Chapter Seven

The limits of historical knowledge

Historians make many claims for their subject, but can any historical account amount to anything more than its author’s personal take on the past? This chapter looks at the debate surrounding the essential nature of historical work and therefore, to some extent, its value. The positivist position sees history as a form of science, in which historians amass facts from hard evidence and draw valid conclusions; the idealists on the other hand stress that the incomplete and imperfect nature of the historical record obliges the historian to employ a considerable degree of human intuition and imagination. Challenging both positions are the Postmodernists, who point to the highly subjective values and assumptions latent not just in the historical record but in the very language that historians use to express their ideas. Does this mean that objective historical accounts are an impossibility, and if so, what is the student to make of a philosophy that questions history’s very existence as a subject?

The earlier chapters of this book were essentially descriptive. They were intended to show how historians go about their work – their guiding assumptions, their handling of the evidence and their presentation of conclusions. The point has now been reached where some fundamental questions about the nature of historical enquiry can be posed: how securely based is our knowledge of the past? Can the facts of history be taken as given? What authority should be attached to attempts at historical explanation? Can historians be objective? Answers to these questions have taken widely divergent forms and have occasioned intense debate, much of it fuelled by criticisms from outside the ranks of historians. The profession is deeply divided about the status of its findings. At one extreme there are those such as G.R. Elton who maintained that humility in the face of the evidence and training in the technicalities of research have steadily enlarged the stock of certain historical knowledge; notwithstanding the arguments which the professionals take such delight in, history is a cumulative discipline.1 At the other extreme, Theodore Zeldin holds that all he (or any historian) can offer his readers is his personal vision of the past, and the materials out of which they in turn can fashion a personal vision that corresponds to their own aspirations and sympathies: ‘everyone has the right to find his own perspective’.2 Although the weight of opinion among academic historians inclines towards Elton’s position, every viewpoint between the two extremes finds adherents within the profession. Historians are in a state of confusion about what exactly they are up to – a confusion not usually apparent in the confident manner with which they often pronounce on major problems of interpretation.

History publishing is a huge business, with thousands of new titles appearing every year. Does this mean that we are closer to the truth about the past, or does it just mean that there are as many histories as there are people prepared to write them? (© James Leynse/CORBIS)


Is history a science?

To ask such questions about history or any other branch of learning is to enter the terrain of philosophy, since what is at issue is the nature of knowledge itself; and the status of historical knowledge has been hotly contested among philosophers since the Renaissance. Most working historians – even those disposed to reflect on the nature of their craft – take little account of these debates, believing with some justification that they often obscure rather than clarify the issues.3 But the intense disagreement that divides historians reflects a tradition of keen debate among philosophers. During the nineteenth century two sharply opposed positions crystallized around the question of whether history was a science; as recently as the 1960s, when E.H. Carr created such a stir, this was still the key epistemological issue in history. In our own day the ground of debate has shifted to the nature of language and the extent of its bearing on the real world, past and present. Both these debates – the scientific and the linguistic – will now be examined in turn.


Relating to the theory of knowledge, how we know things.

The central question in the debate about history and science has always been whether humankind should be studied in the same way as other natural phenomena. Those who answer this question in the affirmative are committed to the methodological unity of all forms of disciplined enquiry into the human and natural order. They argue that history employs the same procedures as the natural sciences and that its findings should be judged by scientific standards. They may differ as to how far history has in fact fulfilled these requirements, but they are agreed that historical knowledge is valid only in so far as it conforms to scientific method. During the twentieth century conceptions of the nature of science have been radically modified, but the nineteenth-century view was straightforward enough. The basis of all scientific knowledge was the meticulous observation of reality by the disinterested, ‘passive’ observer, and the outcome of repeated observations of the same phenomenon was a generalization or ‘law’ that fitted all the known facts and explained the regularity observed. The assumption of this, the ‘inductive’ or ‘empirical’ method, was that generalizations flowed logically from the data, and that scientists approached their task without preconceptions and without moral involvement.


Neutral, objective. Not to be confused with ‘uninterested’.

Positivism: induction from facts

As a result of its immense strides in both pure and applied work, science enjoyed unrivalled prestige during the nineteenth century. If its methods unlocked the secrets of the natural world, might they not prove the key to understanding society and culture?Positivismis the name given to the philosophy of knowledge that expresses this approach in its classic, nineteenth-century form. Its implications for the practice of history are clear. The historian’s first duty is to accumulate factual knowledge about the past – facts that are verified by applying critical method to the primary sources; those facts will in turn determine how the past should be explained or interpreted. In this process the beliefs and values of historians are irrelevant; their sole concern is with the facts and the generalizations to which they logically lead. Auguste Comte, the most influential positivist philosopher of the nineteenth century, believed that historians would in due course uncover the ‘laws’ of historical development. Full-blown professions of positivist faith are still made occasionally,4 but nowadays a watered-down version is preferred. Latter-day positivists maintain that the study of history cannot generate its own laws; rather, the essence of historical explanation lies in the correct application of generalizations derived from other disciplines supposedly based on scientific method, such as economics, sociology and psychology.

Idealism: intuition and empathy

The second position, which corresponds to the school of philosophy known as idealism, rejects the fundamental assumption of positivism. According to this view, human events must be carefully distinguished from natural events because the identity between the enquirer and his or her subject matter opens the way to a fuller understanding than anything that the natural scientist can aspire to. Whereas natural events can only be understood from the outside, human events have an essential ‘inside’ dimension composed of the intentions, feelings and mentality of the actors. Once the enquirer strays into this realm the inductive method is of limited use. The reality of past events must instead be apprehended by an imaginative identification with the people of the past, which depends on intuition and empathy – qualities that have no place in the classical view of scientific method. According to idealists, therefore, historical knowledge is inherently subjective, and the truths that it uncovers are more akin to truth in the artist’s sense than the scientist’s. Furthermore, historians are concerned with the individual, unique event. The generalizations of the social sciences are not applicable to the study of the past, nor does history yield any generalizations or laws of its own.

This outlook came naturally to the nineteenth-century proponents of historicism (see Chapter 1) with their demand that every age be understood in its own terms, and their practical emphasis on political narrative made up of the actions and intentions of ‘great men’. Ranke’s fame as the champion of rigorous source criticism has sometimes been allowed to obscure the emphasis that he laid on contemplation and imagination: ‘after the labour of criticism’, he insisted, ‘intuition is required’.5 In the English-speaking world the most original and sophisticated exponent of the idealist position has been the philosopher and historian R.G. Collingwood. In his posthumously published The Idea of History (1946), he maintained that all history is essentially the history of thought, and that the historian’s task is to re-enact in his or her own mind the thoughts and intentions of individuals in the past. Collingwood’s influence is evident in the case of present-day opponents of ‘scientific’ theory such as Zeldin, who deplores the tendency for history to become ‘a coffee-house in which to discuss the findings of other disciplines in time perspective’ and pleads for a history concerned with individuals and their emotions.6 Conversely, history’s scientific pretensions tend to be taken much more seriously by historians of collective behaviour – voting or consumption for example – because in these spheres regularities are evident that can sometimes form the basis of firm and significant generalizations.

But the implications of the unresolved clash between positivism and idealism go much further than the distinction between traditional political history and the more recent fields of economic and social history. They help to explain why there is so much disagreement among historians about the nature of virtually every aspect of their work from primary source evaluation through to the finished work of interpretation.


An incomplete and tainted record

Much of the professional self-esteem of the new breed of academic historians in the nineteenth century was based on the rigorous techniques that they had perfected for the location and criticism of primary sources. The canons they established have governed the practice of historians ever since, so that the whole edifice of modern historical knowledge is founded on the painstaking evaluation of original documents. But the injunction ‘Be true to your sources’ is less straightforward than it looks, and sceptics have seized on a number of problem areas. First, the primary sources available to the historian are an incomplete record, not only because so much has perished by accident or design but in a more fundamental sense because a great deal that happened left no material trace whatever. This is particularly true of mental processes, both conscious and unconscious. No historical character, however prominent and articulate, has ever set down more than a tiny proportion of his or her thoughts and assumptions; and often some of the most influential beliefs are those that are taken for granted and therefore are not discussed in the documents. Second, the sources are tainted by the less than pure intentions of their authors and – more insidiously – by their confinement within the assumptions of men and women in that time and place. ‘The so-called “sources” of history record only such facts as appeared sufficiently interesting to record’;7 or, more polemically, the historical record is forever rigged in favour of the ruling class, which at all times has created the vast majority of the surviving sources. In some Marxist circles this contention has led to an absolute scepticism about the possibility of knowledge of the past, and history has been put on the intellectual scrap-heap.

