History In Focus logo and homepage link

History in Focus

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

What is History?

Book cover:What is History?

Book Review (reappraisal)


What is History?

E.H. Carr (1892-1982)
Penguin, 1961, Second Edition 1987 (including Preface to Second Edition and a new chapter 'From E.H. Carr's Files: Notes towards a Second Edition of What is History?')

Professor Alun Munslow

University of Staffordshire

Edward Hallett Carr's contribution to the study of Soviet history is widely regarded as highly distinguished. In all probability very few would argue against this assessment of his multi-volume history of Soviet Russia. For the majority of historians he pretty much got the story straight. However, for several years there was disagreement about his contribution to the analytical philosophy of history. His ideas were outlined in What is History? first published in 1961. For many today What is History? is the most influential book on history thinking published in Britain this century. For many years, however, the methodologically foundationalist wing of the history profession regarded the book as espousing a dangerous relativism. This has now all changed. Arguably the central ideas in the book constitute today's mainstream thinking on British historical practice. Most British commentators, if not that many in America, acknowledge the significance and influence of the book. In this review I want to establish why it is What is History? now occupies a central place in British thinking about the relationship between the historian and the past. I conclude that the important message of What is History? - fundamentally misconceived though I believe it to be - lies in its rejection of an opportunity to re-think historical practice. This failure has been most significant in rationalising the epistemologically conservative historical thinking that pervades among British historians today.

John Tosh, in the most recent edition of his own widely read methodological primer The Pursuit of History describes Carr's book as "still unsurpassed as a stimulating and provocative statement by a radically inclined scholar" (Tosh 1991: 234). Keith Jenkins, much less inclined to view Carr as a radical scholar, nevertheless confirms the consequential nature of What is History? suggesting that, along with Geoffrey Elton's The Practice of History both texts are still popularly seen as "'essential introductions' to the 'history question"' (Jenkins 1995: 1-2). Jenkins concludes both Carr and Elton "have long set the agenda for much if not all of the crucially important preliminary thinking about the question of what is history" (Jenkins 1995: 3).

So, according to Tosh and Jenkins, we remain, in Britain at least, in a lively dialogue with What is History?. Why should this be? The reason is, as most British historians know, to be found in the position Carr took on the nature of historical knowledge. A position that brought him into a long conflict with, among others, the Tudor historian and senior Ambassador at the Court of 'Proper' Objectivist History Geoffrey Elton. Again I turn to John Tosh for his comment that "The controversy between Carr and Elton is the best starting-point for the debate about the standing of historical knowledge" (Tosh 1991: 236). Until Jenkins' recent re-appraisal of Carr's philosophy of history, Carr had been misconstrued almost univer among British historians as standing for a very distinctive relativist, if not indeed a sceptical conception of the functioning of the historian.

Explaining Carr's 'radicalism' the philosopher of history Michael Stanford has claimed Carr "insisted that the historian cannot divorce himself from the outlook and interests of his age (sic.)" (Stanford 1994: 86). Stanford quotes Carr's own claim that the historian "is part of history" with a particular "angle of vision over the past" (Stanford 1994: 86). As Stanford points out, Carr's "first answer...to the question 'What is History?"' is that it is a continuous "process of interaction between the historian and his facts, an unending dialogue between the present and the past". While this was not a fresh insight with Carr, it still carved him out for a number of years as someone with a novel stance. However, over time, the effect of his argument (which generated such initial notoriety) was to increasingly balance the excesses of the hard core empiricists. In What is History? Carr propelled British historiography toward a new equilibrium - one that pivoted on a new epistemological certitude.

The claim to epistemological radicalism on behalf of Carr does not seem to me especially convincing. Why? My doubts about the message in What is History? is the product of my present intellectual situatedness as a historian (a writer about the past). Today, with our greater awareness of the frailties and failures of representationalism, referentialism, and inductive inference, more and more history writing is based on the assumption that we can know nothing genuinely truthful about the reality of the past. It would be tempting, but wholly incorrect, to say that history's pendulum has swung far more to the notion of history as a construction or fabrication of the historian. Rather, what has happened, is that our contemporary conditions of existence have created a much deeper uncertainty about the nature of knowledge-creation and its (mis-)uses in the humanities. It is not about swings in intellectual fashion.

