I would prefer a designation that does not signify what is often regarded as representing subversive (defined in a destructive sense) thinking about history. Like the journal for which I am the UK editor (Rethinking History published by Routledge), I would prefer to be regarded as a historian - perhaps no different in this respect to all historians - who rethinks the nature of history in a manner that is productive but which, above all, extends the form and content boundaries and arrangements of the discipline of the-past-as-history.

Having said that I would refer Professor O'Brien to my editorial for the first issue of the Rethinking History. In that I argued the discipline today was widely held not to be "...an autonomous, objective, representationally accurate, or disinterested discipline" (Munslow 1997(a): 1), and that, consequently, the greatest mistake of postmodernism was that it has a tendency to "criticise a historical method that has never really existed" (Munslow 1997(a): 1). History, I suggested, has never been quite so naive as postmodernists often seem to claim. So I think I am here already in agreement, at least on this point, with Professor O'Brien. The postmodernist charge that sceptical empiricist historians fail to acknowledge the ambiguous nature of representation I would agree is somewhat overdrawn. I have always thought it would be difficult especially for sceptical empiricists - to be quite so unaware of the pitfalls of their own language-use and the historicist problems they encounter in decoding their sources as the criticism implies they are. No, I do not think that argument - whether it is postmodern or not - is very convincing. It serves no purpose, as Professor O'Brien says, for anyone intent on criticising history to set up the straw man of empiricist absolutism. The problem with history is, in fact, much deeper than that.

Having, I hope disposed of one area of potential confusion I turn, therefore, to rather more significant criticisms that have been leveled at Professor O'Brien's version of history since the 1980s. In Britain these were first focused by Keith Jenkins' 1991 'postmodernist manifesto' (Rethinking History). And here I am happy to publicly congratulate the history editor at Routledge and the company itself for her and its foresight in publishing the likes of Jenkins, Southgate, Bunzl and McCullagh who have debated over nearly ten years sceptical empiricism's epistemological character. Jenkins is an important, perhaps the most important critic of Professor O'Brien's variety of history. Before my comments perhaps I might paraphrase Jenkins' position? As I read him Jenkins' criticisms are at least three fold: first, that in spite of its claims to a sophisticated and sceptical empiricism, conventional history does in fact sanction a guileless belief in the possibility of the discovery of past reality/truth; second, that it still believes the knowing subject must remain unchallenged (characterised as the disinterested historian - and still primarily masculine); and, finally, endorses explanation through contextualisation via inductive inference as the only way to generate 'truthful interpretations' (an oxymoron as Jenkins has pointed out). The response to these criticisms is usually met - as Professor O'Brien has met them with the comment that much of this has long been known. Well, that may be the case, but if it is why do so many historians persist in writing history on the premise that they are discovering the real history of the past because they make single truth-conditional statements? For example saying the first point of the Chartist Charter was annual Parliaments may be true, but this, as most historians would agree, is not. the whole story.

As Giambattista Vico pointed out in the early eighteenth century, history had cognitive and epistemological or post-empiricist problems then, and I think it still does. Where I diverge from Professor O'Brien's defence of sceptical empiricist history, therefore, and where I start to agree with Jenkins, is that I believe it is predicated on the modernist or Enlightenment- inspired belief that we historians possess a privileged set of procedures that permit us to get ever closer to the likely truth of the-pastas-history. I am not convinced by McCullagh, for instance, that the sceptical empiricist procedures of verification and comparison of evidence, or hypothesis-testing or, for that matter, inference/explanation to the best fit, and all undertaken by the judicious, distanced, and independently- minded observer-historians will necessarily generate truth conditional statements (McCullagh 1984, 1998). But even if I were to grant that they do (and the judicious and distanced historian did exist), I fail to see how the move from the individual truth-conditional or factual statement about annual Parliaments, to the compositional level of a narrative length historical explanation, can retain anything other than a 'sense' of reality what Barthes famously called the reality effect ("The Discourse of History" 1967).

It is Barthes' suggestion that the connection between language and the-past-as-history does not depend on any actual link between evidence and its disposition as historical fact, but rather on the reality-effect created by our belief in the correspondence theory of knowledge. It is this which allows us to reconstruct the-past-as-history. It is my position that the linguistic study of the historical narrative does not, a priori, franchise to history an exclusive epistemological status that discriminates it from fiction in its form. In other words, it is not the content. of history that generates its own peculiar form. If it were then history would be just a collection of truthful statements organised by the appropriate concept that the evidence has thoughtfully provided for us. And it clearly is more than that.

