As a country, Britain is not particularly given to controversy. To the outside world, its institutions have often presented an image of unruffled calm, quite unused to the unseemly clamour of angry disagreements. Among these institutions, those which radiate an air of the most beatific calm are its universities. Populated by studious dons, home only to the quiet rustle of dusty pages, they are commonly viewed as having carefully cultivated their status as temples of learning, removed from even the slightest hint of dispute. Whereas in Europe, intellectuals have been at the forefront of public debate and universities have proved to be fruitful hothouses of dissent, the academic institutions of Britain seem to the outsider to have something of the character of the ivory tower, quite cut off from the noise and passion of the controversial.
In May 2003, however, this image was shattered and the normally quiet quads of Britain's universities echoed with the sound of angry voices raised in protest. This being England, there were, of course, no public bonfires of mortar boards, and there were no gown-bedecked sit-ins staged in Downing Street, but the groundswell of dissent which broke forth nonetheless stirred passions of the strongest kind. No war, no conflict or political shock had caused this outburst, but something which struck to the very core of academic life. Innocuous perhaps to the everyday observer, the event which called forth this trauma touched on the essence of a particular branch of intellectual life and, as such, threatened the identity of a subject itself.
On an official visit to a provincial university, Charles Clarke - then Secretary of State for Education - earned the undying enmity of almost every historian in the country and stirred up the controversy which was to inflame normally quiet common rooms up and down the United Kingdom. Setting out the government's approach to higher education in his speech, he neatly - if incautiously - described his understanding of the criteria for granting state funding to research projects. 'I don't mind there being some medievalists around for ornamental purposes,' he declared, without a hint of irony, 'but there is no reason for the state to pay for them.' Himself a Cambridge-educated mathematician and economist, Clarke wished the state to pay only for subjects of 'clear usefulness.' It was abundantly clear that medieval history did not count as a subject of 'clear usefulness'.
At the time that Charles Clarke made this speech, I was studying a particularly arcane aspect of medieval history and even then I had no doubt that the relationship between Marsilius of Padua's Defensor pacis and Latin Averroism could never be described as having a 'clear usefulness,' even on a good day. Even though I was shocked by the remark, however, I remember being strangely flattered by Clarke's comments: it was rather touching to be described as 'ornamental'. I admit that, horrifying though I found Clarke's statement, I quite liked the idea. Senior members of the university, however, were less compromising in their response. Gillian Evans, a noted medieval scholar from a nearby college, decried Clarke as a 'philistine thug', and in private, others were even more forthright. The cries of outrage became deafening as medieval scholars shouted their disapproval with mounting anger. Neither before nor since have I seen medievalists so impassioned.
It is perhaps amusing that it took so abstract an issue to enrage British academics, but it is nevertheless true that Charles Clarke's comments went to the very heart of what it is to be an historian and touched upon both the merit and function of the subject. Although he was speaking from the limited perspective of government funding for research, his speech was underpinned by a belief that academic subjects can be described in terms of 'use', that they can be measured in terms of social utility. In this regard, it is noteworthy that he did not specifically condemn history tout court, but only medieval history. History, he seemed to imply, was 'useful': medieval history, however, was not. It is difficult to read too much into his brief description of this distinction, but it is a surprisingly common view that medieval history is indeed 'useless', precisely because it concerns only the medieval world. Being far removed from the modern period, it has nothing to tell us about today: it has, in other words, no relevance for the modern world and hence no use. Although many would see this as a perfectly preposterous suggestion and would decry such a distinction, it is the concept of use in history - and the associated notion of relevance - which Clarke's comment reveals that is of particular interest here.
History and Use
Historians have long been preoccupied with the question of history's function and have for the most part not been averse to claiming some use - and even relevance - for their subject. Few would dispute that knowing more about the past is in itself a praiseworthy venture, and there are many who see the discipline of history - as a vehicle for the exercise of particular skills of analysis and argumentation - as a valuable preparation for a career in another field (as discussed in Timothy Stanley's article elsewhere in this issue), but the question of the use of historical research is at some distance from these concerns and has proved to be an unusually challenging problem for the professional historian.
In the eyes of many historians and public figures, the study of history has use primarily insofar as it can teach us something and this seems to have been implicit in Clarke's original comment. More specifically, history is capable of teaching us something about ourselves and about our own day. It can, as a result, serve both the present and the future. This, however, is a deceptively simple statement. Behind this contention lurks a multiplicity of different viewpoints and, while it would be impossible to examine ever shade of opinion, it is worth exploring a few of the most important attempts to develop such a line of argument.
From Antiquity to the Renaissance, writers of historical works (distinct from historians in the modern, professional sense) tended to believe that the value of history lay in its ability to impart knowledge of universal truths about human nature. As Thucydides argued in his account of the Peloponnesian War, and as most of his heirs believed, history was moral philosophy taught by example. For the Roman historian Sallust, the attempt to extol the virtues and vices of great men, would help to inspire his reader to avoid evil and embrace the good. In the fourteenth century, Francesco Petrarca similarly saw historical writing as being closely allied to epideixis, the rhetoric of praise and blame. In his De viris illustribus, Petrarca attempted to reveal truths about virtue and vice so that others might learn by example. Past, present and future were united by the unchanging nature of human nature; the role of history - like that of moral philosophy - was to act as a guide to right action.
In later centuries, this notion of universality solidified into the notion of historical 'laws'. The study of history revealed facets of human nature which allowed the procession of the past to be conceived of in terms of an inescapable obedience to immutable rules that, once uncovered, would give some indication both about the present and about the future. For Hegel, history actualised the movement of the spirit (geist) and the particularities of a society or people at any given moment in the past dissolved within the motion of the spirit. In the Elements of the Philosophy of Right, indeed, Hegel declared that history was 'the necessary development, from the concept of the freedom of the spirit alone, of the moments of reason and hence of spirit's self-consciousness and freedom.' Later, Hegel's thought was taken up and developed by Marx, who replaced this notion with his own 'materialist conception of history.' Rather than being concerned with the metaphysical issue of the development of the spirit, Marx saw history as revealing the course of struggles between classes founded on the competing interests of capital and labour. Despite the problems of interpretation that this poses, Engels believed that Marx had revealed the 'great law of motion in history', a law which was 'analogous in scope and precision' to 'the law of transformation of energy.'