It follows, a growing number of historians believe that we don't 'discover' (the truthful?' 'actual?' 'real?' 'certain?') patterns in apparently contingent events because, instead, we unavoidably impose our own hierarchies of significance on them (this is what we believe/want to see/read in the past). I do not think many historians today are naive realists. Few accept there must be given meaning in the evidence. While we may all agree at the event-level that something happened at a particular time and place in the past, its significance (its meaning as we narrate it) is provided by the historian. Meaning is not immanent in the event itself. Moreover, the challenge to the distinction of fact and fiction as we configure our historical narratives, and further acknowledgments of the cognitive power of rhetoric, style and trope (metaphors are arguments and explanations) provide not only a formal challenge to traditional empiricism, but forces us to acknowledge that as historians we are making moral choices as we describe past reality.

Does all this add up to a more fundamental criticism of historical knowing than Carr imagined in What is History?? I think so. If this catalogue is what historical relativism means today, I believe it provides a much larger agenda for the contemporary historian than Carr's (apparently radical at the time) acceptance that the historian is in a dialogue with the facts, or that sources only become evidence when used by the historian. As Jenkins has pointed out at some length, Carr ultimately accepts the epistemological model of historical explanation as the definitive mode for generating historical understanding and meaning (Jenkins 1995: 1-6, 43-63). This fundamentally devalues the currency of what he has to say, as it does of all reconstructionist empiricists who follow his lead. This judgment is not, of course, widely shared by them. For illustration, rather misunderstanding the nature of "semiotics - the postmodern?" as he querulously describes it, it is the claim of the historian of Latin America Alan Knight that Carr remains significant today precisely because of his warning a generation ago to historians to "interrogate documents and to display a due scepticism as regards their writer's motives" (Knight 1997: 747). To maintain, as Knight does, that Carr is thus in some way pre-empting the postmodern challenge to historical knowing is unhelpful to those who would seriously wish to establish Carr's contribution in What is History?. It would be an act of substantial historical imagination to proclaim Carr as a precursor of post-modernist history.

Carr is also not forgotten by political philosopher and critic of post-modernist history Alex Callinicos, who deploys him somewhat differently. In his defence of theory in interpretation (Marxist constructionism in this case), Callinicos begins with the contribution of a variety of so called relativist historians of which Carr is one (others include Croce, Collingwood, Becker and Beard). Acknowledging the "discursive character of historical facts" (Callinicos 1995: 76) Callinicos quotes Carr's opinion (following Collingwood) that the facts of history never come to us pure, but are always refracted through the mind of the historian. For Callinicos this insight signals the problem of the subjectivity of the historian, but doesn't diminish the role of empirically derived evidence in the process of historical study.

Of course Carr tried to fix the status of evidence with his own objections to what he understood to be the logic of Collingwood's sceptical position. Collingwood's logic could, claims Carr, lead to the dangerous idea that there is no certainty or intrinsicality in historical meaning - there are only (what I would call) the discourses of historians - a situation which Carr refers to as "total scepticism" - a situation where history ends up as "something spun out of the human brain" suggesting there can be no "objective historical truth" (Carr 1961: 26). Carr's objectivist anchor is dropped here. He explicitly rejected Nietzsche's notion that (historical?) truth is effectively defined by fitness for purpose, and the basis for Carr's opinion was his belief in the power of empiricism to deliver the truth, whether it fits or not (Carr 1961: 27). Historians ultimately serve the evidence, not vice versa. This guiding precept thus excludes the possibility that "one interpretation is as good as another" even when we cannot (as we cannot in writing history) guarantee 'objective or truthful interpretation'.