Why? Because, as Professor O'Brien accepts, it is historians rather than the past that generates (writes, composes?) history - a history that the O'Brien type of modernist and liberal humanist subject and knowledge-centred historian assumes is made up of accurately knowable individual events, processes and, above all, agent intentions. But instead of doubting that history really is an epistemology that has privileged access to the past reality of agent-intentionality, Professor O'Brien instead chooses just to reject one doubtful version of it - the robust Hempelian covering law model believing it is unnecessary to the study of the sources or the creation of unique historical facts. The issue between social scientist advocates of such a model (like Karl Popper, Patrick Gardiner, Ernest Nagel and Clayton Roberts) and the O'Brien mainstream constructionist doubters, it seems to me, should not really be over positivism as an epistemology. It should instead be about whether it is the primacy of the content of history (which in the O'Brien version of history, lest we forget, is about understanding agent intentionality) that provides history with its epistemological viability. Apart from the fact that a good number of historians today still believe in the social sciences (if not full-strength covering laws) I am forced to question the assumption that the belief in a knowable agent-intentionality is of itself enough to constitute sceptical-empiricist history as a licit discipline (Snooks 1998)?

Frank Ankersmit has, for example, already noted the attempt by mainstream (O'Brien-type?) constructionist historians to find and defend a weak form of the covering law model that could be deployed to invigorate their historical method. Such a method is that founded on inductive inference. Its appeal rests precisely on the compromise O'Brien makes between the extremes of sceptical-empiricist reconstructionism and positivist history (Ankersmit 1994). The compromise, he notes, comes from the desire widespread among the constructionist mainstream, to reinforce the liberal humanist belief in agent-intentionality with inductive inference as its preferred way of creating knowledge in the humanities. Happily, history depends on the inductive inference of agent intentionality, and agent-intentionality can only be known through induction. Choice and purpose are exercised by the knowing subject whether she be historical agent or historian.

The result of this particular Enlightenment intellectual invention is that laws of human behaviour constitute only possible/probable explanations of the causes of events: explanations to the best fit. It follows historians have plenty of epistemological choices to make within this preferred frame of sceptical-empiricism. They can delight in a theory-free existence if that is what they want (even to the extent of fantasising that they can reconstruct the past as it actually happened), or they can adopt highly complex but essentially non- deterministic (humanistic) explanations with the intention of saving as a good number have, obsolete, utopian or deluded historical agents from the condescension of posterity, or go for some heavy-handed determinist history if that is there personal preference. The majority of historians these days are modernists. They are, like Professor O'Brien, epistemological constructionist middle-of-the-roaders. They see themselves as being sensible explorers of large-scale social/institutional structures and while their imaginative speculations may never stretch to the discovery of absolute individual agent intentionality, will still get pretty close through the process of contextualisation. However, regardless of the kinds of history that sceptical-empiricism has produced over the years most historians converge in the belief that there is only one proper kind of historical procedure, the kind that is the discovery and factualist rendition of agent-intentionality as informed by appropriate conceptual theory. It is this that makes history a truth-acquiring epistemology.

It is here I submit that sceptical-empiricism cannot escape its pre- figurative status. While as Professor O'Brien claims it is about getting closer to the truth of the real past via the sovereignty of its content, it is never actually able to escape from the ways in which its practitioners prefigure or preconceive it - be it philosophically (e.g., agent-intentionality), linguistically (e.g., trope), conceptually (e.g., argument), or ideologically (e.g., Marxist). I do not believe it is just a matter, as I read Professor O'Brien, of noting history's shortcomings as a cultural discourse or practice and then thanking one's lucky stars that the safety net of the ('proper' treatment of the) evidence will ultimately save us from untruthful interpretations. I think the priority for history should be philosophical (and tropic, etc.) more than it is empirical. We should, rather, pose basic questions about history's metaphysical status. What are the implications for the study of the past given that history, though it may be empirically founded, is a heavily authored second-hand knowledge constituted and organised by the historian. This may mean a linguistic turn, or a philosophical turn, or a discursive turn. I welcome any or all of them if they turn historians away from the myopic and single minded belief that they can recover the meaning of the- past-as-history through the mechanism of factualism.