There is an element of truth in both these criticisms, but those who push them to extremes betray an ignorance of how historians actually work. What a researcher can learn from a set of documents is not confined to its explicit meaning; that meaning is first of all scrutinized for bias and then used as the basis for inference. When properly applied, the critical method enables the historian to make allowances for both deliberate distortion and the unthinking reflexes of the writer – to extract meaning ‘against the grain of the documentation’, in Raphael Samuel’s useful phrase.8 Much of the criticism directed against historical method rests on the common misconception that primary sources are the testimonies of witnesses – who like all witnesses are fallible but in this instance are not available for cross-examination. Yet, as was shown in Chapter 5, a great deal of the historian’s documentation is made up of record sources which themselves constitute the event or process under investigation: historians interested, say, in the character of Gladstone or the administrative machinery of the medieval Chancery are not dependent on contemporary reports and impressions (interesting though these may be); they can base their accounts on the private correspondence and diaries of Gladstone himself, or on the records generated in the course of the Chancery’s day-to-day business. Moreover, much of the importance attached to primary sources derives not from the intentions of the writer but from information that was incidental to his or her purpose and yet may provide a flash of insight into an otherwise inaccessible aspect of the past. The historian, in short, is not confined by the categories of thought in which the documents were composed.9

A surfeit of records

But there is a third and more formidable difficulty in the notion that historians simply follow where the documents lead, and this turns on the profusion of the available sources. These sources may, it is true, represent a very incomplete record; yet for all but very remote periods and places they survive in completely unmanageable quantities. This is a problem that has been confronted only during the present century. Nineteenth-century historians, especially those of a positivist turn of mind such as Lord Acton, believed that finality in historical writing would be attained when primary research had brought to light a complete assemblage of the facts; many of these facts might seem obscure and trivial, but they would all tell in the end. These writers were blinded to the limitations of their method by the very narrow way in which they conceived both the content of history and a primary source: when Acton at the end of the nineteenth century wrote, ‘nearly all the evidence that will ever appear is accessible now’,10 he was referring only to the great collections of state records. Since Acton’s day the subject matter of history has been vastly enlarged, and the significance of whole bodies of source material whose existence nineteenth-century historians were scarcely aware of has been established. Faced with the virtually limitless content that history could in theory embrace, modern historians have been compelled to subject the notion of historical ‘fact’ to severe scrutiny.

What are facts?

Objection is sometimes made to the idea of ‘facts’ in history on the grounds that they rest on inadequate standards of proof: most of what pass for the ‘facts’ of history actually depend on inference. Historians read between the lines, or they work out what really happened from several contradictory indications, or they may do no more than establish that the writer was probably telling the truth. But in none of these cases can the historian observe the facts in the way that a physicist can. Historians generally have little time for this kind of critique. Formal proof may be beyond their reach; what matters is the validity of the inferences. In practice historians spend a good deal of time disputing and refining the inferences that can be legitimately drawn from the sources, and the facts of history can be said to rest on inferences whose validity is widely accepted by expert opinion. Who, they ask with some justice, could reasonably ask for more?

Historians are much more troubled by the implications of the apparently limitless number of facts about the past that can be verified in this way. If the entire past of humankind falls within the historian’s scope, then every fact about that past may be said to have some claim on our attention. But historians do not proceed on this assumption – not even the specialist in some limited aspect of a well-defined period. There is in practice no limit to the number of facts that have a bearing on such a problem, and the historian who resolved to be guided solely by the facts would never reach any conclusion. The common-sense idea (and the central tenet of positivism) that historians efface themselves in front of the facts ‘out there’ is therefore an illusion. The facts are not given, they are selected. Despite appearances, they are never left to speak for themselves. However detailed a historical narrative may be, and however committed its author to the re-creation of the past, it never springs from the sources ready-made; many events are omitted as trivial, and those that do find a place in the narrative tend to be seen through the eyes of one particular participant or a small group. Analytical history, in which the writer’s intention is to abstract the factors with greatest explanatory power, is more obviously selective. Historical writing of all kinds is determined as much by what it leaves out as by what it puts in. That is why it makes sense to distinguish with E.H. Carr between the facts of the past and the facts of history. The former are limitless and in their entirety unknowable; the latter represent a selection made by successive historians for the purpose of historical reconstruction and explanation: ‘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’.11

The selection and rejection of facts

If historical facts are selected, it is important to identify the criteria employed in selecting them. Are there commonly shared principles, or is it a matter of personal whim? One answer, much favoured since Ranke’s day, is that historians are concerned to reveal the essence of the events under consideration. Namier expressed this idea metaphorically:

The function of the historian is akin to that of the painter and not of the photographic camera; to discover and set forth, to single out and stress that which is of the nature of the thing, and not to reproduce indiscriminately all that meets the eye.12

But this amounts to little more than a restatement of the original question, for how is the ‘nature of the thing’ to be determined? It makes for less confusion if it is admitted outright that the standards of significance applied by the historian are defined by the nature of the historical problem that he or she is seeking to solve. As M.M. Postan put it:

The facts of history, even those which in historical parlance figure as ‘hard and fast’, are no more than relevances: facets of past phenomena which happen to relate to the preoccupations of historical inquirers at the time of their inquiries.

As new historical facts are accepted into the canon, so old ones pass out of currency except, as Postan mischievously remarks, in textbooks that are full of ‘ex-facts’.13


Features, characteristics.

There is an element of rhetorical exaggeration about this view. Historical knowledge abounds in facts such as the Great Fire of London or the execution of Charles I whose status is for all practical purposes unassailable, and critics such as Elton have seized on this point to discredit the distinction between the facts of the past and the facts of history, which they feel introduces a dangerous element of subjectivity.14 But, as anyone who has sampled the work of professional historians knows, historical writing is never composed entirely, or even principally, of these unassailable facts. The decision whether to include this set of facts rather than that is closely affected by the purpose that informs the historian’s work.

Clearly, then, much depends on the kind of questions that the historian has in mind at the outset of research. As was discussed in Chapter 5, there is something to be said for selecting a rich and previously untapped vein of source material and being guided by whatever questions it throws up (see pp. 120–21). The difficulty with this method is that nobody actually approaches the sources with a completely open mind – the grounding in the standard secondary literature which precedes any research will see to that. Even if no specific questions have been formulated, the researcher will study the sources with certain assumptions that are only too likely to be an unthinking reflection of current orthodoxy, and the result will be merely a clarification of detail or a modification of emphasis within the prevailing framework of interpretation.

Historical hypotheses

Significant advances in historical understanding are more likely to be achieved when a historian puts forward a clearly formulated hypothesis that can be tested against the evidence. The answers may not correspond to the hypothesis, which must then be discarded or modified, but merely to ask new questions has the important effect of alerting historians to unfamiliar aspects of familiar problems and to unsuspected data in well-worked sources. Consider, for example, the origins of the English Civil War. Nineteenth-century historians approached this as a problem of competing political and religious ideologies, and they selected accordingly from the great mass of surviving information about early seventeenth-century England. From the 1930s onwards an increasing number of scholars sought to test a Marxist approach to the conflict, and as a result new material which related to the economic fortunes of the gentry, the aristocracy and the urban bourgeoisie became critically important. More recently several historians have employed a ‘Namierite’ approach in which the constitutional and military conflicts are seen as the expression of rivalry between political factions: hence the networks of patronage and the intrigues at court are now coming more into play.15 The point is not that the Marxist or Namierite position amounts to a rounded explanation of the war but rather that each hypothesis has brought into focus certain previously neglected factors which will have a bearing on any future interpretation. Marc Bloch, whose own work proceeded on the basis of hypotheses, put the issue clearly:

Every historical research supposes that the inquiry has a direction at the very first step. In the beginning, there must be the guiding spirit. Mere passive observation, even supposing such a thing were possible, has never contributed anything productive to any science.16

A new understanding of the nature of science

Significantly, scientists today would themselves mostly agree. The positivist theory still dominates the lay person’s view of science, but it no longer carries much conviction among the scientific community. Inductive thought and passive observation have ceased to be regarded as the hallmarks of scientific method. Rather, all observation whether of the natural or the human world is selective and therefore presupposes a hypothesis or theory, however incoherent it may be. In Karl Popper’s influential view, scientific knowledge consists not of laws but of the best available hypotheses; it is provisional rather than certain knowledge. Our understanding advances through the formulation of new hypotheses that go beyond the evidence currently available and must be tested against further observation, which will either refute or corroborate the hypothesis. And because hypotheses go beyond the evidence, they necessarily involve a flash of insight or an imaginative leap, often the bolder the better. Scientific method, then, is a dialogue between hypothesis and attempted refutation, or between creative and critical thought.17 To historians this is a much more congenial definition of science than the one it has replaced.

Karl Popper (1902–94)

British scientist and philosopher. Popper rejected induction as a basis for science and argued that the proper role of scientific observation was to refute existing theories rather than to try to confirm them.