Carr wished to reinforce the notion that he was a radical. As he said in the preface to the 1987 Second Edition of What is History? "...in recent years I have increasingly come to see myself, and to be seen, as an intellectual dissident' (Carr 1987: 6). But his contribution really lies in the manner in which he failed to be an epistemological radical. In the precise manner of his return to the Cartesian and foundationalist fold lies the importance of What is History? The book's distinction resides in its exploration and rapid rejection of epistemological scepticism - what I call post-empiricism. From the first chapter Carr accepts relativism would an unacceptable price to pay for imposing the historian on the past beyond his narrow definition of dialogue. Dialogue even cast as interrogation is all very well and good, but an intervention that cannot ultimately become objective is quite another matter. After all, Carr argues, it is quite possible to draw a convincing line between the two.

While confirming the ever present interaction between the historian and the events she is describing, Carr was ultimately unwilling to admit that the written history produced by this interaction could possibly be a fictive enterprise - historians if they do it properly, (their inference isn't faulty and/or they don't choose to lie about the evidence) will probably get the story straight. This argument still appeals to many historians today for whom the final defence against the relativism of deconstructionism lies in the technical and forensic study of the sources through the process of their authentication and verification, comparison and colligation.

In Britain, most realist-inspired and empiricist historians thus happily accept the logical rationalisation of Carr's position - that of the provisional nature of historical interpretation. This translates (inevitably and naturally it is argued) as historical revisionism (re-visionism?). The provisionality of historical interpretation is a perfectly normal and natural historian's state-of-affairs that depends on discovering new evidence (and revisiting old evidence for that matter), treating it to fresh modes analysis and conceptualisation, and constantly re- contextualising it. For illustration, in my working career (since the early 1970s) the omission of women in history has been 'rectified', and now has moved through several historiographical layers to reach its present highly sophisticated level of debate about the possibility for a feminist epistemology(ies). So, new evidence and new theories can always offer new interpretations, but revisionist vistas still correspond to the real story of the past because they correspond to the found facts.

In fact, with each revision (narrative version?) it is presumed by some that we know better or see more clearly the nature of the past. So, we are for ever inching our way closer to its truth? Arthur Marwick makes the claim that by standing on "...the powerful shoulders of our illustrious predecessors" we are able both to advance "the quality" and "the 'truthfulness' of history" (Marwick 1970: 21). Standing on the shoulders of other historians is, perhaps, a precarious position not only literally but also in terms of the philosophy of history. No matter how extensive the revisionary interpretation, the empiricist argument maintains that the historical facts remain, and thus we cannot destroy the knowability of past reality even as we re-emphasise or re- configure our descriptions. Marxists and Liberals alike sustain this particular non sequitur which means they can agree on the facts, legitimately reach divergent interpretations and, it follows, be objective. The truth of the past actually exists for them only in their own versions. For both, however, the walls of empiricism remain unbreached. The (empiricist-inspired) Carr- endorsed epistemological theory of knowledge argues that the past is knowable via the evidence, and remains so even as it is constituted into the historical narrative. This is because the 'good' historian is midwife to the facts, and they remain sovereign. They dictate the historian's narrative structure, her form of argumentation, and ultimately determine her ideological position.

For Carr, as much as for those who will not tarry even for the briefest of moments with the notion of epistemological scepticism, Hayden White's argument that the historical narrative is (a story) as much invented as found, is inadmissible because without the existence of a determinate meaning in the evidence, facts cannot emerge as aspects of the truth. Most historians today, and l think it is reasonable to argue Carr also endorses this view in What is History?, accept Louis Mink's judgment that "if alternative emplotments are based only on preference for one poetic trope rather than another, then no way remains for comparing one narrative structure with another in respect of their truth claims as narratives" (Vann 1993: 1). But Carr's unwillingness to accept the ultimate logic of, in this instance, the narrative impositionalism of the historian, and his failure to recognise the representational collapse of history writing, even as he acknowledges that "the use of language forbids him to be neutral" (Carr 1961: 25), has helped blind many among the present generation of British historians to the problematic epistemological nature of the historical enterprise.

Take the vexed issue of facts. Carr's answer to the question "What is a historical fact?" is to argue, pace Collingwood (Collingwood 1994: 245) that facts arise through "...an a priori decision of the historian" (Carr 1961: 11). It is how the historian then arranges the facts as derived from the evidence, and influenced by her knowledge of the context, that constitutes historical meaning. For Carr a fact is like sack, it will not stand up until you put 'something' in it. The 'something' is a question addressed to the evidence. As Carr insists, "The facts speak only when the historian calls on them: it is he who decides to which facts to give the floor, and in what order or context" (Carr 1961: 11).