Although the immense influence exerted by Marxist historiography until the age of détente constitutes the most significant example of the continuity of this understanding of the 'use' of history, it is noteworthy that the idea of historical 'laws' continued to shape approaches to the value of the subject throughout the twentieth century. Arnold Toynbee's monumental thirteen-volume A Study of History, published between 1934 and 1937, is perhaps the most illustrative case study of an attempt to detect 'laws' at play in human history. In this massive analysis, spanning literally thousands of years of civilisation, Toynbee described - in excruciating and almost mathematical detail - the laws which govern the rise and fall of civilisations, and the role played by great men in those same cultures with a view to understanding where, in a period of considerable geopolitical uncertainty, humanity's fate might lie. While Toynbee's work was certainly a product of its time, the idea of historical 'laws' proved unexpectedly enduring. More recently, it has been revisited by Francis Fukuyama in The End of History, written under the influence of the Marxist/Hegelian scholar Alexandre Kojève. Predicting the inevitable triumph of liberal capitalist democracy, Fukuyama claimed that '[m]odern natural science has provided us with a Mechanism (sic.) whose progressive unfolding gives both a directionality and a coherence to human history over the past several centuries.' Economic development, working in tandem with a universal 'desire for recognition' both explained and predicted what Fukuyama perceived to be an inevitable march towards liberal capitalist democracy across the globe.
Amongst historians, however, the idea of universal 'laws' is now extremely rare and - as a result of the multiplicity of obvious intellectual and methodological problems to which it is prey - it has largely faded from view, consigned (appropriately) to the dustbin of history. With the advent of the professionalisation of history, however, there emerged a more concerted attempt to understand the use or value of the subject in term related to the logical processes intrinsic to its operating methods. Particularly in the aftermath of the Second World War, the view emerged that the study of history is ultimately based on the notion of causality. Occupied with uncovering the causal relations between particular occurrences, history is capable not merely of telling us more about the past, but is also able to give us both an understanding of the present and a reasonable ability to predict the future. By knowing how we came to be here now, it is suggested, we may perhaps gain some insight into where we are going in future.
This view has been most ably represented - albeit in radically different ways - by two historians who both taught at Cambridge from the mid-1950s onwards: E. H. Carr and Sir Geoffrey Elton. Although, as we shall see, they disagreed profoundly over many issues (especially the historian's relationship to his 'facts'), they were nevertheless united by a common belief that the role of causation in history allowed it to serve a valuable purpose both in comprehending the heritage of the present, and in positing the likely course of the future.
Opposing those like Isaiah Berlin who placed great emphasis on the role of chance in human affairs, E. H. Carr contended that despite the role of accidence, historical thinking 'is always teleological'. The historian, Carr claimed, must always ask both 'why?' and 'whither?' As such, the practice of history was necessarily bound up with a notion of continual change and development, a sense of progress. 'History properly so-called,' he argued in What is History?, 'can be written only by those who find and accept a sense of direction in history itself. The belief that we have come from somewhere is closely linked with the belief that we are going somewhere.' Causation, in other words, explained the relevance of history for society. The modern man, Carr asserted, 'peers eagerly back into the twilight out of which he has come, in the hope that its faint beams will illuminate the obscurity into which he is going; and, conversely, his aspirations ad anxieties about the path that lies ahead quicken his insights into what lies behind. Past, present and future are linked together in the endless chain of history.'
Although he was scornful of Carr's understanding of 'progress' and associated himself with a more relentlessly empirical approach to the subject, Sir Geoffrey Elton shared the belief that the historian must think of the subjects of his particular analysis 'as steps in a chain of events, as matters explanatory of a sequence of happenings. He will have to concentrate on understanding change, which is the essential content of historical analysis and description.' While Elton pointed out that the evidence may force the historian to recognise that some sequential events were not causally related, his acceptance that there was a role for causation in history underpinned a significant part of his conception of the 'use' of history. Although history's 'lessons are not straightforward didactic precepts, either instructions for action ... or universal norms ... a sound acquaintance with the prehistory of a situation or problem does illuminate them and does assist in making present decisions; and though history cannot prophesy, it can often make reasonable predictions. Historical knowledge gives solidity to the understanding of the present and may suggest guiding lines for the future.'
While the notion that causation is in some way intrinsic to the study of history is undoubtedly the most widely supported explanation of the subject's use, a very different perspective has been provided by those who have adopted what is sometimes characterised as an 'idealist' conception of the historian's task. Despite the fact this point of view enjoyed something of a renaissance during the 1950s and found many very able exponents, its most notable exponents remain Benedetto Croce and R. G. Collingwood.
Writing in the shadow of early twentieth-century German philosophers, Croce and Collingwood each came to regard the contention that an objective knowledge of the past was possible with extreme scepticism. Both Croce and Collingwood recognised that the historian could never approach his source materials with an entirely clear mind: in attempting to construct an interpretation of the past, he inevitably came to select what documents or facts were 'important' and which were not. The historian, in effect, created 'historical facts'. Although this was to have some influence on E. H. Carr in What is History?, Croce and Collingwood developed the idea in a manner which related the past and the present most specifically in the mind of the historian himself. The historian could not help but view the past, and select evidence concerning the past, from the perspective of the present. In History as the Story of Liberty, Croce contended that '[t]he practical requirements which underlie every historical judgement give to all history the character of "contemporary history", because, however remote in time events thus recounted may seem to be, the history in reality refers to present needs and present situations wherein those events vibrate.'
Since historians consciously or unconsciously select their evidence in pursuit of grand narratives or causal explanations, the true business of history lay in the meaning behind statements, in reconstructing the mental processes of agents in the past. Like Croce, Collingwood conceived of history as 're-enactment' or, to use a term borrowed from Wilhelm Dilthey, as 'hermeneutic understanding' (Verstehen). Indeed, in The Idea of History, Collingwood suggested that historical understanding was based on consciously attempting to re-think the same thoughts as historical agents. 'History,' he claimed, 'is the re-enactment in the historians mind of the thought whose history he is studying.' Through hermeneutic understanding, the past effectively lived again in the present. Past and present, for both Croce and Collingwood, could become one in the practice of history. In casting light on the thought of the past, the historian would also come to illuminate his own thought, and in this lay the use of history. 'History is for human self-knowledge,' as Collingwood put it. '[It's] value is that it teaches us what man has done, and thus what man is.'