The importance of imagination

But although history and the natural sciences may converge in some of their fundamental methodological assumptions, important differences remain. First, far greater play is allowed to the imagination in history. It is by no means confined to the formulation of hypotheses but permeates the historian’s thinking. Historians are not, after all, only concerned to explain the past; they also seek to reconstruct or re-create it – to show how life was experienced as well as how it may be understood – and this requires an imaginative engagement with the mentality and atmosphere of the past. As Joseph C. Miller puts it:

History turns data into evidence not by pursuing the technical attributes of data but by substituting a distinctively intuitive, humanistic, holistic strategy for the experimental method of science.18

In maintaining that all history is the history of thought, Collingwood unduly confined the scope of the subject. But it is certainly true that the evaluation of documentary sources depends on a reconstruction of the thought behind them; before anything else can be achieved, the historian must first try to enter the mental world of those who created the sources.

Furthermore, although idealists from Ranke to Collingwood have placed an exaggerated emphasis on ‘unique’ events, individuals are certainly a legitimate and necessary object of historical study, and the variety and unpredictability of individual behaviour (as opposed to the regularities of mass behaviour) demand qualities of empathy and intuition in the enquirer as well as logical and critical skills. And whereas scientists can often create their own data by experiment, historians are time and again confronted by gaps in the evidence which they can make good only by developing a sensitivity as to what might have happened, derived from an imagined picture that has taken shape in the course of becoming immersed in the surviving documentation. In all these ways imagination is vital to the historian. It not only generates fruitful hypotheses; it is also deployed in the reconstruction of past events and situations by which those hypotheses are tested.

The impossibility of consensus

The second and even more critical distinction to be made between history and the natural sciences is that the standing of explanations put forward by historians is very much inferior to that of scientific explanation. It may be that scientific explanations are no more than provisional hypotheses, but they are for the most part hypotheses on which all people qualified to judge are in agreement; they may be superseded one day, but for the time being they represent the nearest possible approximation to the truth and are commonly recognized as such. In matters of historical explanation, on the other hand, a scholarly consensus scarcely exists. The known facts may not be in doubt, but how to interpret or explain them is a matter of endless debate, as my example of the English Civil War illustrated. The ‘faction hypothesis’ has not superseded the ‘class-conflict hypothesis’ or the ‘ideology hypothesis’; all are very much alive and receive varying emphasis from different historians.

The reason for this diversity of opinion lies in the complex texture of historical change. We saw in Chapter 6 how both individual and collective behaviour are influenced by an immense range of contrasting factors. What needs stressing here is that each historical situation is unique in the sense that the exact configuration of causal factors is unrepeatable. It might be argued, for instance, that the reasons why the European powers withdrew from most of their African colonies during the 1950s and 1960s were common to some thirty-odd different territories. But this would be valid only as a very broad-brush statement. The respective strength of the colonial power and the nationalist movement varied from one country to another according to its value to the metropolis, its experience of social change, the size of the resident European community, and so on.19 In practice, therefore, each situation has to be investigated afresh, with the strong possibility of different findings, and as a result the basis for a comprehensive theory of historical causation simply does not exist.

A multiplicity of hypotheses

Perhaps this would not matter if certainty was attainable in explaining particular events. But this more modest objective eludes historians as well. The problem here is that the evidence is never sufficiently full and unambiguous to place a causal interpretation beyond doubt. This is true of even the best-documented events. In a case like the origins of the First World War, the sources provide ample evidence of the motives of the protagonists, the sequence of diplomatic moves, the state of public opinion, the upward spiral of the arms race, the relative economic strength of all the nations involved, and so on. But what the evidence alone cannot do is tell us the relative importance of all these varied factors, or present a comprehensive picture of how they interacted with each other.20 In many instances the sources do not directly address the central issues of historical explanation at all. Some of the influences on human conduct, such as the natural environment or the neurotic and irrational, are apprehended subconsciously; others may be experienced directly but not disclosed in the sources. Questions of historical explanation cannot, therefore, be resolved solely by reference to the evidence. Historians are also guided by their intuitive sense of what was possible in a given historical context, by their reading of human nature, and by the claims of intellectual coherence. In each of these areas they are unlikely to concur. As a result, several different hypotheses can hold the field at any one time. Burckhardt frankly acknowledged the problem in the Preface to hisCivilization of the Renaissance in Italy (1860):

In the wide ocean upon which we venture, the possible ways and directions are many; and the same studies which have served for this work might easily, in other hands, not only receive a wholly different treatment and application, but lead also to essentially different conclusions.21

The area of knowledge beyond dispute is both smaller and much less significant in history than it is in the natural sciences. This is a crucial limitation which is not properly confronted by present-day champions of ‘objectivity’ in history.22

Burckhardt (1818–97)

Jakob Burckhardt, Swiss historian. He is credited with having coined the term ‘Renaissance’ (French: ‘rebirth’) to describe the cultural changes and revival of classical form in fifteenth-century Italy.


The historian as selector

This comparison between history and natural sciences is perhaps somewhat contrived, given that the assumptions most people make about the standing of scientific knowledge are an outdated residue of nineteenth-century positivism; scientific knowledge is in reality less certain and less objective than is commonly supposed. But what the comparison does bring out is the extent to which our knowledge of the past depends on choices freely exercised by the historian. The common-sense notion that the business of historians is simply to uncover the past and display what they have found will not stand up. The essence of historical enquiry is selection – of ‘relevant’ sources, of ‘historical’ facts and of ‘significant’ interpretations. At every stage both the direction and the destination of the enquiry are determined as much by the enquirer as by the data. Clearly, the rigid segregation of fact and value demanded by the positivists is unworkable in history. In this sense, historical knowledge is not, and cannot be, ‘objective’ (that is,empirically derived in its entirety from the object of the enquiry). This does not mean, as sceptics might suppose, that it is therefore arbitrary or illusory. But it does follow that the assumptions and attitudes of historians themselves have to be carefully assessed before we can come to any conclusion about the real status of historical knowledge.


Reasoning from experiment and experience, rather than from theoretical principles. Although strict scientific experimentation is a form of empiricism, so too is deduction based on ill-defined ‘common sense’, which can lend empiricism an ambiguous intellectual status.

The historian in context

Up to a point those standards can be seen as the property of the individual historian. The experience of research is a personal and often very private one, and no two historians will share the same imaginative response to their material. As Richard Cobb put it, ‘the writing of history is one of the fullest and most rewarding expressions of an individual personality’.23 But however rarefied the atmosphere that historians breathe, they are, like everyone else, affected by the assumptions and values of their own society. It is more illuminating to see historical interpretation as moulded by social rather than individual experience. And because social values change, it follows that historical interpretation is subject to constant revision. What one age finds worthy of note in the past may well be different from what previous ages found worthy. This principle can be illustrated many times over within the relatively short span of time since the emergence of the academic profession of history. For Ranke and his contemporaries the sovereign nation-states which dominated the Europe of their day seemed the climax of the historical process; the state was the principal agent of historical change, and human destiny was largely determined by the shifting balance of power between states. This world view was seriously eroded by the First World War: after 1919, against the background of optimism engendered by the League of Nations, history teaching in Britain tended to stress rather the growth of internationalism over the centuries.

League of Nations

The international organization set up at the end of the First World War to settle international disputes without recourse to war. It inspired enormous levels of optimism, especially in Britain, in its early years.

More recently, the way in which historians study the world beyond Europe and the United States has been transformed in the light of the changes they have lived through. Fifty years ago the history of Africa was still treated as an aspect of the expansion of Europe, in which the indigenous peoples scarcely featured except as the object of white policies and attitudes. Today the perspective is very different. African history exists in its own right, embracing both the pre-colonial past and the African experience of – and response to – colonial rule, and stressing the continuities of African historical development, which had previously been completely obscured by the stress on the European occupation. And those continuities have already been reassessed: whereas in the 1960s historians of Africa were mainly concerned with placing African nationalism in a historical perspective of pre-colonial state formation and resistance to colonial rule, they are now, after forty years’ disillusionment with the fruits of independence, preoccupied with the historical antecedents of Africa’s deepening poverty. Twice in the course of a single lifetime the standards of significance applied by historians to the African past have been substantially revised.

However, to say that history is rewritten by each generation (or decade) is only part of the truth – and positively misleading if it suggests the replacement of one consensus by another. In the case of history written during the High Middle Ages or the Renaissance it might be appropriate to speak of a scholarly consensus, since historians and their audience were drawn from a very restricted sector of society, and at this distance in time the differences between historians seem much less significant than the values they held in common. But the attainment of universal literacy and the extension of education in Western society in the twentieth century mean that historical writing now reflects a much wider range of values and assumptions. The towering political personalities of the past such as Oliver Cromwell or Napoleon Bonaparte are interpreted in widely divergent ways by professional historians as well as lay people, partly according to their own political values.24 Liberal or conservative historians such as Peter Laslett tend to conceive of social relations in pre-industrial England as reciprocal, while radically inclined historians such as E.P. Thompson see them as exploitative.25 Michael Howard has made public confession of a bias that is widely shared – a bias in favour of a liberal political order in which alone the historian has been permitted to work without censorship.26 Many other historians, however, would set a higher value on material progress or equality in social relations than on freedom of thought and expression. Historical interpretation is a matter of value judgements, moulded to a greater or lesser degree by moral and political attitudes. At the beginning of the twentieth century Acton’s successor at Cambridge, J.B. Bury, looked forward to the dawn of scientific history with these words: ‘Though there be many schools of political philosophy, there will no longer be divers schools of history’.27 It would be nearer the truth to say that for as long as there are many schools of political philosophy there will be divers schools of history. Paradoxically there is an element of present-mindedness about all historical enquiry.

