The plot thickens

History is fiction, argues Alun Munslow, but that does not mean it cannot tell truths

What do historians do? When I was an undergraduate in the 1960s I read good-practice primers, among them E. H. Carr's What Is History? (1961), Geoffrey Elton's The Practice of History (1967) and Arthur Marwick's The Nature of History (1970). They argued that the Enlightenment legacy of empirical observation that permitted history to emerge as a discipline in the 19th century sustains it to this day.

History as a discipline has not, of course, stood still during my career. Its borders have constantly shifted as historians have borrowed methods from adjacent fields of study. History has subdivided, creating hybrids such as postcolonial and subaltern studies. But these developments have been accompanied by a growing identity crisis that questions history's very being as a discipline. What are historians supposed to do, and using what philosophical assumptions do they do it?

I learned from my undergraduate reading that the real past was discoverable, but that, unlike scientists, historians can never "prove" anything. We always have to infer the meaning of the past from the evidence of unique events. The past is reconstructable as history, however, only as long as we do not ask tricky questions or use too much "theory".

I learned that my historical narrative would correspond to the truth of the past if I followed the proper technical rules of evidence (comparison, verification and contextualisation) and relied on the tried and trusted empirical method. This was the historical method and it made history an artisan's discipline - the painstaking inference of the meaning to be found in the evidence. There was no need, therefore, to question its essential philosophical underpinnings.

I also learned what appeared to be some fairly obvious corollaries to this position: there is always a clear distinction between fact and value, history and fiction are not the same and there is a division between the knower and that which is known. I assumed our task as historians was to understand the connections between events in the past and express the meaning of those connections in a narrative form.

Today's crisis of identity centres on the refusal of historians to address history's problematic character. The historian may be best qualified in the "practice" of the rules of evidence, but once a substantial number of practitioners question the validity of history's inferential and referential nature, then, by extension, they are confronting its claim to be a discipline.

One of the most influential practitioners to break ranks with the "artisanal establishment" was Keith Jenkins in Rethinking History (1991). He argued that there was a difference between the past and history. This hit at the naive empiricist belief in a close association between the past and historians' narrative representation of it. Many historians now refuse to accept that our technical treatment of the evidence can overcome the undeniably subjective nature of our linguistic construction of the reality we supposedly discover back there.

In the wake of the intellectual developments of the past 25 years a new history is emerging. Historians are breaking out of the traditional paradigm by employing postmodern perspectives. Younger historians are now asking: "Can history adequately represent the past as it happened? History is confronted by its "failure" of representation, and its increasingly threadbare claims to truthfulness.

The question is no longer whether we can reconstruct the past as it really was, by inferring meaning from its traces, but whether any historical narrative can ever be accurate. Many "new-wave" historians in the pages of the leading journals have moved away from traditional empiricism to exploit philosophical theorising. They place a fresh emphasis on the nature of language in exploring the relationship of form and content in our sources, acknowledging the unavoidable relativism of historical understanding. History as a form of knowledge has become a contested terrain.

In 1974, the American Renaissance historian Hayden White posed his now famous question: "Why do historians persist in failing to consider historical narratives . . . as what they most manifestly are - verbal fictions, the contents of which are as much invented as found and the forms of which have more in common with their counterparts in literature than they have with those in science?" White's question was a time bomb. Most historians were then still dedicated to the principle of discovering "the truth" based on the presumption that the past existed as an external reality beyond our description of it. Thanks to White we are increasingly challenging that facile belief. We are witnessing the collapse of the old order of empirical history and experiencing the rise of a new understanding of what history is and what it can do.

Within history as a discipline our general condition of postmodern intellectual uncertainty has been focused by White's insistence that it is the historical narrative that creates meaning and knowledge, rather than the results of our reductionist trawls through the archives. Historians always write of the past in the form of a literary genre such as romance, tragedy, or farce, and we deploy figurative devices to create our meaning. Often the historian's concepts and categories are metaphorical. The past is not inherently tragic or farcical, but history probably is.

History is not a discipline that can generate truthful interpretations. The very notion is an oxymoron. What this means is that history should not be constructed exclusively as an objectivised empiricist enterprise. I am arguing for the new history primarily as a literary project, one that must self-reflexively take account of the narrative imposition of historians, while serving the evidence, as well as deploying analysis and argument. In other words, we should imagine history first as what it palpably is, a narrative of the past.

When I recommend the methodology primers of Carr, Elton, and Marwick to my undergraduates, it is to alert them that being self-reflexive about history is not an optional extra. The new history demands that we write the past fully aware of the inadequacy of our encryption and encoding. Just because the new history denies any necessary correspondence between knowledge and historical truth does not mean we have to embrace total relativism or become nihilists.

Equally we do not have to assume that fiction cannot tell truths. Nor less is it to accept that we can never know anything meaningful about the past. The new history does not mean we no longer have grounds for telling right from wrong, nor for claiming the Holocaust never occurred. What it does mean is that we must fully recognise the role of the historian as a writer of the past. The new history is not just about factual statements, it is about the historian's imagination.

The crisis of history today lies in our failure to come to terms with the fact that our access to the past is far more problematic than our traditional method allows. History, like the past, is a site under construction.

Alun Munslow is the author of Deconstructing History and editor of Rethinking History: the Journal of Theory and Practice. He teaches history at Staffordshire University.