This essay was published in the TLS of September 26, 2008

Most historians are familiar with Samuel Butler's remark that "though God cannot alter the past, historians can; it is perhaps because they can be useful to Him in this respect that He tolerates their existence". Most, too, have been asked why their own views of the past differ from those of other historians. Leaving divine revisionism aside, we tend to refer to the discovery of new evidence, the selection of past evidence, the development of new ways of doing research, different perspectives on agency and the creation of evidence, and the varying conceptual frameworks within which historians operate. Given enough time, the explanation might move on to a brief discussion about the unattainability of objectivity, and postmodernist perspectives on past hierarchies and attitudes, as well as modern understandings of what evidence is understood to be true, representative, or "right".

These factors all relate to history as a "dialogue between the past and the present". This concept of an epistemological "dialogue" was first outlined by R. G. Collingwood in The Idea of History (1946): that history is neither "the past by itself" nor "the historian's thought about it by itself" but "by the two things in their mutual relations". This was succinctly paraphrased by E. H. Carr in What is History? (1960), where he wrote that history "is a continuous interaction between the historian and his facts, an unending dialogue between the past and the present". Subsquently this idea of an epistemological "dialogue" has been adapted by many writers, including E. P. Thompson, Dominick LaCapra, and Adrian Wilson. It has a long and impressive track record. But this dialogue between past and present is only one of several in historiographical development. A second dialogue - a social one, with no direct reference to the past - takes place between historians and their contemporary readers. Obviously this incorporates changes in langauge and social attitudes. Times change: those reading a modern history book do not expect to see historians display certain prejudices and judgments from a few years earlier. They do expect certain historians to display other prejudices (for example, academic historians should eschew dramatic or emotive language, even when describing a dramatic battle or a romantic betrayal). Subjects of public interest wax and wane, and with them fluctuate the fortunes of the writers of scholarly as well as popular history.

These epistemological and social dialogues represent two forces for change in history. Indeed, they both demand that views of the past change, and in that sense of "change" they represent one form of "originality", that is to say, they constitute the development of existing ideas. The assimilation of new facts continually causes the historian to revise and modify his or her theoretical understandings of the past. At the same time, the social dialogue between historians and society is no less an agent for change. Politically it becomes more or less dangerous to write about certain subjects in certain ways (for example, the Holocaust, the slave trade, or the genocides committed by the past leaders of one's own nation). Similarly the financial rewards of writing about certain individuals and themes alter. Much of the change, and thus much of the variation between views of history, is a complex mix of these two dialogues which involve the three parties that matter both epistemologically and socio-economically: the past, the historian and the reader.

These forces for change are not the whole story. If they were, originality in history would be simply a reactive process in which historians respond to new discoveries and to shifts in public opinion, political control, or scholarly debate. This is a cynical view of history: it assumes that historians only write something new to please a certain trend in popular culture, or to jump on a scholarly bandwagon, or as a consequence of their job description. But at the heart of some historical work there is a quite uncynical and very powerful self-questioning. It amounts to a third "dialogue" - a philosophical or conceptual one - this being a writer's recognition of some aspect of human nature within his or her own self, or within his or her own experience. Alternatively it might be referred to as "vision", or even "character". Ultimately it results in an idea or set of ideas which is not rooted in past evidence nor in an awareness of the historian's potential readership, but in the historian's own understanding of humanity.

At first sight this is a subjective platform, and has little or nothing to do with history, being the very antithesis of the cool objectivity traditionally praised and pursued by the academic side of the profession. On further reflection, however, it seems to be the root of true originality in history writing. The first two dialogues are both responses to factors external to the historian. In the epistemological dialogue, it is an engagement with the evidence which causes a historian to revise his or her understanding and theoretical framework. Once modified, the revised framework allows further evidence to be understood more clearly or fully, and the framework to be revised still further. In the social dialogue, it is the historian's engagement with contemporary society which causes him or her to write in a certain way, and to revise the language, attitudes and literary style employed. The variations and novelties in historical understanding and expression which result from the two dialogues are routine consequences of the historian doing his or her job along prescribed, professional lines. One may draw an analogy with chess: every game of chess is unique, and thus original, but each game is original in a routine way, for it is conducted according to a single set of rules. The invention of the rules of chess, by contrast, was truly original.