Peter Laslett (1915–2001)

British historian. He pioneered the study of the history of the English family. His ground-breaking work of social history The World We Have Lost (1965) overturned many common assumptions about everyday life in early modern England.

Sir Michael Howard (1922–)

British military historian, Regius Professor of Modern History at Oxford 1980–9.

The search for origins

The problem, of course, is to determine at what point presentmindedness conflicts with the historian’s aspiration to be true to the past. The conflict is clearest in the case of those writers who ransack the past for material to fuel a particular ideology, or who falsify it in support of a political programme, as Nazi historians did under the Third Reich and supporters of Holocaust denial do today. Such works are propaganda, not history, and it is usually clear to the professional – and sometimes the lay person – that evidence has been suppressed or manufactured. Among historians themselves present-mindedness commonly takes two forms. The first is an interest in the historical origins of the modern world, or some particularly salient feature of it – say the nuclear family household or parliamentary democracy. In itself this is a positive response to the claims of social relevance, and it has the merit of providing a clear principle of selection leading to an intelligible picture of the past. But it also carries risks of superficiality and distortion. The problem with seeking the historical antecedents of some characteristically ‘modern’ feature is that the outcome can so easily seem to be predetermined, instead of being the result of complex historical processes. Abstracting one strand of development to be traced back to its origins too often means an indifference to historical context; the further back the enquiry proceeds, the more likely will a stress on linear descent obscure the contemporary significance of the institution or convention in question. Thus the Whig historians of the nineteenth century completely misunderstood the structure of medieval English government because of their obsessive interest in the origins of Parliament. A comparable criticism has been levelled at recent work on the medieval and early modern history of family relations and sexuality.28 As Butterfield put it in The Whig Interpretation of History (1931) – probably the most influential polemic ever written against present-minded history – ‘the study of the past with one eye, so to speak, upon the present is the source of all sins and sophistries in history, starting with the simplest of them, the anachronism’.29 ‘Whig’ history exhibits a tendency to underestimate the differences between past and present – to project modern ways of thought backwards in time and to discount those aspects of past experience that are alien to modern ideas. In this way it reduces history’s social value, which derives largely from its being a storehouse of past experiences contrasted to our own.

A voice for the oppressed

Today a second variant of present-minded history (or ‘presentism’) is much more prevalent. This is the history written out of political commitment to a social group that has previously been marginalized by the prevalent historiography. As explained in Chapter 1, effective political action in the present requires an articulate social memory, and to supply this has been one of the main objectives of black historians and women’s historians in Britain and the United States. It is said that the purpose of these radical histories is not just to uncover what was previously ‘hidden from history’30 but to demonstrate historical experience of a predetermined kind – in this case oppression and resistance – to the exclusion of material that fits less neatly with the political programme of the writer. Thus the complicity of West African societies in the transatlantic slave trade may be omitted, or the sexual conservatism of much nineteenth-century feminism. When ethnic particularism or gender loyalty provides the decisive impetus for research, the differences between ‘then’ and ‘now’ may be downplayed in the cause of forging an identity across the ages, while no serious effort may be made to understand the experience of other groups with a part in the story. The way is then open for a reactive historiography marked by a more explicit and hard-nosed defence of the established order than that which existed before.

‘Everyman his own historian’

If the outcome of historical enquiry is so heavily conditioned by the preferences of the enquirer and can so easily be altered by the intervention of another enquirer, how can it merit any credibility as a serious contribution to knowledge? If fact and value areinextricably tied together, how can a distinction be drawn between sound and unsound history? Between the two World Wars it was the fashion in some quarters to concede most, if not all, of the sceptics’ case. Historical interpretation, these historians averred, should be considered true only in relation to the needs of the age in which it was written. With the phrase ‘Everyman his own historian’,31 the American scholar Carl M. Becker renounced the aspirations to definitive history that had characterized the profession since Ranke. More recently the case has been succinctly put by Gordon Connell-Smith and Howell Lloyd:

History is not ‘the past’, nor yet the surviving past. It is a reconstruction of certain parts of the past (from surviving evidence) which in some way have had relevance for the present circumstances of the historian who reconstructed them.32

The unattainability of the past

The implications of this position are disturbing. Not surprisingly historians are reluctant to allow their discipline’s claim to academic respectability to be so lightly abandoned. Over the past forty years the orthodox response to relativism has been to make what is essentially a restatement of historicism. Historians, the argument goes, must renounce any standards or priorities external to the age they are studying. Their aim is to understand the past in its own terms, or in Elton’s words ‘to understand a given problem from the inside’.33 Historians should be steeped in the values of the age and should attempt to see events from the standpoint of those who participated in them. Only then will they be true to their material and their vocation. But this claim to speak with the voice of the past will not bear inspection. On the face of it, historians may appear to be strikingly successful in assimilating the values of those they write about: diplomatic historians usually accept the ethics of raison d’état which have governed the conduct of international relations in Europe since the Renaissance, and the historian of a political movement may well be able to achieve an empathy with the outlook and aspirations of its members. However, as soon as historians cast their net more widely to embrace an entire society, ‘the standards of the age’ becomes a question-begging phrase. Whose standards should be adopted – those of the rich or the poor, the colonized or the colonizers, Protestant or Catholic? It is a fallacy to suppose that historians who renounce all claim to ‘relevance’ thereby ensure the objectivity of their work. In practice their writing is exposed to two dangers. On the one hand they may find themselves confined by the priorities and assumptions of those who created the sources; on the other, the end-product is quite likely to be influenced – if only unconsciously – by their own values, which are difficult to make allowances for because they are undeclared. Elton’s work illustrates both these tendencies: his Tudor England is seen through the spectacles of the authoritarian paternalist bureaucracy whose records Elton knew so intimately and whose outlook was evidently congenial to his own conservative convictions.34 Re-creative history is a legitimate pursuit, but it is a mistake to suppose that it can ever be completely realized, or that it carries the promise of objective knowledge about the past.

G.R. Elton (1921–94)

Sir Geoffrey Elton first made his name with a detailed study, based on his Ph.D. thesis, of what he called The Tudor Revolution in Government. He held that Thomas Cromwell had instituted such a strikingly modern system of bureaucracy at Henry VIII’s court that it amounted in effect to an administrative revolution. However, Elton was sometimes accused of seeing everything in Tudor England as if it related to bureaucratic administration.

History and hindsight

There is another serious difficulty encountered by the strictly historicist approach. We can never recapture the authentic flavour of a historical moment as it was experienced by people at the time because we, unlike them, know what happened next; and the significance which we accord to a particular incident is inescapably conditioned by that knowledge. This is one of the most telling objections that can be made against Collingwood’s idea that historians re-enact the thought of individuals in the past. Like it or not, the historian approaches the past with a superior vision conferred by hindsight. Some historians do their best to renounce this superior vision by confining their research to a few years or even months of history, for which they can give a blow-by-blow account with a minimum of selection or interpretation, but the total divestment of hindsight is not intellectually possible. Besides, should not hindsight be viewed as an asset to be exploited rather than a disability to be overcome? It is precisely our position in time relative to the subject of our enquiry that enables us to make sense of the past – to identify conditioning factors of which the historical participants were unaware, and to see consequences for what they were rather than what they were intended to be. Strictly interpreted, ‘history for its own sake’ would entail surrendering most of what makes the subject worth pursuing at all, without achieving the desired goal of complete detachment. The problems of historical objectivity cannot be evaded by a retreat into the past for the past’s sake.


The challenge of Postmodernism

So far this evaluation of historical enquiry has implied a hierarchy of approaches in which positivist science stands as the ultimate yardstick of intellectual rigour. Scientific method is here viewed as the only means of gaining direct knowledge of reality, past or present. The procedures of historicism offer a scarcely tenable defence, and to the extent that they fall short of scientific method must be deemed inferior. This debate has been running for as long as history has been seriously studied, and it shows no sign of being resolved. However, in the past three decades the hand of the sceptics has been strengthened by a major intellectual shift within the humanities that has rejected historicism as the basis for history and all other text-based disciplines. This is Postmodernism. Its hallmark is the prioritization of language over experience, leading to outright scepticism as to the human capacity to observe and interpret the external world, and especially the human world. The implications of Postmodernism for the standing of historical work are potentially serious and must be addressed with some care.