So, I do have some basic doubts about the assumed connections made by Professor O'Brien between evidence, concept, interpretation, and meaning. I would certainly not wish to affront either Professor O'Brien or my friends and colleagues in the profession by suggesting they should be cast into the dustbin of the humanities. However, I am worried by the 'yes, but' attitude of many of them. Yes, there are any number of epistemological and ontological problems with history: it is culturally situated, it is emplotted/narrated, it is dependent upon evidence selected by the historian, it is unavoidably ideological, it is inferential, it is not positivist, its meaning is never finished, truth is always deferred. Yes, there are all these problems, but things can correlate with words and concepts.

Now I do understand Professor O'Brien's practical and realist position and I have a certain, though be it limited, sympathy with it. I do not believe, for example, that the knowable world sprang into existence less than a minute ago when I started typing this paragraph - I have too many memories (and well-attested debts) that suggest that I have little other than a very tangible past. There is a real past back there and it speaks to me in the single existential bank statement. But memories and debts, even when both are accumulated over many years, will never furnish me with the true meaning of my life. Historians can make any number of truth-conditional individual statements about any person's life, but when we crank them all up to the level of conceptualisation and interpretation a qualitative change occurs that inevitably leaves factualism a long way back in its wake. So, I have my doubts about the extent to which history can be studied for its own factual sake - which seems to be the conclusion of conventional O'Brien history.

Another thrust of my criticism of Professor O'Brien's kind of history is that while it nods in the direction of the importance of its narrative form, it does so in order to avoid acknowledging history's fundamental epistemological problem - its incapacity to faithfully render truthful (and therefore non-racialised, non-Eurocentric, non-masculinised, non-class, non-nationalistic, etc.) narratives because of its modernist bourgeois liberal humanist assumptions of knowing subject and objective knowledge. Professor O'Brien notes but dismisses the argument that the form that we impose on history gives it its meaning. This is not going to butter many parsnips with feminist post-structuralists and for good reason I would think. Now, even if we conveniently ignore the issue of whether history is a modernist project that has outlived its usefulness or, for that matter, that postmodernism can only offer straw men arguments, I agree with Professor O'Brien that history is a literary artifact and I also agree with his conclusion that the guild still pays far too much attention to its evidential base at the cost of usefully considering how it is represented. But this acknowledgment falls far short of recognising the full implications of how we prefigure history by trope and emplotment (not to mention argument, gender, ethnocultural, or ideological preference).

While the historian can learn things about her use of language, style, and metaphor, the bottom line for Professor O'Brien is always one of referentiality, truth, and the right protocol of investigation. But as I suggested earlier this ignores the problem that when the referential is translated into narrative it effectively ceases to exist. The referential is now under the sway of imaginative interpretation. Truth, as a verified existential statement, is effectively occluded by its own textualisation - a textualisation that is generated by the author-historian and her (or his) gender, political, ethical, tropic, epistemic or whatever preferences. As Professor O'Brien admits, truth is not something that historians talk about as an absolute, if indeed they talk about it all. The absence of discussions about truth in the polite company of historians is, of course, supposed to be the result of our sophisticated practical realist sceptical empiricism. I believe it is more the result of the polite refusal to acknowledge, let alone accept, the epistemological implications of a past that is always textualised if you like unavoidably pre-shrunk with prior, deferred and imposed meaning(s). History comes with historians as standard equipment, all of whom endorse different theories about the nature of change, and all of whom write about it in ways dedicated to the creation of their own versions of the- past-as-history though be it in the name of objectivity.

I sincerely doubt that the kind of minimal linguistic self-reflexiveness suggested by Professor O'Brien will do much to change the essential commitment of his ideal-type historian to her liberal humanist investment in agent-intentionality and the knowing (male!?) subject. History remains for Professor O'Brien inflexibly about acquiring the truth(s) of past reality through the correspondence of the document to the historian's written interpretation. This is taken as axiomatic because the thinking behind it remains inured to the poststructuralist revolution. But we really should try to address the consequences of the fact that history is generated through the imposition by historians of a personally chosen narrative form on the past - and that it is a process influencing more than the writing-up stage. Elsewhere I have called this the deconstructive consciousness (Munslow(b) 1997).

By this I mean the kind of thinking about history that recognises its form as a written narrative and suggests that form also provides a model for the study of the past itself. Why? Because we are narrativised creatures. I agree with Professor O'Brien historians do not just read the data of the past, instead - and on this I also go along with Vico and Kant - we negotiate it through our concepts. History's a priori is the human power to conceptualise through narrative. History for me is not just about lived experience as retold by historians as a narrative, but the experience of existing within particular narratives whether they be gendered, subaltern, race, class or whatever. The past did exist, but does not have meaning until the historian writes it as history.