This distinction between "routine originality" and "true originality" in history may be likened to T. S. Eliot's distinction between "spurious originality" and "true originality" in his introduction to Ezra Pound's Selected Poems (1928). In Eliot's view, "spurious originality" arises when the poet writes a new poem but derives the form and perhaps even the substance from other poets. It is essentially derivative - the writer develops what has gone before for the sake of continuing a tradition. "True originality", according to Eliot, comes from the writer who goes direct to life - his or her own experiences - and tries to describe what he finds outside literature and then brings it within a literary form. The obligation on historians to be mindful of previous work on the period under study, and to incorporate previously discovered facts, means that the term "spurious originality" is inappropriate: much of what Eliot identifies as "derivative" in poetry is simply good practice in history. However, in most other respects, "routine originality" and "true originality" are distinct from one another in history just as Eliot's "spurious originality" and "true originality" are distinct from one another in poetry. In both respects it is those writers who are "truly original" who have the most significant and profound effects on society.

It is nothing new to say that a historian's own life experiences condition the history that he or she writes. Carr, for example, exhorted his readers to "study the historian". But there is a more urgent need to distinguish between true and routine originality in history writing - or between historiographical revolution and mere historiographical development. In the Da Vinci Code trial, when two of the authors of the popular history The Holy Blood and the Holy Grail accused the novelist Dan Brown of appropriating their ideas, the very possibility of historical originality came under attack - not in the courtroom but among the journalists who covered the event. There was a widespread view that historical "facts", including ideas, cannot be original at all. No copyright can pertain to a historical fact for no one can copyright what is perceived to be an element of the past. One journalist at the time (Joel Rickett, the then-deputy editor of The Bookseller) went so far as to say that Michael Baigent and Richard Leigh, in pressing their case, were "admitting their work has elements of fiction to it. If it was pure history, how could they copyright history? When historians discover something they can't copyright it". The implication is that members of the public may reasonably divide what they see as facts (presumably that is what is meant here by "pure history") from historical originality, and that the latter may be assumed to be fictitious, as it stems from the historian rather than directly from the evidence.

But true originality in history writing does exist. It is not necessarily fictive, or fictitious; it is philosophical and conceptual, and sometimes poetic. Consider the following four historio-conceptual names. "The Hundred Years War" is a term which would have meant nothing to those fighting in it, least of all Edward III, who started it. The word "Renaissance" would have perplexed Leonardo. The only reason why people in 1349 would have understood what you meant by "the Black Death" was because they had very recently lost a large number of their friends and relatives and were themselves going greyer by the minute.

"Bastard Feudalism" is not a concept you could have discussed with John of Gaunt. All these terms are concepts developed by historians in later centuries to explain an aspect of the past. The "factual" basis of each term is nothing more or less than a historian's desire to describe the past in a way which has both meaning and resonance. And because they are modern inventions these concepts are just as "truly original" and as distinct from the past itself as the sentiments expressed in Seamus Heaney's poems about the Tollund Man or Andrew Motion's "Anne Frank Huis", which are also descriptions of individuals who actually existed.

Names as representations of historical concepts illustrate how modern originality can come to be universally associated with the past and yet not be fictive. No one would refer to the Renaissance as either fictional or imaginary. But what needs to be stressed is that it is not just the name which is truly original but the concept it embodies. The "Renaissance" is not just a name for a period, it embodies the nineteenth-century idea of the rebirth of scientific achievement (later extended to artistic and cultural achivement) after the Middle Ages.

Furthermore, the concept of the Renaissance has come to represent the historical years to which it relates in much the same way that "history" has come to represent "the past". True originality can be much more than just a poetic quip which happens to prove popular. It can be a way of envisioning an entire period, and perhaps even a way of envisioning the entire human past.