It is easy to see why Elton and others like Arthur Marwick misconstrue the (Collingwood-) Carr position when Carr says such things because, if pushed a little further allows historians to run the risk of subjectivity through their intervention in the reconstruction of the past. Carr, of course, denies that risk through his objectivist bottom line. There is clear daylight between this position and that occupied by Hayden White. It is that while historical events may be taken as given, what Carr calls historical facts are derived within the process of narrative construction. They are not accurate representations of the story immanent in the evidence and which have been brought forth (set free?) as a result of the toil, travail, and exertion of the forensic and juridical historian.

Since the 1960's Carr's arguments have moved to a central place in British thinking and now constitute the dominant paradigm for moderate reconstructionist historians. This is because, as Keith Jenkins has demonstrated, Carr pulls back from the relativism which his own logic, as well as that of Collingwood, pushes him. In the end Carr realises how close to the postempiricist wind he is running, so he rejects Collingwood's insistence on the empathic and constitutive historian, replacing her with another who, while accepting the model of a dialogue between past events and future trends, still believes a sort of objectivity can be achieved. This then is not the crude Eltonian position. It is a claim to objectivity because it is position leavened by a certain minimum self-reflexivity. This is a conception of the role of the historian affirmed by the most influential recent American commentators Joyce Appleby, Lynn Hunt and Margaret Jacob who claim there can be no postmodern history by repeating (almost exactly) Carr's fastidious empiricist position. Carr received only one oblique reference in their book Telling the Truth About History which may help explain why they re-packed Carr's position as practical realism (Appleby, Hunt and Jacob 1994: 237, 241-309 passim). Is it that his position is so central to the intellectual culture of mainstream history that it wasn't even necessary to reference him? In the early 1990's the historian Andrew Norman endorsed the Carr mainstream position more directly by arguing writing history necessitates historians engaging directly with the evidence "A good historian will interact dialogically with the historical record" (Norman 1991: 132). Facts in history are thus constituted out of the evidence when the historian selects sources contextually in order to interpret and explain that to which they refer, rather than in the narrative about which they describe.

It is because Carr remains at the end of the day a convinced objectivist despite (or because of?) his dalliance with relativism - that his legacy in What is History? is still so potent among British historians. His objectivist appeal in What is History? is potent because it is not of the naive variety. We know the Carr historian cannot stand outside history, cannot be non-ideological, cannot be disinterested, or be unconnected to her material because she is dispassionate. But she is telling us what actually happened because she can overcome those obstacles. She knows that the significance of the evidence is not found solely in the evidence. The historian, as he said, "does not deal in absolutes of this kind" (Carr 1961: 120). There can be no transcendental objective measures of truth. However, while accepting the "facts of history cannot be purely objective, since they become facts of history only in virtue of the significance attached to them by the historian" (Carr 1961: 120), Carr was forced by his naked objectivist desire to underplay the problems of historical form and the situatedness of the historian. he did this by arguing that the standard for objectivity in history was the historian's "sense of the direction in history" by which he meant the historian selected facts based not on personal bias, but on the historian's ability to choose "the right facts, or, in other words, that he applies the right standard of significance" (Carr 1961: 123).

Carr's philosophical sleight-of-hand produced the objective historian who "has a capacity to rise above the limited vision of his own situation in society and history" and also possesses the capacity to "project his vision into the future in such a way as to give him a m-ore profound and more lasting insight into the past than can be attained by those historians whose outlook is entirely bounded by their own immediate situation" (Carr 1961: 123). The objective historian is also the historian who "penetrates most deeply" into the reciprocal process of fact and value, who understands that facts and values are not necessarily opposites with differences in values emerging from differences of historical fact, and vice versa. This objective historian also recognises the limitations of historical theory. As Carr says a compass "is a valuable and indeed indispensable guide. But it is not a chart of the route" (Carr 1961: 116).