History and Truth
Despite their radical difference, each of these various attempts to describe a use for history ultimately rests on the underlying assumption that the historian's task lies in uncovering an underlying truth. For universal laws to be valid, or for causal explanations to have application to the present and future, truth must be vested in the practice of history. By the same token, in order for the thought of past actors to be 're-enacted' through hermeneutic understanding, the true meaning of a statement must be accessible to the historian.
Although it is embedded in much earlier forms of historical writing, the idea that history necessarily pursues the truth about the past has been a central claim of academic historians since the professionalisation of the subject in the early nineteenth century. Leopold von Ranke, is often seen as having been a seminal figure in this regard, and his association of history and truth has on occasions acquired the status almost of a credo for academic historians. In the preface to his Histories of the Latin and Germanic Nations, written in 1824, von Ranke boldly claimed that the task of history was to show 'how it essentially was' - 'Wie es eigentlich gewesen'. The reality and character of the past were to be sought relentlessly in the ruthless pursuit of the truth. History became a wissenschaft, what we might in English call a 'science' (albeit rather inaccurately). The past was, at root, knowable through the facts at the historian's disposal and the facts alone would reveal the truth of the past. 'Strict presentation of facts,' he asserted, 'no matter how conditional and unattractive they might be, is undoubtedly the supreme law.' The claim that history consists in the mere 'presentation of facts' is perhaps a little stringent, but it is nevertheless a relatively commonplace historical orthodoxy that history is an attempt to construct a plausible interpretation based on known facts.
Facts, however, are tricky things for the historian. Although Carr - drawing on Croce and Collingwood - seems to have believed that a past event became an historical fact only by virtue of the fact that the historical constituted it as such, Richard Evans has correctly pointed out that this represents a semantic misunderstanding and relies confusion between the means of verifying a fact and the fact itself. An historical fact, Evans has argued, is something that happened in history. It is a past reality which exists 'entirely independently of the historian'. There would, I think, be few who would now disagree with this view. Everything that occurred in the past, everything that was thought or built or done, is an autonomous and unalterable fact. The reality of the past is immutable. As Aristotle reported the Athenian poet Agathon as saying, 'even God cannot change the past'.
Although the reality of facts may not be in question, the problem for the historian is in determining how it is possible to gain knowledge of those facts which are the essential materials for his enterprise. In answering this question, it is necessary to engage with what is meant by 'evidence'. In the highly readable In Defence of History, Richard Evans - who has recently been appointed Regius Professor of Modern History at Cambridge University - has argued that while facts are conceptually distinct from evidence, facts can be used as evidence to test a particular theory. As Evans puts it (following Nancy Partner): '[t]he historian formulates a thesis, goes looking for evidence and discovers facts.' This distinction between facts and evidence is, for Evans, based on the understanding that facts can be used to evidence very different points that may have relevance for a multiplicity of different theses. Facts, in other words, are discrete entities; evidence, on the other hand, sheds light on the 'interconnectedness' of various facts.
Evans' objective in distinguishing facts from evidence is not merely to rectify a semantic confusion, but to concentrate attention on the way in which the historian constructs an interpretation of the fact. The use of facts as evidence for the 'interconnectedness' of past realities is, he observes, the subject of important theoretical questions about the means by which such 'interconnections' are discerned and the manner in which historical interpretations are built. These questions have rightly been debated with great intensity in recent years and particularly interesting issues have been raised by postmodernist approaches to the reading of texts. While the susceptibility of the historical interpretation of facts itself calls into question the validity of 'using' history to aid our understanding of the present and future, Evans' distinction between facts and evidence omits to consider a more fundamental problem confronting the historian's search for truth and, as a result, the 'use' of history more generally.
In conceiving of evidence as the way of 'reading' a fact to reveal its relationship to other facts, Evans has little interest in how facts themselves may be known and in this regard perhaps betrays the influence of his own background in the richly documented field of modern German history and interest in the political events of his period. It is indeed true that - once known - a fact may be 'read' in a variety of ways and thus become evidence in support of a multiplicity of interpretations of different issues, but it is impossible to ignore that in order to reach that level of historiographical thought, one must first have an understanding of how an historian can know a fact. In this respect, we must return both to the nature of an historical fact and to the meaning of the word 'evidence'.
At face value, the question of how the historian can know an historical fact appears to be a rather abstruse and abstract question, more germane to epistemological debate than the study of history. A wide range of past events can indeed be known as facts. It is perfectly obvious that John Kennedy defeated Richard Nixon in the 1960 presidential election, and that William the Conqueror won the Battle of Hastings in 1066. So too, past realities other than events can be known as facts. It is plain that many houses in the Tudor period were constructed using wooden beams as a frame for wattle and daub, and that Brunelleschi was responsible for many technical innovations in designing the dome for the Duomo in Florence. These are facts that can be known without too much difficulty. There is nothing which suggests that we should have cause to question their veracity. The historian can draw on a huge range of sources - both documentary and otherwise - which unambiguously testify to their truthfulness. The careful study of source materials can uncover a huge range of facts and part of the historian's task is to search continually for new factual information which can be used in the testing of various hypotheses.
Although a wide range of materials - from works of art to archaeological finds - are at the historian's disposal, however, he relies most firmly on documentary sources to recover information about the past. In that they are written by fallible human beings, documents present a number of problems to the scholar. Far from being windows through which we may see the past clearly, documents allow only an occluded view of a particular historical moment. Every schoolchild is taught that documents are sometimes 'biased' or may 'distort' reality. In many cases - indeed, in most cases - they do not necessarily represent the past perfectly, but provide only an unclear reflection of a particular author's preoccupations and are frequently coloured by potentially misleading authorial interests. Some documents may contradict others, extraneous details may be introduced, exaggeration may be employed, and telling information may be ignored or passed over.