The tyranny of language

Modern theories of language stand in a tradition first laid out by Ferdinand de Saussure at the beginning of the twentieth century. Saussure declared that, far from being a neutral and passive medium of expression, language is governed by its own internal structure. The relationship between a word and the object or idea it denotes – or between ‘signifier’ and ‘signified’ in Saussure’s terminology – is in the last resort arbitrary. No two languages have an identical match between words and things; certain patterns of thought or observation that are possible in one language are beyond the resources of another. From this Saussure drew the conclusion that language is non-referential – that speech and writing should be understood as a linguistic structure governed by its own laws, not as a reflection of reality: language is not a window on the world but a structure that determines our perception of the world. This way of understanding language has the immediate effect of downgrading the status of the writer: if the structure of the language is so constraining, the meaning of a text will have as much to do with the formal properties of the language as with the intentions of the writer, and perhaps more. Any notion that writers can accurately convey ‘their’ meaning to their readers falls to the ground. In a much-quoted phrase, Roland Barthes spoke of ‘the death of the author’.35 One might equally speak of the death of the textual critic in the traditional sense, since those who interpret texts have as little autonomy as those who wrote them. There can be no objective historical method standing outside the text, only an interpretative point of address fashioned from the linguistic resources available to the interpreter. The historian (or literary critic) does not speak from a privileged vantage point.

However, it is simplistic to speak of the ‘language’ of any society in the singular, if by this we mean to suggest a common structure and uniform conventions. Any language is a complex system of meanings – a multiple code in which words often signify different meanings to different audiences; indeed the power of language partly resides in the unintended layers of meaning it conveys. The kind of textual analysis in which the immediate or ‘surface’ meaning is set aside in favour of the less obvious is called in Postmodern circles ‘deconstruction’ – a term coined by Jacques Derrida. Deconstruction covers a bewildering mass of daring and dissonant readings. If Saussure’s severance of signifier from signified is treated as an absolute principle, there is after all no limit to the range of permitted readings. The creative approach to interpreting texts – playful, ironic and subversive by turns – is a hallmark of Postmodern scholarship.36


Also known as constructionists, the literary forebears of historical Postmodernists. Inspired by the French literary scholar Jacques Derrida, they stressed the importance of analysing not just the wording of a text but the hidden assumptions and social or moral values within its vocabulary, even questioning whether text actually denotes what its words theoretically mean.

Intertextuality: text and context

For most exponents of the linguistic turn, however, some limit is placed on the freedom with which we can ‘read’ texts by the constraints of ‘intertextuality’. According to this perspective, the texts of the past should not be viewed in isolation, because no text has ever been composed in isolation. All writers employ a language that has already served purposes similar to their own, and their audience may interpret what they write with reference to yet other conventions of language use. At any given time the world of texts is composed of diverse forms of production, each with its own cultural rationale, conceptual categories and patterns of usage. Each text belongs, in short, to a ‘discourse’ or body of language practice. Today the term ‘discourse’ is best known in the distinctive twist given to it by French philosopher Michel Foucault. For him ‘discourse’ meant not just a pattern of language use but a form of ‘power/knowledge’, pointing to the way in which people are confined within the regulatory scope of specific discourses. He showed how new, more restrictive discourses of madness, punishment and sexuality became established in Western Europe between 1750 and 1850, challenging the conventional interpretation of this period as one of social and intellectual progress.37 Foucault was unusual among the founding fathers of Postmodernism in conveying a strong sense of period. But as used by most literary scholars, ‘discourse’ and ‘intertextuality’ have a tendency to float free of any anchorage in the ‘real’ world, thus bearing out Derrida’s celebrated aphorism, ‘there is nothing outside the text’.38

Michel Foucault (1926–84)

French philosopher and social historian. Foucault’s studies of restrictive or oppressive institutions, such as nineteenth-century hospitals, prisons and mental asylums, have led to a new understanding of the power relationship between the individual and the state.

Relativism: nothing is certain

Analysing discourse, like all the critical procedures associated with modern linguistics, is founded on relativism. Its champions dismiss the idea that language reflects reality as the representational fallacy. Language, they assert, is inherently unstable, variable in its meanings over time, and contested in its own time. If accepted at face value, that indeterminacy is fatal to traditional notions of historical enquiry. It becomes meaningless to attempt a distinction between the events of the past and the discourse in which they are represented; as Raphael Samuel put it in a neat summary of Roland Barthes, history becomes ‘a parade of signifiers masquerading as a collection of facts’.39 As we saw in Chapter 5, historians certainly do not regard their primary sources as infallible, and they are accustomed to reading them against the grain for implicit meanings. But underlying their scholarly practice is the belief that the sources can yield up some, at least, of the meaning they held for those who wrote and read them originally. That isanathema to the deconstructionist, for whom no amount of technical expertise can remove the subjectivity and indeterminacy inherent in the reading of texts. Deconstructionists offer us instead the pleasure of finding any meanings we like, provided we do not claim authority for any of them. No amount of scholarship can give us a privileged vantage point. All that is available to us is a free interaction between reader and text, in which there are no approved procedures and no court of appeal. To claim any more is naïvety or – in the moreintemperate Postmodernist statements – a deception practised on the innocent reader.


Completely unacceptable. The term comes from the Roman Catholic Church, where it is used to denote ideas and beliefs that are entirely incompatible with Catholic doctrine.

The negation of history

Because historians claim vastly more than this, every aspect of their practice is open to challenge by Postmodernism. Once the validity of the historical method of interpreting texts is undermined, all the procedures erected on that foundation are called into question. The Rankean project of re-creating the past collapses, because it depends on a privileged, ‘authentic’ reading of the primary sources. In place of historical explanation, Postmodernist history can only offer intertextuality, which deals in discursive relations between texts, not causal relations between events; historical explanation is dismissed as no more than a chimera to comfort those who cannot face a world without meaning.40 The conventional actors of history fare no better. If the author is dead, so too is the unified historical subject, whether conceived of as an individual or as a collectivity (such as class or nation): according to the Postmodernist view, identity is constructed by language – fractured and unstable because it is the focus of competing discourses. Perhaps most important of all, deconstructing the individuals and groups who have been the traditional actors in history means that history no longer has a big story to tell. The nation, the working class, even the idea of progress, all dissolve into discursive constructions. Continuity and evolution are rejected in favour of discontinuity, as for example in Foucault’s conception of four unconnected historical epochs (or ‘epistemes’) since the sixteenth century.41 Postmodernists are generally scathing about the ‘grand narratives’ or ‘metanarratives’ of historians – such as the rise of capitalism or the growth of free thought and toleration. The most they will concede is that the past can be arranged into a multiplicity of stories, just as individual texts are open to a plurality of readings.

A reappraisal as radical as this has major implications for how we understand the activity of being a historian. Postmodernists have brought two important perspectives to bear on this. First, they emphasize that historical writing is a form of literary production which, like any other genre, operates within certain rhetorical conventions. In his very influential Metahistory (1973), Hayden White analyses these conventions in aesthetic terms and classifies historical writing according to twelve stylistic permutations and four underlying ‘tropes’. The specifics of this elaborate analysis are less important than White’s theoretical conclusion, that the character of any work of history is determined not so much by the author’s scholarship or ideology as by the aesthetic choices that he or she makes (usually unconsciously) at the outset of the enquiry and that inform the discursive strategies of the text. With its privileging of the aesthetic over the ideological, this is a somewhat purist position. Postmodernism is currently more strongly identified with a second perspective, in which the historian is seen as the vector of a range of political positions rooted in the here and now. Because the documentary residue of the past is open to so many readings, and because historians employ language that is ideologically tainted, history writing is never innocent. There being no shape to history, historians cannot reconstruct and delineate it from outside. The stories they tell, and the human subjects they write about, are merely subjective preference, drawn from an infinity of possible strategies. Historians are embedded in the messy reality they seek to represent, and hence always bear its ideological imprint. They may do no more than replicate the dominant or ‘hegemonic’ ideology; alternatively, they may identify with one of a number of radical or subversive ideologies; but all are equally rooted in the politics of today.


A metaphor or figure of speech.


Artistic or relating to art or beauty.


Dominant, exercising power over a region or domain.

From this angle all versions of history are ‘presentist’, not just the politically committed ones. In Keith Jenkins’s phrase, history becomes ‘a discursive practice that enables present-minded people(s) to go to the past, there to delve around and reorganise it appropriately to their needs’.42 Since those needs are diverse, and even mutually exclusive, there can be no community of historians and no dialogue between those who hold to different perspectives. Forty years ago, E.H. Carr represented the limits of scepticism in the historical profession when he acknowledged the dialogue between present and past that animates any work of history. Postmodernists take a big step closer to relativism by accepting – even celebrating – a plurality of concurrent interpretations, all equally valid (or invalid). ‘One must face the fact’, writes Hayden White, ‘that, when it comes to the historical record, there are no grounds to be found in the record itself for preferring one way of construing its meaning rather than another’.43 Historians, it is said, do not uncover the past; they invent it. And the time-honoured distinction between fact and fiction is blurred.