Social theory historians (constructionists) understand past events through a variety of methods statistical and/or econometric, and/or by devising deductive covering laws, and/or by making anthropological and sociological deductive-inductive generalisations. For hard-core reconstructionist-empiricists on the other hand, the evidence proffers the truth only through the forensic study of its detail without question-begging theory. These two views are compromised by Carr's insistence that the objective historian reads and interprets the evidence at the same time and cannot avoid some form of prior conceptualisation - what he chooses simply (or deliberately loosely?) to call "writing" (Carr 1961: 28). By this I think he means the rapid movement between context and source which will be influenced by the structures and patterns (theories/models/concepts of class, race, gender, and so forth) found, or discovered, in the evidence.

For Carr the evidence suggests certain appropriate explanatory models of human behaviour to the objective historian which will then allow for ever more truthful historical explanation. This sleight-of-hand still has a certain appeal for a good number of historians today. The American historian James D. Winn accepts this Carr model of the objective historian when he says that deconstructionist historians "...tend to flog extremely dead horses" as they accuse other historians of believing history is knowable, that words reflect reality, and their un-reflexive colleagues still insist on seeing the facts of history objectively. Few historians today, thanks to Carr, work from these principles in pursuit of, as Winn says "...the illusory Holy Grail of objective truth" but strive only to ground "...an inevitably subjective interpretation on the best collection of material facts we can gather" (Winn 1993: 867-68). At the end of the day, this position is not very much different to the hard line reconstructionist-empiricist.

What Carr is doing then in What is History? is setting up the parameters of the historical method - conceived on the ground of empiricism as a process of questions suggested to the historian by the evidence, with answers from the evidence midwifed by the application to the evidence of testable theory as judged appropriate. The appropriate social theory is a presumption or series of connected presumptions, of how people in the past acted intentionally and related to their social contexts. For most objective historians of the Carr variety, his thinking provides a more sympathetic definition of history than the positivist one it has replaced, simply because it is more conducive to the empirical historical method, and one which appears to be a reasoned and legitimate riposte to the deconstructive turn.

For such historians Carr also deals most satisfactorily with the tricky problem of why they choose to be historians and write history. The motivation behind the work of the historian is found in the questions they ask of the evidence, and it is not, automatically to be associated with any naked ideological self- indulgence. Any worries of deconstructionists about either ideology, or inductive inference, or failures of narrative form has little validity so long as historians do not preconceive patterns of interpretation and order facts to fit those preconceptions. Carr would, I think, eagerly challenge the argument that historians are incapable of writing down (reasonably) truthful narrative representations of the past. The position that there is no uninterpreted source would not be a particularly significant argument for Carr because historians always compare their interpretations with the evidence they have about the subject of their inquiry. This process it is believed will then generate the (most likely and therefore the most accurate) interpretation.

So, when we write history (according to the Carr model) our motivation is disinterestedly to re-tell the events of the past with forms of explanation already in our minds created for us through our prior research in the archive. 'Naturally' we are not slaves to one theory of social action or philosophy of history - unless we fall from objectivist grace to write history as an act of faith (presumably very few of us do this? Do you do this?). Instead we maintain our models are generally no more than 'concepts' which aid our understanding of the evidence indeed, which grow out of the evidence. We insist our interpretations are independent of any self-serving theory or master narrative imposed or forced on the evidence. It is the 'common sense' wish of the historian to establish the veracity and accuracy of the evidence, and then put it all into an interpretative fine focus by employing some organising concepts as we write it. We do it like this to discover the truth of the past.

To conclude, Carr's legacy, therefore, shades the distinction between reconstructionism and constructionism by arguing we historians do not go about our task in two separate ways with research in the sources for the facts, and then offering an interpretation using concepts or models of explanation. Rather the historian sets off, as Carr says "...on a few of what I take to be the capital sources" and then "inevitably gets the itch to write". This I take to mean to compose an interpretation and "...thereafter, reading and writing go on simultaneously" (Carr 1961; 28). For Carr this suggests the "...untenable theory of history as an objective compilation of facts...and an equally untenable theory of history as the subjective product of the mind of the historian..." is much less of a problem than any hard-nosed reconstructionists might fear. It is in fact the way in which human beings operate in everyday life, a "...reflection of the nature of man" as Carr suggests. (Carr 1961: 29). Historians, like Everywoman and Everyman work on the evidence and infer its most likely meaning - unlike non-historians we are blessed with the intellectual capacity to overcome the gravitational pull of our earthly tethers.