For all the interpretative difficulties which they occasionally present, however, documents customarily retain historians' confidence. As Elton believed, it is usually the case that, if read with sufficient care, documents can furnish the historian with a reasonable understanding of facts regarding the past. Even when uncertainty is unavoidable due to the limitations of the available documentary sources, that uncertainty can itself be illuminating. While far from perfect, documents can, in other words, allow access to a measure of truth.
The confidence which historians have shown in their documents is often strongest in particular fields. Those interested in political history, and particularly modern political history, are usually among those who express the greatest faith in documentary sources, and this faith is not unrelated to the relationship between processes of thought and external reality in the past. It is true of all fields in the study of history that the thought of historical actors is of great importance. While events and occurrences are interesting and even revealing, the connections between them, their causes and effects, are concerned primarily with the thought of particular people at a particular time. It would not be entirely unjustified to suggest - as many, including Elton and Carr, have done - that the thought of historical actors may be seen as the 'medium' of history itself. Where the historian is interested in political history, for example, this may not in itself be especially troubling, and the same might perhaps be said of economic history, military history and even social history. In these fields, thought is in many ways manifested externally, in events and occurrences: where it is not directly visible, it may be inferred with a reasonable degree of certainty. Intelligently employed, documents can be used to provide evidence for patterns of thought which are, at the very least, plausible. The Holy Roman Emperor Charles IV's promulgation of the Golden Bull in 1356 can, for example, be understood in the context of the confusion and conflict which had arisen from uncertainties in the imperial constitution in earlier decades. By the same token, it is possible to understand Caesar's decision to cross the Rubicon in 49BC in terms of the danger of prosecution which awaited him in Rome as a consequence of his actions during his first consulship and the collapse of the First Triumvirate. Similarly, the reasons for the success of the Christian Democrats in post-war Italy can be inferred from the social and economic conditions of the period and from the adoption of policies which contrasted strongly with those of their main rivals, the PCI and the PSI. A certain confidence is, in other words, justified. It is perhaps no coincidence that Elton was a scholar of Tudor government, Carr specialised in the history of the Soviet Union, and Evans' work has concentrated on twentieth-century German politics and society.
Intellectual History, Language and Truth
This confidence is, however, not shared by those working in all fields. As Evans himself notes rather dismissively in In Defence of History, intellectual historians have tended to adopt a more sceptical attitude towards documents. Unlike his colleagues, the intellectual historian is concerned with thought at a more abstract level. Although some areas of intellectual history - such as the history of political thought - are not unrelated to the external realities of a period, others - such as the history of philosophy or the history of the classical tradition - are at a greater remove from the nuts and bolts of everyday life. While biographical knowledge and an understanding of the immediate conditions of the period may provide some insight, the historian engaged in attempting to reconstruct the thought of even a single figure is presented with problems of a more extreme variety than those which confront his colleagues in other fields. These problems are concerned, above all else, with the means by which the 'facts' about a person's ideas may be known through documents.
It may, of course, be taken for granted that, at the point of writing, an historical actor is conscious of his thoughts and attempts to record them to one degree or another. As a result, this conscious thought process is itself a 'fact' which the intellectual historian aims to recover. It is, however, important to observe that the actor's thought exist for the historian only through the medium of language. While it may at first seem innocuous, however, this apparently simple observation contains within it the source of innumerable problems.
For a considerable period of time, historians of ideas were barely troubled with this problem. Until the early 1960s, it was generally the case that such scholars were accustomed to approach their subject by examining the manner in which a canonical group of texts engaged with 'perennial ideas' such as 'liberty' and 'democracy'. It was held that the most effective means of doing this was to study exactly what a particular text says about a given concept.
This approach rested on a belief in the over-riding importance of uncovering the meaning of a text within the framework of a broad discourse on a commonly-understood theme. Language was held both to be almost entirely representational (i.e. words stand for particular concepts or things) and, with regard to 'perennial ideas', made use of a lexicon which was comprehensible across the expanse of history. That an historical actor may have lived several centuries ago and expressed himself using the linguistic norms of his period were not believed to present any significant challenges and it was generally presumed that the historian could reconstruct an actor's thought's simply by establishing what he was saying in a particular text in relation to the general, pan-historical themes that was in question.
In an article first published in 1969, however, Quentin Skinner identified a number of flaws in this approach and, in doing so, catalysed a revolution in the way in which historians understand the relationship between thought and text. In his article, Skinner (who preceded Richard Evans as Regius Professor of Modern History at Cambridge) first demonstrated that the language in which a text was written employed a vocabulary that was specific to the period. The lexicon of a fourteenth-century humanist, for example, was manifestly not the same as that of an eighteenth-century antiquarian, and it is more than possible that notions which later came to dominate intellectual life were simply not part of the cultural vocabulary of an earlier period. Even if two figures separated by many centuries employed the same word in their writings, it may be that it connotes a very different meaning for each actor. Vocabulary, in other words, changes over time. This being so, Skinner argued, it is invalid to approach the history of ideas from the perspective of so-called 'perennial ideas'. There was, however, a second, closely related and - for our present purposes - more important element to Skinner's article. Drawing on the linguistic philosophy of Ludwig Wittgenstein and J. L. Austin, Skinner contended that the search for meaning alone was an insufficient basis for the recovery of an agent's thought. A text, he suggested, consisted not merely of words which mean something, but more fundamentally of words which were intended to mean something. The words an author employed in expressing his thoughts were inscribed with the intention of conveying a specific meaning and with the intention that this be apprehensible to his reader. 'The understanding of texts,' Skinner argued,
'presupposes the grasp of what they were intended to mean and of how that meaning was intended to be taken. To understand a text must at least be to understand both the intention to be understood and the intention that this intention be understood, which the text as an intended act of communication must have embodied. The question we accordingly need to confront in studying such texts is what their authors - writing at a time when they wrote for the specific audience they had in mind - could in practice have intended to communicate by issuing their given utterances.'