Postmodernism in context

How should historians respond to this onslaught? One task for which they are well equipped is to place Postmodernism itself in historical context. This means recognizing that it is located in a particular cultural moment. As the name implies, Postmodernism is a reactive phenomenon. ‘Modernism’ denotes the core beliefs that underpinned the evolution of modern industrial societies from the mid-nineteenth to the mid-twentieth century, especially the belief in progress and faith in the efficacy of disciplined, rational enquiry. In throwing them over, Postmodernists signal their desire for the new and for their emancipation from the previous generation. But the appeal of Postmodernism is best explained by its resonance with some of the defining tendencies in contemporary thought. For some time now the view has gained currency that much that the West has traditionally stood for has come to a dead end: its global supremacy is in decline, its technological flair has become a liability (as in the arms race), and its much-vaunted monopoly of reason is held to be irrelevant to an increasing range of human problems, from the understanding of the psyche to the care of the environment. The Holocaust, instead of being treated as an aberration, is now taken to be a grimly ironical commentary on the conventional equation of progress with Western civilization. There is widespread disillusion with the previously uncontested virtues of scientific method. Postmodernism is the theoretical stance that best illustrates these tendencies. By calling into question the possibility of objective enquiry, it undermines the authority of science. By denying shape and purpose to history, it distances us from all that we find hardest to face in our past – as well as that in which we used to take pride. If, as Postmodernism asserts, history really has no meaning, it follows that we must become fully responsible for finding meaning in our own lives, bleak and demanding though the task may be. History as traditionally conceived becomes not only impractical but irrelevant.

The precursors of Postmodernism

This is not the first time that the credentials of history as a serious discipline have been called into question. The emphasis placed by Postmodernists on the indeterminacy of language and the pervading tone of cultural pessimism are very contemporary, but their denial of historical truth has a very familiar ring about it. In the era of religious wars in Europe in the sixteenth and seventeenth centuries, historians were dismissed by philosophers as credulous impostors, and their much-vaunted sources written off as unreliable. The nineteenth-century historicists, despite their more rigorous standards of scholarship, were soon being attacked by relativists who argued that absolute historical truth was a chimera. In fact there have been sceptics for as long as history has been written. Doubts about the status of the ‘real’, and our ability to apprehend it in the past or the present, have been part of the Western philosophical tradition since the ancient Greeks. Historians themselves have participated in these debates. Postmodernism is less of a novelty than its proponents sometimes claim.

History adapts

Nor is the relationship between history and Postmodernism quite so antagonistic as my account so far implies. It may be, as some Postmodernists argue, that the Rankean documentary ideal is finished and that history as we know it is destined for the scrap-heap.44But what this gloomy prognosis overlooks is that historians are already in the process of assimilating aspects of the Postmodernist perspective. As has so often been the case in the past, root-and-branch critiques of the discipline have a tendency to attack astraw man. Historians have always shown a capacity to engage with critics of the truth claims of their discipline and to take on board some of their arguments. They are not nearly so committed to the unified historical subject as some critics have supposed; it is now rare for scholarly writers to structure a book around ‘the nation’ or ‘the working class’ without carefully analysing the changing and contested significance of these labels.45 Equally, many of the ‘grand narratives’ of Western history – such as the Whig interpretation of English history or the Industrial Revolution – have been subjected to much more devastating attack by empirically minded historians than they have by Postmodernists.46

straw man

An old term for an idea or body that is not as strong as it looks.

Historical writing has also been directly influenced by the linguistic turn in the humanities. Recognizing the structural constraints that language may impose on its users has proved a particularly helpful insight. Gareth Stedman Jones proved as much in his reassessment of Chartism in Languages of Class (1983). The failure of the Chartists to sustain a mass campaign for popular democratic rights after the middle-class agenda had been met in the Reform Act of 1832 has been explained in various ways by historians. Stedman Jones concludes that the movement essentially failed because its politics was constituted by a discourse inherited from the past, which was inappropriate to a rapidly changing political landscape. It is a powerful (though not undisputed) case for ‘an analysis of Chartism which assigns some autonomous weight to the language within which it was conceived’.47 Historians are also sympathetic to the notion that texts embody more than one level of meaning, and that the implicit or unconscious meaning may be what gives the text its power. In late nineteenth-century Britain, for example, the popular language of the New Imperialism was obviously about nationalism and racism; but with its stress on ‘manliness’ and ‘character’ it also carried a heavy charge of masculine insecurity, which arose from changes in women’s position in the family and the workplace. When politicians used that language, they both reflected and intensified an uncertain sense of manhood, almost certainly without meaning to.48 Determining the discourse to which a particular text belongs, and its relation to other relevant discourses, is a task that goes beyond the procedures of source criticism as traditionally understood. As a result, historians now tend to be more sensitive to the counter-currents of meaning in their sources, pushing Marc Bloch’s well-known aphorism about ‘witnesses in spite of themselves’ in a new and rewarding direction.

Reform Act

The pioneering measure of parliamentary reform, which was finally passed into law in 1832. Historians point out that its provisions were relatively modest but that its symbolic importance was immense.

Language and cultural hegemony

Equally, the Postmodern critique of historical writing has met with some positive responses among historians. In particular, Hayden White’s dissection of the literary conventions embedded in historical narrative has resulted in a renewed awareness of historical writing as a literary form and a greater readiness to experiment.49 Even more promising, the Postmodern deconstruction of discourse as a form of cultural power has made it harder to ignore the fact that history writing itself can be an expression of cultural hegemony, and this in turn has opened up opportunities for radical contestation by groups previously excluded from the record. Edward Said’s interest in how language is formed and how a subject is constituted has gone hand in hand with his investigation of the Arab and the Palestinian in Western discourse; his path-breaking Orientalism (1976) proved to be a turning point in the emergence of a post-colonial or multicultural history (see Chapter 10). Feminists, in their ambition to penetrate the limitations of ‘man-made language’, have acknowledged a comparable debt to the linguistic turn.50 These instances go some way to support the Postmodernists’ contention that their perspective holds out the prospect of democratic empowerment. When to that is added the pervasive influence of language-led theory on the development of cultural history in recent years (as discussed in Chapter 9), it is clear that the encounter between Postmodernism and more traditional theories of history has been quite fruitful.


The limitations of Postmodernism

However, there is a limit beyond which most historians will not go in embracing Postmodernism. Many welcome a greater sophistication in interpreting texts and a heightened awareness of the cultural significance of historical writing. But few are prepared to join in a rejection of the truth claims of history as usually practised. Confronted by the full force of the deconstructionist critique, historians tend to be confirmed in their preference for experience and observation over first principles. In theory an impeccable case can be made for the proposition that all human language is self-referential rather than representational. But daily life tells us that language works extremely well in many situations where meaning is clearly communicated and correctly inferred. On any other assumption human interaction would break down completely. If language demonstrably serves these practical functions in the present, there is no reason why it should not be understood in a similar spirit when preserved in documents dating from the past. Of course there is anelement of indeterminacy about all language; the lapse of time serves to increase it, and a 300-year-old text straddling two or three discourses may be very difficult to pin down. Historians frequently acknowledge that they cannot fathom all the levels of meaning contained in their documents. But to maintain that no text from the past can be read as an accurate reflection of something outside itself flies in the face of common experience. In a set of trade figures or a census return the relation between text and reality is palpable (which is not to say that it is necessarily accurate). A carefully considered literary production such as an autobiography or a political tract disguised as a sermon presents much more complex problems, but it is still important to recognize that their authors were attempting a real engagement with their readers, and to get as close as we can to the spirit of that engagement.

political tract

A tract is a small booklet, larger than a pamphlet but smaller than a book, which puts across an argued case. Tracts were widely used in the nineteenth century by church and religious reform groups, but there were plenty of political tracts as well.

It is at this point that historians invoke the discipline of historical context. The meanings that link words and things are not arbitrary and infinite but follow conventions created by real culture and real social relations. The task of scholarship is to identify these conventions in their historical specificity and to take full account of them in interpreting the sources. Whereas exponents of the linguistic approach treat ‘context’ as meaning other texts only, with the further complication that they too invite a variety of readings, historians insist that texts should be set in the full context of their time. That means taking seriously not just the resources of the language but the identity and background of the author, the conditions of production of texts, the intended readership, the cultural attitudes of the time, and the social relations that enveloped writer and readers. Every text is socially situated in specific historical conditions; in the useful phrase of Gabrielle Spiegel, there is a ‘social logic of the text’ which is open to demonstration by historical enquiry.51 So, for example, my reading of the language of late nineteenth-century imperialism can be taken seriously because the strains in gender relations at that time are very well documented, and because the cultural identification of empire with masculinity bore some relation to imperial realities. No doubt deconstruction could yield other interpretations, more elegant and intriguing than this; but unless they have a firm anchorage in historical context, they amount to an imposition by the critic on the text. Respect for the historicity of the sources is fundamental to the historical project; the point at which it is breached is where historians part company with the deconstructionists. Historians do not claim that in all cases their method can uncover every dimension of textual meaning; in order for historical work to be done, it is sufficient to demonstrate that some of the original meaning can be reclaimed, so that we can look beyond discourse to the material and social world in which the texts were created. The verification of historical events and the discipline of historical context mean that historians can distinguish between what happened in history and the discourse in which it is represented.