I am happy to agree with Professor O'Brien that history is not about universal meanings - which is an important poststructuralist point. Upon this we would seem to be agreed. But where we differ is on how we reach that conclusion. In other words, the nature of the sceptical empiricist fallback position versus the poststructuralist contention that we do not posses the procedures that will allow us authentic knowledge of the past. It is not a matter of historians being shy about their situatedness, nor less about not making a big deal out of their biases. It is a matter of recognising that if one accepts sceptical-empiricism is up to doing the task of epistemology then there is only one way to research and write history. Anything else is not history. In fact it is likely to be nihilistic, relativist and very possibly dangerous nonsense.

History viewed as an empirical methodology founded on an acceptable level of correspondence between the past and our narrative construction of it is, I suggest, far too narrow a judgement of the job of the historian. Unlike Professor O'Brien, rather than starting with the past we should begin with the thinking behind it, that is, history as its representation. In this way we can confront the foundational sceptical empiricist idea that through proper history it is possible to discover and accurately represent the probable reality of the past. As one commentator said recently in the context of gender history post- structuralism demands that historians challenge the "apparent common sense of thought itself and how thought is inseparably knitted to words" (Baker 1998: 373). I suspect Professor O'Brien might agree with the second point but not the first. Historians are common sensical people. Indeed, to lapse into aphorism briefly, sensible history is sensible history.

Based on what I have said so far our understanding of the past - the facts of history Professor O'Brien mentions - are likewise always presented to the historian in a written or literary form. Facts are never brute always being created by the historian either selected or deselected teleologically for their functionality to the argument, or as mediations of the historian's situation/ideology. As Professor O'Brien implies, facts are never innocent - but their innocence or otherwise is the result of their contextualisation with a whole range of other deliberately selected facts - a preconfigured contextualisation which is part of the constructed process of colligation that eventually produces the interpretation/meaning. It is in this ordering of the evidence in meaningful relationships that the historian's situatedness appears. In writing history we cannot distance ourselves from the creation of meaning in narrative - we conceive historical knowledge as we organise and emplot the data and we infer a meaning.

As Michel Foucault has suggested, the legacy of postmodernism for writing history is the threat posed to the conception of representation as a transparent mode of communication and vehicle for adequately carrying understanding and generating meaning. This is not just at the level of understanding the definition of pauperism as deployed in the early nineteenth century, or grasping when the concept of class became a common- place term, but at the level of a philosophical challenge to the modernist notion that understanding emanates from a centre - the independent knowledge-centred individual subject which we designate variously as the evidence, or the author, or Man. Meaning in history is always created, never found in only the empirical locus. History, because it is an act of will, embodies dissent as well as conformity (and all the spaces between). History is and should be disruptive. History should confront all norms of thinking and believing - not just the what of history but the how and why as well. History, if it is not always a philosophical undertaking that questions how we think and rethink the relationships between subject and object, form and content, fact and fiction, truth and perspective, observer and observed, then I have my doubts as to whether it has been worth all the years I have spent in the archive.

Perhaps I should conclude by summarising what I think my version history is like? Although I have done this in the past maybe I should say it again (Munslow 1997(a)(b))? My history is structured by several assumptions and consequently possesses (what proper historians will likely regard as) peculiar features. I begin with the assumption that history is long overdue a rethinking of its metaphysical status. Doing history is not just a matter of epistemology. It is not just about the procedures through which we know the past. If it were then we could only have Professor O'Brien's version of it - what Keith Jenkins recently called ownsakism (Jenkins 1998: 409-412). This history for its own sake demands a necessary distance between knower and known in writing the-past-as-history. It is imperious in its command for an observer status for the historian. I prefer to think of history as an ontological issue. The essential point about history is, for me, that there is no distance in writing the-pastas-history. The historian is an author rather than a distanced observer/discoverer. I do not want to pretend it is not me creating history when the result of doing so would be to confirm the belief that by spiriting myself out of the process truth immediately fills the vacuum.

This leads me to another attitude I have toward history, one that may or may not be postmodern, namely that unlike Professor O'Brien, who ultimately demands we do it his way, I really do not bother too much of my colleagues pursue history through the sceptical-empiricist method without dwelling too much on its consequences - that is up to them. I suppose I am happy as long as I am allowed to rethink history's boundaries in my own way - and experimentally put form before content if I believe this will extend the frontiers of what it is that historians do and think. To lapse again, this time into irony, allow me to exercise my own agent intentionality. At the end of the day Professor O'Brien does not have to read my history and I do not have to read his. I hope that does not happen. By reading each other we can debate interpretative frameworks and the nature and consequences of our postist intellectual environment.