As Skinner was to explain further in what was soon to become a classical series of later articles, the process of reconstructing intentional meaning (or what he called, after Austin, the 'intentional force') of a text involved two distinct steps. First, the historian had to 'delineate the full range of communications that could have been conventionally performed on the given occasion by the issuing of a given utterance.' In other words, the historian had to establish all the things that an author could have been doing in making particular statements at the time of writing. In this, Skinner - like Austin and, to an extent, Wittgenstein - regarded statements as being essentially equivalent to actions. Second, the historian must 'trace the relationship between the given utterance and [the] wider linguistic context as a means of decoding the intentions of the given writer ... The social context figures as the ultimate framework for helping to decide what conventionally recognisable meanings it might in principle have been possible for someone to have intended to communicate.' Having established all of the various things that an author could have been doing in making a particular statement, the historian could look to the author's immediate social and linguistic context to establish exactly which of these possibilities was most plausible. Regarding statements in historical texts as a form of action, Skinner was proposing that the way in which people of the time understood particular terms in relation to contemporary events and phenomena could be used to 'decode' the author's intended meaning, in other words. This process has often been called 'closing the context'.
Although, in distinguishing between meaning and intention, Skinner seemed to weaken the historian's ability to recover the 'fact' of an agent's thought, the heuristic method he described was intended rather to strengthen this capacity by bringing greater methodological clarity to the study of texts. Equipped with sturdy tools with which to dig up the 'truth' of the past, history could still potentially have a 'use'.
History and Language: Beyond Truth and Skinner
Despite its many merits, however, I would like tentatively to suggest that Skinner's method is subject to some question. I am not the first to do this. Primarily under the influence of postmodernists such as Jacques Derrida who have proudly proclaimed the 'death of the author', there have been scholars (most notably W. K. Wimsatt and M. C. Beardsley) who have attacked what they perceive to be Skinner's 'intentional fallacy' and challenged the importance which he attached to the distinction between meaning and intention. It is not, however, my intention to follow such an approach. In what follows, I fully accept that the key to the understanding of texts in the history of ideas is to make such a distinction. What I would like to suggest, however, is that in many cases, an author's intention cannot be recovered by thinking of a statement as the equivalent of an action or by 'closing the context', and, as consequence, that there are limitations to the historian's capacity to reconstruct past thought.
Particularly insofar as the historian is dealing with texts in the history of political thought, it initially seems that Skinner's approach is remarkably robust. If we remember the philosophical background to Skinner's argument, the reasons for this become clear. Outlining the idea of language as use, Austin and Wittgenstein both suggested that the context in which a statement is uttered given an indication of the intentional meaning attached to that statement. The same utterance in different contexts could be intended to mean very different things. Wittgenstein, for example, gave the example of the word 'brick' uttered alone as a declaration. Standing in front of a half-completed wall with hand outstretched to his workmate, a bricklayer who said 'brick' would probably intend it as a request, and indicate the action of a brick being passed to him with the statement. If a pedestrian said 'brick' in an agitated voice to a man reading a newspaper near some scaffolding, however, we might assume that it is intended to warn the bystander that a brick was falling towards him, and thus denote both the dangerous action of the falling brick and the need to avoid said brick. In relation to past events, the context of a written statement is no less important in determining intentional meaning and reveals the peculiar position of the historian of political thought. In many ways, a work of political thought can be considered as having been intended to be equivalent to a political action: a text is written with the intention that its intended meaning be understood and, in most cases, that intended meaning is reflective of a contemporary political situation. Although the historian necessarily views the context of a political utterance through a glass darkly, it is nevertheless true that the 'intentional force' of a statement can be understood with a fair degree of certainty by examining the political realities and linguistic norms of the period in which it was written. If a hypothetical author writing in France in late 1788 were to have declared that 'the people demand the king's blood,' it would be reasonable to argue on the basis of the intended audience of the text and the author's own background that the word 'people' referred to a particular section of the populace and that the statement could best be understood in relation to the act of executing a monarch. Further analysis of the linguistic, political and social context may yield further information about the extremity of the force attached to this declaration, the immediate causes for the demand and the degree to which it was representative of sections of the population at the time.
Although Skinner's approach seems perfectly valid for many documents, the relationship between text and context which his method presupposes is in some ways problematic. In transposing the notion of statements as performative actions propounded by Wittgenstein and Austin from immediate utterances to historical documents, Skinner relies on the possibility that the context of a document can be known independent of the text itself, and used to inform the reading of the author's intentional meaning. With a great many documents in the history of modern political thought, this is seldom a cause for concern, but, particularly with texts concerning other aspects of intellectual history in more remote periods, the provenance of a document may not be immediately apparent, and it is possible only to make comparatively vague guesses at the date and place of composition from, for example, the orthography or written style. In such cases, the only means of reconstructing the context of a document is through an analysis of the text itself. By examining references and allusions in the text, it may be possible to attempt establish its provenance with greater precision. Superficially, this appears to remove any problems. With the benefit of this 'reconstructed' context, it would then seem possible to apply Skinner's method and recover the author's intentional meaning. Yet to reconstruct a document's context on the basis of the text itself, and then to use that context to 're-read' the text is inherently dangerous and even flawed. The heuristic circularity involved is plain and, inadvertently, such a method forces Skinner's argument to become reliant both on an initial, speculative assumption of the author's intentional meaning and on a conception of what 'characterised' a period or location. On the basis of what may be an extremely limited supply of direct evidence in the text itself, and reliant on indirect references and allusions, the historian is compelled, first, to establish their presumed meaning, and, second, to relate that to what he may believe to be characteristic of a specific place or time. In the first stage of this process, the historian is forced to narrow the range of an utterance's possible intentional meanings without the benefit of a known context, a practice which Skinner's method seems designed to avoid. In the second stage, the historian is obliged to rely on the methodologically challenging process of determining the cultural and intellectual nature of what is - by necessity - an ill-defined collection of individuals at a particular historical moment, and is thus pray to the dangers inherent to notions of social or cultural identity.