Consider Keith Thomas's classic Religion and the Decline of Magic (1971), which presents the idea that religion and superstition provided explanations for, and antidotes to, the adversities of life in pre-industrial society. These explanations and antidotes, Thomas argues, only declined as scientific discoveries made them experimentally unsustainable and socially unnecessary. Crammed with useful and fascinating details, it is "truly original" in many respects. Most importantly, it is truly original in its juxtaposition of science and superstition as diametric opposites, and its central thesis, that rationalism and unsubstantiated beliefs were long in conflict with each other. The fact that that central thesis may be deeply flawed - on the grounds that science and faith are not necessarily diametrically opposed, and religion is not always in conflict with science, and that the divine nature of Creation inspired many people to follow scientific paths, and to experiment with elements of Creation as a religious exercise - is unimportant in assessing the book as work of true originality. Even if its central thesis were to prove unsatisfactory in some respects, it would not become "fictitious" or fictive. It stands on a list the TLS published in 1995 of the "hundred most influential books published since the war", and rightly so, for it is a hugely informative and enlightening view of our culture in the early modern period.

Many other books on that 1995 list (drawn up under the auspices of the Central and East European Publishing Project) are also examples of truly original historical works. R. G. Collingwood's The Idea of History reconsiders the nature of history through the author's own awareness of the process of creating a historical thought, and what others have done in pursuit of knowledge about the past. This process is not one of "routine originality", for the whole idea of examining the historiographical tradition entailed Collingwood stepping outside it.

At some point he must have realized that his own understandings of history and ways of describing the past were different from those of his colleagues and predecessors. Hence the originality of his work is not derived not from any evidence or pre-existing debate, nor his desire to correct or applaud others, but his own vision, insight, character and confidence.

While historical works concerned with the philosophical foundations of history are very likely to be truly original, the same does not apply to those which adopt (consciously or unconsciously) a philosophy of history from a pre-existing debate, school or tradition. Nevertheless, historical studies (unlike poems) can be significant landmarks in the historiography of their chosen subject even if they are routinely original throughout. A classic example is J. H. Wylie's four volume A History of England under Henry the Fourth (1884-98). A colossal compendium of information about the reign, it includes almost no theoretical explanation of the forces at work in it, no account of agency, no understanding of the character of the King, and almost nothing of the philosophy or character of the author, except his incredible determination to gather and coordinate all the facts he could to do with the reign. As a result, scholars still often refer to it, even though no one has a good word to say about its conceptual framework. This is not to say it contains no analysis: it contains many examples where Dr Wylie found a conflict of evidence in his sources and sought to establish which was correct. On top of these there are many more examples of inferences about why the King was in a certain place on a certain day. But in these matters the originality is entirely routine and derivative in form as well as content - scissors-and-paste history, as Collingwood would have called it - being a mass of quotations from primary sources and paraphrases of key passages.

Wylie's History serves to show why routine originality does not prove influential. Its scope is too limited; it is has no implications outside the subject of England under Henry IV. Thus it is no surprise to see that all the other history books on the TLS "most influential" list are works of true originality. Michel Foucault's Madness and Civilization (1961) is a striking example, correlating the rise of madness with the decline of leprosy in Europe and the emptying of the leper houses, allowing medieval people to locate other unwanted people within these social and physical spaces of exclusion. This is not a thesis which is in any way routinely original: its originality rather lies in the philosophy that one can juxtapose and explore one form of social exclusion in relation to another, and as a feature of human behaviour in all ages. Arthur Toynbee's A Study of History (1934-61) is hardly less original for its emphasis on world history patterns and the rise and fall of civilizations. In each case the third, philosophical "dialogue" of the historian results in a new way of understanding the development of the human world, or at least an aspect of it. As with Religion and the Decline of Magic, even if these bold statements are wrong, they are still "truly original", and influential in their time. And most certainly they are not fictitious. A fact might be false, an explanation might be wrong, but fiction cannot be sustained accidentally by the historian any more than history can be sustained accidentally by the novelist.