The idée fixe of mainstream British historians today is to accept history as this inferential and interpretative process that can achieve truth through objectivism. Getting the story straight (from the evidence). The unresolved paradox in this is the dubious legacy of What is History?. I assume a good number of historians recommend Carr to their students as the starting point of methodological and philosophical sophistication, and a security vouchsafed by the symmetry between factualism, objectivism and the dialogic historian. While I am unconvinced by its message, I think this is why What is History? remains, for the majority of British historians, a comforting bulwark against post- constructive and post-empirical history.

November 1997


Appleby, Joyce, Hunt, Lynn, and Jacob, Margaret (1994) Telling the Truth About History, W.W. Norton and Co., London.

Callinicos, Alex (1995) Theories and Narratives: Reflections on the Philosophy of History, Cambridge, Polity Press.

Carr, E.H. (1961) What is History? London, Penguin.

------------ (1987) What is History? (Second Edition) London, Penguin.

Collingwood R.G. (1994) The Idea of History (First published 1946) Oxford, Oxford University Press.

Iggers, Georg, G. (1997) Historiography in the Twentieth Century: From Scientific Objectivity to the Postmodern Challenge, Hanover, NH, Wesleyan University Press.

Jenkins, Keith (1995) On 'What is History?', London, Routledge.
----------- (1997) Postmodern History Reader, London, Routledge.

Knight, Alan (1997) "Latin America" in Bentley, Michael (ed.) Companion to Historiography, London, Routledge.

Marwick, Arthur, (1970) The Nature of History, London, Macmillan.

Munslow, Alun (1997) Deconstructing History, London, Routledge.

------------- (1997) "Authority and Reality in the Representation of the Past" Rethinking History: The Journal of Theory and Practice, Vol. 1, No. 1, Summer, pp. 75-87.

Norman, Andrew (1991) "Telling it Like it Was: Historical Narratives on Their Own Terms", History and Theory Vol. 30, pp. 119-135.

Novick Peter (1988) That Noble Dream: The 'Objectivity Question' and the American Historical Profession, Cambridge University Press, Cambridge.

Stanford, Michael (1994) A Companion to the Study of History, Oxford, Basil Blackwell

Stromberg, Roland N. (1994, Sixth Edition) European Intellectual History Since 1789 Englewood Cliffs, N.J., Prentice Hall.

Tosh, John (1991) The Pursuit of History London, Longman.

Vann, Richard T. (1987) "Louis Mink's Linguistic Turn," History and Theory Vol. 26, No. 1, pp. 1-14.

Winn, James A. (1993) "An Old Historian Looks at the New Historicism," Comparative Studies in Society and History, Vol. 35 No. 4, pp. 859-870.

Carr's What is History? is referenced relatively little in United States' works on historiography. Unlike G.R. Elton, Carr is not referenced in George G. Iggers (1997) Historiography in the Twentieth Century: From Scientific Objectivity to the Postmodern Challenge, Hanover, NH, Wesleyan University Press, or Roland N. Stromberg (1994, Sixth Edition) European Intellectual History Since 1789 Englewood Cliffs, N.J., Prentice Hall, nor Peter Novick (1988) That Noble Dream: The 'Objectivity Ouestion' and the American Historical Profession, Cambridge University Press, Cambridge. As I note later Carr receives only one brief reference in Joyce Appleby, Lynn Hunt, and Margaret Jacob (1994) Telling the Truth About History, W.W. Norton and Co., London. Carr has also disappeared from the postmodernist reckoning. He is not referenced nor indexed in Keith Jenkins (1997) Postmodern History Reader, London, Routledge.

Back to reviews index

Back to top