Even when applied to documents written by an individual author whose identity is known, the problems of Skinner's method are apparent. One example is sufficient to illustrate this. More than a thousand letters written by the fourteenth-century Italian humanist Francesco Petrarca are known to survive. Although Petrarca dated many of these letters, a significant number are undated and contain no direct indication of where they were written. What is more, in gathering his letters into two large collections, Petrarca is known to have engaged in a process of revision and often dramatic alteration. The same is true of his treatises on moral philosophy, including the Secretum, the De vita solitaria and the De otio religioso. To complicate the matter further, Petrarca continually encountered new works of classical and patristic literature which, at different stages in his life, had a profound influence on his outlook. While these tendencies have led Umberto Bosco to claim that Petrarca was an author 'senza storia', many scholars interested in reconstructing his thought and intellectual development are necessarily deeply concerned with the dating of his works and there relationship to his assimilation of other modes of thought. In that it is related to his reading and to underlying thought patterns, the date at which Petrarca may have composed a particular work is of particular importance in apprehending his attitude towards a particular issue.
In attempting to understand and date Petrarca's works, however, the historian is faced with a difficult and problematic task. Looking at the Secretum, for example, scholars have attempted to apprehend his intentional meaning by examining the text as a collage of overlapping compositional layers. Using essentially philological techniques, these compositional layers have been identified by attempting to detect shifts in argumentation. Those passages which are held to be more reflective of Stoic thought, for example, and which bristle with allusions and references to Cicero and Seneca, are contrasted with those which are believed to manifest the theology of St. Augustine more clearly. The tenor of these passages is then used as the basis for dating the various revisions to the text in relation to moments at which he discovered particular works of classical literature, or is thought to have altered in his sympathies. Some scholars, such as Hans Baron and Francisco Rico have seen the 'Augustinian' passages as having been written under the influence of the saint's De vera religione in an early period of Petrarca's life, and the Stoic passages as the products of his later reading of Cicero's letters to Atticus. Other scholars, however, have suggested precisely the opposite, and have seen purportedly 'Stoical' passages as being part of an earlier draft. For both groups of scholars, however, the identity of these compositional layers and the dating of revisions to the text are integral to understanding the intentional meaning of the Secretum.
Such a method is replete with methodological problems. From the disagreement between scholars regarding the dating of the various compositional layers in the Secretum, it is plain even to the layman that there are manifold difficulties in characterising Petrarca's thought during any given period of his life, difficulties which are made more extreme by the confusion which reigns over the dating of most of his other works. Insofar as we are primarily interested with the use of the text to establish context, however, it is of far greater relevance that such a method is dependent on a preliminary reading of the meaning of particular passages of the Secretum in order to relate them to specific moments in Petrarca's intellectual development. In concentrating so explicitly on the text as a source of information about its context, scholars have universally over-emphasised the literalism with which Petrarca referred to works of classical and patristic literature, and have dramatically under-estimated the compatibility of purported compositional layers which are believed to express such radically different philosophical positions. As a result, the intentional meaning of the text has been drastically misrepresented. Where a 'preliminary' reading of the text is abandoned, and preconceptions about the context of the work are set aside, it is possible to read the work in a more logically satisfying manner. Although further work has yet to be done on its implications for the dating of the text itself, there is a strong argument to suggest that the 'Stoic' and 'Augustinian' passages which historians have contrasted so vividly are in actual fact complementary components of a coherent moral sentiment, with Stoic quotations being integrated into an argument derived primarily from the early works of St. Augustine, who was himself profoundly influenced by Stoicism.
In the case of Petrarca's Secretum, it would not be unjust to suggest that the use of the text to establish context generates a misleading impression of the work's intentional meaning, and that the method of examining the text in its own terms, as a network of interconnected concepts, allows a more satisfactory understanding of what Petrarca intended to be understood from his dialogue. This method of attempting to recover the intentional meaning of a text through reference only to the 'conceptual network' of the text itself is equally, if not more valid when we are concerned to establish not the intentional meaning of an entire work, but the intentional meaning of single terms employed in purely philosophical discourse, without reference to physical reality.
Where the historian is interested in establishing exactly what an author intended to mean in using a word such as 'conscience', it initially seems difficult to look 'beyond' the text for an answer since it is implausible to suggest that, in real terms, it corresponds to a literal action in the same way as the word 'brick' in Wittgenstein's example. It may, however, be argued that although there is no physical action per se, and thus no context of real social action, the word 'conscience' in our example nevertheless has a linguistic context and the utterance containing the term can be seen as relating to its current uses, perhaps in other philosophical works. It may therefore be regarded as a purely linguistic 'action' insofar as the utterance containing the word can be understood by the reader from within a particular intellectual and linguistic context, and was intended to be understood in such a manner by the author.
As with Petrarca's Secretum, however, there is a problem. It may be the case that in attempting to reconstruct the intentional meaning of the word 'conscience' on the basis of a supposed linguistic context, the historian is leading himself into error. The way in which a word is commonly used at a particular time or in a particular discipline may not relate to the meaning which the author intended to communicate. The author, indeed, may have used the word 'conscience' knowing its common usage, but with the specific intention of attaching new meaning to the term. Indeed, it may be the case that many contemporaries, if presented with the text, may have misinterpreted the utterance containing the word. Authors, especially of philosophical or highly literary texts, are at liberty to attach new meanings to even the most familiar words and to alter their use radically. In these cases, 'closing the context' may at best tell us that the meaning of the word does not relate to common usages, but is more likely, in fact, to lead to a mistaken interpretation of the text's intentional meaning. Indeed, 'closing the context' may blind the historian to both subtle and important changes in meaning across time.
If the example of an author having attached new meaning to a commonly-used word is problematic for Skinner's method, the case of neologisms is yet more troublesome. Some apparently simple neologisms can, of course, be understood without much difficulty as mere compounds of common words or particles. Persius' 'semipaganus' is a good example of such a 'straightforward' neologism. There are, however, some neologisms which are entirely new and, though perhaps superficially similar to words in common use, are sui generis. In modern times, perhaps the best examples of such neologisms are provided by Heidegger. In Being and Time, Heidegger deliberately created a novel philosophical vocabulary to suit his ends and words such as 'Existentiell' and 'Existentiale' appeared for the first time. Terms such as these are - by their very nature - without a broader context of linguistic usage and, in that they are terms employed in an entirely philosophical discourse (in contrast to Wittgenstein's 'brick'), are also without social contexts which would allow the reader to derive a sense of any intentional illocutionary force.
Where the historian is confronted with the task of recovering the meaning of such neologisms, it seems difficult to see how Skinner's strategy could be applied effectively. The very absence of a linguistic or social context renders the notion of 'closing the context' almost irrelevant, even when one attempts to minimise their importance.