The need for historical explanation

Historians are no more willing to jettison the truth claims of the accounts that they themselves construct. It is one thing to acknowledge the rhetorical aspects of historical writing but quite another to treat it as only – or largely – rhetoric. Historical narratives are certainly moulded by the historian’s aesthetic sense, but they are not inventions: some, like the major revolutionary upheavals, arise partly from the consciousness of those who lived through them; others fall into shape through the benefit of historical hindsight. The stories we tell ourselves about the past may not be completely coherent or completely convincing, but they are rooted in the fact that human beings not only believe them but enact them on the assumption that social action is a continuum through past, present and future. The task of historical explanation is similarly one that cannot be shirked. It represents not an escape from the real world, as the bleaker versions of Postmodernism insist, but an essential application of reason, based on patterns of cause and consequence which go beyond the confined domain of intertextuality. As for the emancipatory potential of competing narratives, this amounts to little if the ambitions of each identity group are confined to producing a history that is ‘true’ only for its own members. Real empowerment comes from writing history that carries conviction beyond one’s own community, and this means conforming to the scholarly procedures that historians of all communities respect. That, rather than the consolation prize of a permissive relativism, has been the objective of most ‘multicultural’ historians. Despite the pessimism of some conservative commentators,52 pluralism does not necessarily mean relativism.




The idea that all codes of values or ethics are equivalent and exist in relation to their context; it is therefore not possible to say that any one of them is in any sense ‘better’ than any other.

The nub of the Postmodernist critique is that historicism is dead and should be abandoned as a serious intellectual endeavour. In fending off this attack, historians point out not only that the weaknesses of historical enquiry have been grossly exaggerated but that a broadly historicist stance towards the past is culturally indispensable. It is a precondition of critical social thought about the present and the future. As Joyce Appleby, Lynn Hunt and Margaret Jacob put it, ‘Rejecting all meta-narratives cannot make sense, because narratives and meta-narratives are the kinds of stories that make action in the world possible’.53 A consciousness of the past as ‘other’, a set of coherent narratives linking past and present, and an explanatory mode of historical writing are all practical necessities. If the ambition to know the past is completely surrendered, we shall never be able to determine how the present came to be. The social function of history is not to be so lightly abandoned.


Theoretical objections, practical answers

In questioning the credentials of historical knowledge, Postmodernism has breathed fresh life into a strand of scepticism that stretches back to the Renaissance. The fallibility (or ‘indeterminacy’) of the sources, the gap between validated facts and the explanations that endow them with meaning, and the personal and political investment that historians bring to their work, have long been hostages to fortune. Positivism condemned them as damning departures from scientific rigour; Postmodernism subsumes them in a larger refutation of rational enquiry. Whether viewed from a positivist or a Postmodern standpoint, the epistemological credentials of history do not look impressive. Primarily this is because abstract theories are best tested in carefully controlled conditions, whereas history is a hybrid discipline that defies simple pigeonholing. The divergent and sometimes contradictory objectives that historians pursue are what gives the subject its distinctive character, but they also lay it open to theoretical attack.

Though some historians still seek refuge in an untenable empiricism,54 the more thoughtful defenders of the discipline concede that it is open to major theoretical objections. Commentators such as Appleby, Hunt and Jacob, or Richard J. Evans know that historical knowledge always involves an encounter between present and past in which the present may weigh too heavily on the past. They know that the sources do not ‘speak’ directly, that facts are selected, not given, that historical explanation depends on the application of hindsight, and that every historical account is in some sense moulded by the aesthetic and political preferences of the writer. Their defence rests on the contention that, while in theory these features may invalidate historical work, in practice they can be – and are – confined to manageable proportions. History is neither an exemplar of realism nor a victim to relativism. It occupies a middle ground in which scholarly procedures are upheld in order to keep the avenues of enquiry as close to the ‘real’ and as far removed from the ‘relative’ as possible.55 Historians are members of a profession, one of whose principal functions is to enforce standards of scholarship and to restrain waywardness of interpretation. Peer-group scrutiny operates as a powerful mechanism for ensuring that within the area of enquiry they find significant, historians are as true as they can be to the surviving evidence of the past.

The historian’s safeguards: self-awareness and peer review

Three requirements stand out in this respect. First, the historian should scrutinize his or her own assumptions and values in order to see how they relate to the enquiry in hand. One of the attractions of E.P. Thompson is that he made no secret of his sympathies – even acknowledging that one chapter in The Making of the English Working Class was polemic.56 This kind of awareness is particularly important in the case of those historians who have no particular axe to grind but can all too easily be the unconscious vector of values taken for granted by people of their own background. That is one reason why, as emphasized by Zeldin, self-knowledge is a desirable trait among historians (see p. 168) – and also why the confessional mode of historical writing should be welcomed, at least in the author’s preface or introduction. Second, the risk of assimilating findings to expectations is reduced if the direction imparted to the enquiry is cast in the form of an explicit hypothesis, to be accepted, rejected or modified in the light of the evidence – with the author always the first to try to pick holes in his or her interpretation. The appropriate conduct for historians is not to avoid social relevance but to be fully aware of why they are attracted to their particular slice of history and to show as much respect for contrary as for supporting evidence. It is sometimes forgotten by non-practising critics that much of the excitement of historical research comes from finding results that were not anticipated and pushing one’s thesis into a new direction. Third and above all, historians must submit their work to the discipline of historical context. The case against ‘presentism’ and deconstructionism is that they remove events and personalities from their real time and place, forcing them into a conceptual framework that would have meant nothing to the age in question. In fact historians have much less excuse for falling into this trap than they used to. The enlargement of the scope of historical studies during the past fifty years, and the way in which the best historical syntheses reflect this enlargement, means that historians today should have a much better-developed sense of context than their predecessors did; peer review operates particularly effectively in this area.


An angry and impassioned argument.

peer review

Academic work is usually scrutinized in detail by other academics in a process known as peer review.

Respect for these three injunctions does much to limit the amount of distortion in historical writing. It does not, however, put an end to debate and disagreement. It would be wrong to suppose that if all historians could only attain a high degree of self-awareness, make their working hypotheses explicit and maintain a scrupulous respect for historical context they would then concur in their historical judgements. Nobody can become completely dispassionate about his or her own assumptions or those of earlier ages; the evidence can usually be read in support of conflicting hypotheses; and, since the sources never recapture a past situation in its entirety, the sense of historical context depends also on an imaginative flair that will vary according to the insight and experience of the individual scholar. The nature of historical enquiry is such that, however rigorously professional the approach, there will always be a plurality of interpretation. That should be counted as a strength rather than a weakness. For advances in historical knowledge arise as much from the play of debate between rival interpretations as from the efforts of the individual scholar. And the same debates that enliven the historical profession are intimately connected with the alternative visions we hold of our society in the present and the future. If history was uncontested it would fail to provide the materials for critical debate on the social issues of the day. Plurality of historical interpretation is an essential – if underestimated – prerequisite for a mature democratic politics. The past will never be placed beyond controversy; nor should it be.

Inductive reasoning

Historians have to achieve a balance between deductive and inductive reasoning. Deductive reasoning is employed when the conclusion is entirely supported by the information on which it is based, known as the premise. Thus, if the premises are a) that all cats are vertebrates and b) that Toby is a cat, we can safely deduce that Toby is a vertebrate; indeed no other deduction is possible. Deductive reasoning is well illustrated in the methods adopted by the fictional detective Sherlock Holmes, who once remarked that, when all other possibilities have been eliminated, what is left, however unlikely, must be the solution. The point of the italics is that many people allow their expectations, drawn from everyday experience, to influence their interpretation of data. This is inductive reasoning. For example, a visitor to London who saw a host of red buses and black taxis might, understandably but still wrongly, conclude that all London buses are red and all London taxis are black. Mathematicians rely entirely on deductive reasoning and can be impatient of scientists’ tendency to slip into the inductive; for historians, with their often patchy and incomplete evidence, the temptation to rely too much on purely inductive reasoning is even stronger. Every time a historian generalizes from a single incident or example he or she is employing inductive reasoning, which further evidence might well show to be mistaken.

Holocaust denial

In the closing stages of the Second World War, allied armies overran German concentration camps and uncovered evidence of the Nazis’ policy of exterminating Jews, Gypsies and other groups. However, a move also got under way fuelled by extreme right-wing and anti-semitic groups in different countries to deny that the Holocaust had ever happened. A characteristic approach of Holocaust denial is to adopt an outwardly respectable, academic manner and to appear to subject the evidence for the Holocaust to careful and objective scrutiny. For example, it is often pointed out that no document has survived with Hitler’s signature on it ordering the murder of Jews. From this it is argued, implausibly, that Hitler can have known nothing of the Holocaust. Holocaust denial is based upon the systematic suppression or distortion of the evidence. (For the court case involving David Irving, see Chapter 2.)

