History is best done by historians, argues Arthur Marwick, who are often blind to the reasons why they write what they do, replies Hayden White.

Marx excoriated the philosophers of his day for merely seeking to understand the world; to him, the problem was "to change it''. In these three little words lies the key to the central academic debate between, loosely speaking, post-modern and empirical approaches to knowledge. The debate runs and runs, as in the good-natured (I hope) exchanges between myself and the Californian academic Hayden White (and some British disciples) which began in October 1993 at the Open University.

We must always start from the two fundamental questions (for robust intellectual health, ask them three times a day after meals): why study history; and, why do historians think they know what they think they know? We study history because of the desperate importance of the human past: what happened in the past (often the very remote past) governs the world we live in today, and created the many problems which beset us. It follows that we need a history which is as accurate and thoroughly substantiated as mere mortals can make it. For, as I have insisted over and over again, "history'' is not the past itself, it is "the past as we know it through the work of historians''. Knowledge of the past is not something that simply exists (for, say, philosophers, or political propagandists to fool around with); we only have knowledge of the past through the activities of historians.

Why should we regard this knowledge as reliable? There have been many myths about the past, many inadequate accounts, constructed to serve, say, religious or political ends. Professional history only began to develop towards the beginning of this century: it is the duty of the professional historian to get it as right as is possible (as it is the duty of the scientist to get her science as right as possible). There are many reasons for historians falling short of their duty: career pressures aligning them with one particular thesis, desire for adulation in the posh Sunday press, sheer incompetence. Above all, as historians themselves are better equipped than anyone else to understand, historians will inevitably be influenced (not controlled) by the attitudes and values of the culture within which they live.

ÖйúAƬ

All each individual historian produces is contributions to knowledge, tentative and fallible, which will be attacked, debated, qualified and amplified by colleagues. As with all bodies of knowledge, the production of history is cumulative. Sure, what eventually gets through to students and to laypersons is very much under the tutelage of the community of historians. How could it be otherwise? Better, surely, history produced by historians than history produced by ballet-dancers? Football, after all, is "what (fully recognising the social and career pressures) is played by footballers''. That, too, is as it should be (certain English teams excepted).

The past has passed, it has gone for good. But it has left an immense variety of relics and traces, the primary sources. The only way of producing knowledge of the past is through the study and interpretation of these unforthcoming and fragmentary sources, a task calling for complex professional skills (which have to be learned, since we are not born with them) and great powers of analysis and reflection. The critics should try it -- but they never do: they prefer to sit on their backsides and conjure up utterly misleading accounts of what historians do, conveniently shaped to their political purposes (radical socialism, radical feminism, radical gay rights).

ÖйúAƬ

And so I return to the nub. To change the world, we have first to understand it. It is those who are hell-bent on change, not on the basis of hard-won knowledge, but in accordance with a few metaphysical speculations, who cause the havoc or would if they could (all they actually manage are their puerile attacks on history -- or science). By "neo-Marxists'' I mean those who have absorbed the shibboleths of post-structuralism. Most of the British camp followers were formerly unreconstructed Marxists, and some openly state their objective as the destruction of "bourgeois liberal humanist empiricism''.

Hayden White (professor at gorgeous Santa Cruz) is an erudite and compelling figure, standing somewhat apart from the battle (in his very essence a liberal Marxist, he explicitly condemns revolution). He is the most persuasive proponent of the assertion (you cannot call it an argument) that language operates to a set of rules, arising from bourgeois dominance and bourgeois desire, which are quite independent of individual human agency. White pays no attention to the problems of research, but imagines that events and "facts'' are somehow already known, and that all the historian does is "narrativise'' them. In "narrativising'', historians, he declares, will be unable to avoid the "tropics of discourse'' ("tricks of the trade'', to you and me) which makes it impossible for them to give any account other than one deeply contaminated by the values of bourgeois society.

While recognising that too many historians are extremely sloppy in their use of language, I believe that, given the discipline and the expertise, it is possible for historians to control their use of language, and always be precise and explicit. I stress that research and the writing-up of research form an integrated, iterative process, in which, contrary to the assertions of the post-structuralists (who, incidentally, get no support from those who really know, like Chomsky, Lyons and Pinker), language is used as a basic tool: only by trying out our thoughts and conclusions in written form can we refine and clarify them and relate them to each other.

The neo-Marxist post-structuralists retain from Marxism the delusion of "history'' as objective material process unfolding in successive epochs (currently, apparently, we are in the age of "post-modernity''), the arrogant ontological claims, the highly dubious concept of "ideology'', and the political programme, which instead of arguing over as citizens, they allow to permeate their academic work. In substituting slogans for meticulous research they certainly make history easier (why read all these books if they are merely "ideology"?). In making history subserve politics and metaphysics, they deprive it of its fundamental, and critically important, purpose: helping us to understand the past.

ÖйúAƬ

Arthur Marwick is professor of history at the Open University.

These articles are based on two much longer essays, which will appear in the January and April 1995 issues respectively of the Journal of Contemporary History. The debate will also be aired at the Social History Society conference in January and a video on the same subject will be previewed at the Institute of Historical Research on February 3.