It is true that in some ways neologisms may be seen to function as either 'new pieces' whose identity can be guessed from the moves they perform on a familiar playing board, or as missing pieces in a jigsaw puzzles whose shape can be derived from the gap which they leave. The milieu of the 'playing board' or the almost-complete jigsaw puzzle may indeed be of such a variety that it can be related to the linguistic or even social norms of a given community. As a result, despite their uniqueness, it may be suggested that neologisms can still just be fitted within Skinner's general strategy.
In that it relies on the notion of a 'preliminary' reading, however, this argument suffers from the same weakness as that which may be made to defend the application of Skinner's method to a document of unknown provenance or to words invested with new meanings. In order to establish a 'context' for a neologism which can then be used to narrow down its intentional meaning, I must first have an idea of what the intentional meaning of that inherently uncertain word might be and, moreover, of what context it might fit within. The opportunities for unnecessary preconceptions creeping into such a method are legion and it is not difficult to see how the circularity of the strategy can lead to profound misunderstandings.
A further weakness of Skinner's method in relation to neologisms lies in the fact that in many cases, historical actors sometimes demonstrate that the creation of specialist vocabulary is a self-fuelling enterprise. Once again, this is particularly true of Heidegger. Finding himself entering new philosophical territory, Heidegger discovered the need to develop a new lexicon to describe the path he was following. In coining neologisms such as Existentiell and Existentiale, however, he frequently found the conventional vocabulary of philosophical usage insufficient to illustrate the meaning of such novel terms. If the cartographic analogy can be extended, Heidegger found that he required not merely one or two new symbols to describe the territory he was exploring, but also an entirely new method of projection. As a consequence, neologisms such as Existentiell are often defined in relation to other, equally puzzling neologisms such as ontic. The role played by words familiar to other linguistic contexts rapidly becomes less significant and is fast overshadowed by the network of connections which link Heidegger's neologisms. In such a situation, it seems reasonable to suggest that the method which Skinner describes is all but powerless to assist the historian in his task.
Although it must be granted that the historian will seldom - if ever - be confronted with a text about whose ambient language he knows nothing whatsoever, it seems apparent that attempts to recover the intentional meaning of works such as Heidegger's Being and Time require a very different methodology to that proposed by Skinner. Having observed the difficulties intrinsic to any attempt to rely on the context - or, rather, the supposed context - of a particular work, it is perhaps necessary to concentrate rather more closely on the text itself to determine the meaning which the author intended to be understood. In texts such as Being and Time, which are riddled with neologisms, a particular utterance may be comprehensible only by attempting to relate its potentially unfamiliar terms to other similarly uncertain expressions elsewhere in the text. By constructing a plausible 'semantic network' to model the inter-connections between novel terms in a given text, it seems possible to determine the intentional meaning of a work from within, by inhabiting the author's own vocabulary in a very real sense. Since the linguistic 'game' is being played with so many new 'pieces', the role of each piece can be understood only by reconstructing the rules in relation to the movements of all of the pieces from inside the game itself, and it may, of course, become apparent that familiar pieces from other 'games' are introduced into new roles within this novel activity.
The most significant problem with such a methodology, however, is that the possibility of certainty is greatly reduced. In that its heuristic framework is restricted to the language of the text itself, and the whole is the product of only a single actor's mind, there is nothing outside the text to which appeal can be made to verify the conclusions drawn in an absolutely satisfying manner. Although a reconstructed 'semantic network' might be plausible, and allow for an apparently logical reading of the text, it may be only one of many possible reconstructions, and may not accurately reflect the author's intentional meaning: yet, because there is no outside source of authority to which to appeal, it is difficult to verify or disprove such a reading except through further, possibly inconclusive analysis. While the concept of truth remains, such a strategy reduces it to the status merely of a necessary predicate of study. There is no sure way of knowing the truth about someone's thought through a text of this variety. Indeed, it is tempting to paraphrase Wittgenstein himself in suggesting that a text of this type will only be understood by someone who has already had the thoughts that are expressed in it, but even then, such a person would never be able to be entirely sure that he had had those same thoughts.
It might, of course, be objected that since texts containing a large number of neologisms or involving significantly novel usages, or of unknown provenance are relatively rare, there is no need to regard the weaknesses of Skinner's method as too troubling, or to attempt to substitute a form of linguistic network analysis for 'closing the context'. Yet this is perhaps to ignore the fact that an examination of texts of unknown provenance, or containing a large number of novel usages or neologisms reveals wider issues about the nature of language and the methods of reading available to the historian. While such texts are indeed uncommon, it is nevertheless true not only that there are dangers intrinsic to 'closing the context', but also that the appearance of new usages and new words is inherent to the erratic and unpredictable evolution of language. Every day, conversations around the world bristle with new meanings and new uses - meanings and uses which may be apparent in the immediate context of a spoken conversation but which, if employed later in writing, may in all respects seem to conform to previous meanings and uses. Although in some cases - such as with Heidegger's Being and Time - it is possible to identify neologisms and new usages without too much difficulty, it may often be more difficult to discern alterations in the intentional meaning of apparently familiar words where the emphasis is on the context, rather than the content of the text in question. Although Skinner's method was originally proposed as a challenge to the study of 'universal ideas' in the canon of political thought, and was designed to react against the imposition of anachronistic readings, his own emphasis on context may actually be seen to suffer from a comparable insensitivity for linguistic variation. What we may term a linguistic network analysis is perhaps significantly more sensitive to such changes, and thus may lead the historian closer to an author's intentional meaning in much wider range of cases than might first appear.
If Skinner's method is pursued, it may be that significant alterations in language - and hence in intentional meaning - may be missed, or misrepresented. Although its emphasis on context seems to offer more opportunity to verify an hypothesis, that hypothesis may be radically wide of the mark even once 'verified' using a potentially unsound method. If, however, the importance of detecting linguistic change to the recovery of thought is accepted, it must simultaneously be granted that the method which may lead the historian closer to truth will at the same time severely limit his capacity to recognise the truth he may detect. As a consequence, it is thus important to observe that - regardless of the method which is pursued - the practice of history is less a science capable of determining how things essentially were, as von Ranke wished it to be, than an enterprise in which truth exists merely as a necessary, but forever unattainable ideal. The historian is thus perhaps much like a prisoner in Plato's cave. Condemned always to see the shadows of the world flicker on the wall, he will never know in what relation such shifting images stand to the true forms beyond.