Whig history

The Whigs were a political group that dominated British politics in the eighteenth and nineteenth centuries. Whiggery developed out of the parliamentary side in the English Civil War and retained a strong attachment to the rights and privileges of Parliament, whose virtues, the Whigs believed, could be explained by a particular reading of English history. The Whigs interpreted it as a long battle with the Crown for the restoration of the ancient rights of Parliament, which they fondly believed had been enjoyed in Anglo-Saxon times but lost at the time of the Norman Conquest. Historians have long since shown the Whig version of events to be myth, but the general theme of progress towards a pinnacle in the present day remained very popular. Herbert Butterfield attacked Whig history for looking at medieval or Tudor institutions in entirely modern terms and not taking the contemporary context into account; left-wing historians have rejected Whig history for its over-confident, patriotic tone. The general Whig tendency, however, to see modern conditions or attitudes as the peak of perfection and then to look to the past to see how we attained it, is by no means confined to constitutional history and can be found in such diverse fields as women’s history or the history of science and medicine.

Further reading

E.H. Carr, What is History?, Penguin, 1961.

G.R. Elton, The Practice of History, Fontana, 1969.

W.H. Walsh, An Introduction to Philosophy of History, 3rd edn, Hutchinson, 1967.

Richard Evans, In Defence of History, Granta, 1997.

Ludmilla Jordanova, History in Practice, Arnold, 2000.

R.G. Collingwood, The Idea of History, Oxford University Press, 1946.

Keith Jenkins, On ‘What is History?’, Routledge, 1995.

Alun Munslow, Deconstructing History, Routledge, 1997.

Beverley Southgate, Postmodernism in History, Routledge, 2003.

Hayden White, The Content of the Form, Johns Hopkins University Press, 1987.

Joyce Appleby, Lynn Hunt & Margaret Jacob, Telling the Truth about History, Norton, 1997.


  1  G.R. Elton, The Practice of History, Fontana, 1969.

  2  Theodore Zeldin, ‘Ourselves as we see us’, Times Literary Supplement, 31 December 1982. See also his article, ‘After Braudel’, The Listener, 5 November 1981.

  3  See, for example, Elton, The Practice of History, pp. vii–viii.

  4  Lee Benson, Toward the Scientific Study of History, Lippincott, 1972.

  5  L. von Ranke, quoted in Peter Novick, That Noble Dream: The ‘Objectivity Question’ and the American Historical Profession, Cambridge University Press, 1988, p. 28.

  6  Zeldin, ‘After Braudel’. See also his article, ‘Social and total history’, Journal of Social History, X, 1976, pp. 237–45.

  7  K.R. Popper, The Open Society and its Enemies, vol. II, 5th edn, Routledge & Kegan Paul, 1966, p. 265.

  8  Raphael Samuel (ed.), People’s History and Socialist Theory, Routledge & Kegan Paul, 1981, editor’s introduction, p. xlv.

  9  E.H. Carr, What is History?, Penguin, 1964, p. 16, rather surprisingly falls into this error.

10  Lord Acton, letter to the contributors to the Cambridge Modern History, 1896, reprinted in Fritz Stern (ed.), Varieties of History, 2nd edn, Macmillan, 1970, p. 247.

11  Carr, What is History? p. 120.

12  L.B. Namier, Avenues of History, Hamish Hamilton, 1952, p. 8.

13  M.M. Postan, Fact and Relevance, Cambridge University Press, 1970, p. 51.

14  Elton, Practice of History, pp. 74–82.

15  See R.C. Richardson, The Debate on the English Revolution Revisited, Routledge, 1988.

16  Marc Bloch, The Historian’s Craft, Manchester University Press, 1954, p. 65.

17  Popper’s views are lucidly expounded in Bryan Magee, Popper, Fontana, 1973.

18  Joseph C. Miller ‘History and Africa/Africa and History’, American Historical Review, CIV, 1999, p. 27.

19  R.F. Holland, European Decolonization 1918–81, Macmillan, 1985.

20  James Joll, The Origins of the First World War, Longman, 1984.

21  Jakob Burckhardt, The Civilization of the Renaissance in Italy, Phaidon, 1960, p. 1.

22  This is particularly true of Elton, Practice of History.

23  Richard Cobb, A Second Identity, Oxford University Press, 1969, p. 47. See also Zeldin’s comments in the same vein in France 1848–1945, vol. I, Oxford University Press, 1973, p. 7.

24  See, for example, Pieter Geyl, Napoleon: For and Against, 2nd edn, Cape, 1964.

25  Compare, for example, Peter Laslett, The World We Have Lost, 2nd edn, Methuen, 1971, with E.P. Thompson, Whigs and Hunters, Penguin, 1977.

26  Michael Howard, The Lessons of History, Oxford University Press, 1981, p. 21.

27  J.B. Bury, ‘The science of history’, 1902, reprinted in Stern, Varieties of History, p. 215.

28  Adrian Wilson, ‘The infancy of the history of childhood: an appraisal of Philippe Ariès’, History and Theory, XIX, 1980, pp. 132–53.

29  H. Butterfield, The Whig Interpretation of History, Penguin, 1973, p. 30.

30  Cf. Sheila Rowbotham, Hidden from History, Pluto Press, 1973.

31  Cited in J.H. Hexter, On Historians, Collins, 1979, p. 15.

32  Gordon Connell-Smith and Howell A. Lloyd, The Relevance of History, Heinemann, 1972, p. 41.

33  Elton, Practice of History, p. 31.

34  Elton’s conservative convictions are most clearly set out in his two inaugural lectures, ‘The future of the past’ (1968) and ‘The history of England’ (1984), reprinted in his Return to Essentials, Cambridge University Press, 1991.

35  Roland Barthes, Image, Music, Text, Fontana, 1977, pp. 42–8.

36  The textual theories that have grown up in the wake of Saussure are usefully set out in Raman Selden, Peter Widdowson and Peter Brookes, A Reader’s Guide to Contemporary Literary Theory, 4th edn, Prentice Hall, 1997.

37  For a good introduction, see P. Rabinow (ed.), The Foucault Reader, Penguin, 1991.

38  Jacques Derrida, Of Grammatology, Johns Hopkins University Press, 1976, p. 158.

39  Raphael Samuel, ‘Reading the signs’, History Workshop Journal, xxxii, 1991, p. 93.

40  Hayden White, The Content of the Form, Johns Hopkins University Press, 1987, p. 72.

41  Michel Foucault, The Archaeology of Knowledge, Tavistock, 1972.

42  Keith Jenkins, Re-Thinking History, Routledge, 1991, p. 68.

43  Hayden White, quoted in Novick, That Noble Dream, p. 601.

44  See, for example, Alun Munslow, Deconstructing History, Routledge, 1997; Keith Jenkins, On ‘What is History?’, Routledge, 1995.

45  Linda Colley, Britons: Forging the Nation, 1707–1837, Yale University Press, 1992, is a good example of a highly critical analysis uninfluenced by Postmodernism.

46  For attacks on the Whig interpretation of history, see J.C.D. Clark, English Society 1688–1832: Ideology, Social Structure and Political Practice during the Ancien Régime, Cambridge University Press, 1985, and Conrad Russell, The Causes of the English Civil War, Oxford University Press, 1990. For attacks on the concept of the Industrial Revolution, see R. Floud and D. McCloskey (eds), The Economic History of Britain since 1700, 2 vols, Cambridge University Press, 1981.

47  Gareth Stedman Jones, Languages of Class: Studies in English Working Class History 1832–1982, Cambridge University Press, 1983, p. 107.

48  H. John Field, Toward a Programme of Imperial Life, Clio Press, 1982; John Tosh, ‘What should historians do with masculinity? Reflections on nineteenth-century Britain’, History Workshop Journal, XXXVIII, 1994, pp. 179–202.

49  For a review of these trends, see Peter Burke (ed.), New Perspectives on Historical Writing, Polity Press, 1991.

50  See, for example, Joan Scott, Gender and the Politics of History, Columbia University Press, 1988.

51  Gabrielle M. Spiegel, ‘History, historicism, and the social logic of the text’, Speculum, LXV (1990), pp. 59–86.

52  Gertrude Himmelfarb, On Looking Into the Abyss, Knopf, 1994.

53  Joyce Appleby, Lynn Hunt and Margaret Jacob, Telling the Truth About History, Norton, 1994, p. 236.

54  Elton, Return to Essentials; Arthur Marwick, ‘Two approaches to historical study: the metaphysical (including “Postmodernism”) and the historical’, Journal of Contemporary History, XXX, 1995, pp. 5–35.

55  Appleby, Hunt and Jacob, Telling the Truth About History; Richard J. Evans, In Defence of History, Granta, 1997.

56  E.P. Thompson, The Making of the English Working Class, revised edn, Penguin, 1968, p. 916.

If you find an error or have any questions, please email us at Thank you!