In this light it is somewhat worrying to realize that Religion and the Decline of Magic is the second most recent of the dozen or so historical texts on that TLS list, even though it was published as long ago as 1971. The only other historical text first published in the period 1970-95 on that list is Paul Kennedy's The Rise and Fall of the Great Powers (1987). History is the second most popular single theme in the first half of the period (after fiction). But even if the list were to be continued to the year 2005, with another twenty titles for those extra ten years, which history books might be added? Perhaps a space might be found for Eamon Duffy's Voices of Morebath (2001) or Orlando Figes's A People's Tragedy (1996). But it is difficult to think of many other historians who would appear on a "most influential" list. It is much easier to think of important works which were left off the 1995 one, and, apart from Simon Schama's Citizens (1989), they mostly date from the years before 1980 - for example, E. H. Carr, What is History? (1960); Steven Runciman, History of the Crusades (1951-4); Peter Laslett, The World We Have Lost (1965); Hayden White, Metahistory (1973).

The history titles that do appear on the TLS list are all startlingly and truly original. However, although true originality gave historians a large measure of influence before 1971, it seems this has not been sustained. The explanation may lie in the series of critiques which history as a discipline received from postmodernists and critical theorists: few historians were able to turn these critiques to their advantage, lacking either the philosophical ambition or the literary skills. Also in those same years history declined in popularity in UK universities - from a position third only to medicine and law in the 1970s to one well outside the top ten. An optimist might respond that just as many history books are published now as in 1970 - in fact many more are - and history departments in universities and bookshops alike are never empty. But it seems undeniable that history is no longer the strategic intellectual command centre it was before 1971.

It is in this context of decline that everything I have argued above about true originality comes to take on a more significant meaning. It is hardly surprising that in the modern world, it can be assumed that true originality on the part of the historian must be tantamount to writing something fictitious. It has been all but banished from higher education - hunted down by the trappers of postmodernism and excluded from its natural habitat by the financiers of the research councils.

The vast majority of university-employed historians are dependent on teaching students, continuing academic research traditions, following institutional initiatives, and meeting funding bodies' priorities and selection criteria.

Applications for research are structured in such a way as to maximize the chances of success - by appealing to a grant awarding body's guidelines. Paid historians are no longer paid to be truly original; free thinkers are expected to work for free.

One might argue that the rituals of academia are so deeply ingrained now that many people cannot distinguish what free thinking in history amounts to, or even what history is, outside academia. Academics are paid to teach, to administer their departments, to research and publish, and to be experts in their prescribed fields. They are not required to produce works of true originality, or even to try to do so. It is simply a happy coincidence if they do. We do not fund historians to use their imaginations, but to do a job. That some "truly original" books do still appear - as the Wolfson History Prize admirably testifies - is almost entirely due to the fact that a few dozen academics are more powerfully motivated by their ideas than the scramble for research funding and, outside higher education, a clutch of scholars are prepared to risk financial insecurity in return for intellectual freedom.

Historians can, if they want, get through their whole careers by being nothing more than routinely original. It is hardly a crime to be a researcher and a teacher, revising this theory or that series of facts within an academic tradition. Similarly popular historians, if they so wish, can choose to do nothing more profound than reflect popular tastes in their stories about the past. Those who fall into either of these two categories are still performing the admirable function of imparting knowledge, but if they succeed in inspiring anyone at all, it will be because their subject is inspirational, not them. It is only those

historians and scholars of true originality who have a significant and influential part to play in modern society, for only they can put into their works something which is rooted in life, not evidence. Through them, people may come to understand the human past differently, and what mankind has done differently, and thereby achieve a new vision of what mankind is. From that standpoint, famously outlined by Collingwood, true originality in history may even allow us to break the barrier of the past and affect the way we develop in the future, as Orwell predicted.

Poets may not be the only unacknowledged legislators of the world.