History and Use, Revisited
Denied the facility to see anything more than a shadow of past truths and shorn of the capacity to recognise even those few sharp images which are granted to him, the historian is simultaneously divested of use.
As we have seen, those scholars who have attempted to suggest that history can act as a guide to the future have based their arguments on the assertion that the discipline has the capacity to uncover underlying truths about past occurrences. While it may be granted that something can be known about certain events in the past and that this can indeed be classed as a form of truth, it would be invalid to conclude from this that this constitutes the foundation of any meaningful predictive power. Just as the present is more than a series of events, so our construction of the past must aim to be more than the enumeration of happenings. At root, history remains a study of human beings, of men and women throughout the ages with thoughts and feelings and impulses, and not merely the analysis of actions. What was done, what happened is only the visible emanation of the ideas and emotions from which they sprang, and in that it seeks to understand the past, the substance of history ultimately lies in the reconstruction of these thoughts and sentiments. Yet, while the historian has at his disposal traces of a portion of these thoughts and sentiments, the very fact that such traces are cloaked in language limits his ability to discern their true nature and further compromises his capacity to recognise if he is indeed accessing that underlying truth. Without reliable access to the true identity of the motivations and thoughts of past actors, historians of all varieties are unable to posit anything more than the most general connections between the remnants of the past. Consequently, it seems clear that any attempt to derive predictions about the future, or conclusions about the root nature of humanity from such general connections is not merely a fruitless, but perhaps even a dangerous enterprise. History, then, is without use.
If history is indeed use-less, however, what is its value? Are historians condemned to being 'merely ornamental', in Charles Clarke's words?
In the sense of its relevance to society generally, I freely admit that I see no value in history. Beyond the tenuous claim that it is somehow 'good' to know something about the past, and the suggestion that some things somehow 'deserve' to be remembered, it is difficult to see how an enterprise based on so fragile a relationship with truth can possibly be described as 'valuable'. Yet while I would grant that history is socially valueless - and am, in fact, rather proud of this fact - that is not to say that I do not believe that history has a value at all. On the contrary: I am firmly of the opinion that, insofar as the individual historian himself is concerned, history is of immense value.
At one level, the practice of history carries with it a degree of self-analysis that leads to almost unparalleled intellectual liberty. This is a consequence of the nature of the quest for an eternally elusive truth.
In attempting to reconstruct past thoughts in conditions of continual uncertainty, the historian is obliged constantly to engage in a struggle not just with doubt regarding his sources, but also between empathy and objectivity. Confronted with documents which conceal intentional meanings which are perhaps always inaccessible, the historian is required at every turn to question precisely what conclusions the words before him may support. In interrogating the language of his documents, however, the historian is simultaneously required both to attempt to place himself inside the mind of the author, and to retain some degree of scholarly objectivity. The tension which this generates creates the need for the historian always to ask himself why he thinks what he does in response to a particular document: does he favour a particular interpretation because the language itself truly supports such a reading, or does he favour it because - in trying to place himself inside the mind of the author - he is unconsciously bringing to bear more obscure and less easily-defined attitudes of his own? Is he, indeed, asking the correct questions of his document, or is he interrogating it in a manner which is not merely anachronistic, but even the product of some other, hidden impulse?
In searching for an eternally inaccessible truth in this way, the historian is forced to question himself as much as, if not more than the documents themselves. Indeed, every aspect of the historian's outlook must necessarily be brought into question. Unlike other fields of endeavour, such as the study of law, the practice of history requires that there are no boundaries to what can be queried. Operating in conditions of perpetual uncertainty, everything the historian may think - and everything that other historians have thought before - has the capacity to be challenged, and there is something close to an obligation to defy any given interpretation. If it is to be made more perfect, more satisfying and complete, the temple must ever be torn down anew.
In that it requires each and every assumption, interpretation and question to be challenged, the practice of history is thus a process of mounting intellectual liberty. Anything that can be thought, anything that can be conceived can and must be questioned; nothing, but nothing, is sacrosanct. There are no totems, no holy images which cannot be brought down. There are no inviolable principles, save one: the absolute, over-riding obligation to question everything, and the self most searchingly. Once self-imposed limits to thought are removed by historical study, anything - outside the field of history - is possible: nothing becomes unthinkable. History, then, is intellectual liberty.
At another level, however, history is a true humanism. Although there is a humanist undertone in the notion of history as intellectual liberty, the humanistic value of the subject is described most fully by its very uselessness, and in this, it is necessary to move beyond the discipline itself into a wider realm.
It is perhaps one of the most difficult things imaginable to define what it is to be human, but for me, the essence of humanity lies in the useless; that is to say, in those enterprises which have no absolute use. While we are unusual in many ways, most of the characteristics which are often held up as intrinsic to human identity are, in fact, shared by other animals. Although perhaps to a lesser extent, other creatures share our ability for logical reasoning, our capacity to nurture and educate our young, and our complex social structures. Yet there is one characteristic which is uniquely ours. Of all creation, human beings are the only animals which have both the capacity and the willingness to pursue enterprises which transcend function. Peacocks may develop fine feathers, chimps may use rudimentary tools, and meerkats may tutor their young, but all such attributes are strictly related to a particular function associated with survival above all else. Humans alone indulge the useless: we, uniquely, cultivate those enterprises which contribute nothing to survival.
History has no use: without truth, it is useless. It can leave us no richer, no poorer, no happier, no more content: it cannot build higher skyscrapers, salvage economic collapse or affect medical advances. It contributes nothing to our survival as individuals, or as a species. Yet, precisely because it has no function, it is a valuable affirmation of our humanity. In practicing history, we continually restate our basic human identity, proclaim our uniqueness, and declare our status as human beings. In that it plays always with uncertainty, history remains always a game without function, but in that such a lack of function confirms our humanity, it is indeed a beautiful game.