But even more important than this I do believe history is about ethics and taking up moral positions. Because I have a moral understanding that certain things are right and others wrong, an important feature of my post-empiricist history is that I do not expect my evidence to point me in the direction of objectivised knowledge - the answer. Answers come from moral reasoning as much as empirical realities. This is in part predicated on the ideas of Hayden White and other anti-representationalist philosophers like W.O. Quine, Wilfred Sellars and Richard Rorty, but it primarily derives from my ontological predisposition. History is a story constructed not only by evidence, and argumentation but also by ethical positioning. So my history is historicist in that I want to see history as a contemporary and emancipatory cultural practice, and I do not think a narrow concentration on finding out the empirical reality is the only thing required in order to do that - you can use moral argument and non-empiricist positions just as well to know something about the past.

It is in the space created by the diminution of sceptical- empiricism that possibilities open up for writing the-past-as- history, possibilities that explicitly call into question how the history text is organised as a structure of knowledge with truth claims. The point of doing this is not just to raise epistemological questions (crucially important though they are) but also questions about power. History is a political act, and for me it is about challenging various hegemonies such as class, race, or gender. I think it is especially important for those colleagues who write from a critical and materialist perspective intent on deploying history to recover the marginalised and exploited, to understand that conventional historical thinking compresses the possibilities of meaning in the-past-as-history precisely because of the reasoning of sceptical-empiricism. Not only do the 'proper' procedures of history narrow the discipline's boundaries but they disguise those boundaries as universals.

My history is just another cultural practice that studies cultural practice. It is relativist and that in no way worries me. My history collapses knowledge and representation, and representation and being, and enjoys the permeable relationship of past reality and the present. To recognise that we are textualised creatures does not constrain but rather liberates us. I am content that while our interpretations possess referentiality they do not access reality, and so history can never be what it once was.

February 1999


F.R. Ankersmit, History and Tropology: The Rise and Fall of Metaphor (Berkeley, Cal., University of California Press, 1994).

Mary Baker, "Feminist Post-Structuralist Engagements with History", Rethinking History, Vol. 2, No. 3, Autumn, 1998, pp. 371-378.

Roland Barthes, "El discourse de l'histoire," Information sur les sciences sociales, Vol. 6, No. 4, (1967), pp. 65-75 translated with an introduction by Stephen Bann, Comparative Criticism - A Yearbook, Vol. 3, 1981, pp. 3-20.

Isaiah Berlin, The Sense of Reality (New York, Farrar, Straus and Giroux, 1996).

Martin Bunzl, Real History (London, Routledge, 1997).

R.G. Collingwood, The Idea of History (1946, Revised Edition 1994), pp. 302, 395-395.

Carl Hempel, "The Function of General Laws in History," The Journal of Philosophy, Vol. 34, 1942, reprinted in Patrick Gardiner (Ed.) Theories of History (Glencoe, Ill. The Free Press, 1959).

Anne Janowitz, Lyric and Labour in the Romantic Tradition (Cambridge, Cambridge University Press, 1998).

Keith Jenkins, Rethinking History (London, Routledge, 1991).

-------------------, Review of David Harlan The Degradation of American History, in Rethinking History, Vol. 2, No. 3, 1998, pp. 409-412.

Chris Lorenz, "Can Histories be True? Narrativism, Positivism, and the 'Metaphorical Turn"', History and Theory, Vol. 37, No. 3, 1998, pp. 309-329.

C. Behan McCullagh, Justifying Historical Descriptions (Cambridge, Cambridge University Press, 1984).

----------------------------, The truth of History (London, Routledge, 1998).

Alun Munslow(a), Editorial, Rethinking History, Vol. 1, No. 2, Autumn 1997, pp. 1-20.

----------------------(b), Deconstructing History (London, Routledge, 1997).

Richard Rorty, Objectivity, Relativism and Truth: Philosophical Papers Vol. 1 (Cambridge, Cambridge University Press, 1991).

Graeme Donald Snooks, The Laws of History (London, Routledge, 1998).

Dorothy Thompson, The Chartists: Popular Politics in the Industrial Revolution (New York, Pantheon, 1984).

Original review

Back to reviews index

Back to top