I cannot agree this categorical distinction between historical documents and historical novels in terms of their evidentiary value.
However I do agree that the method by which one works with documents whether putatively historical or fictional is a key element in evaluating both types of documents for their historical value.
This value itself stems in large part from the quality and type of questions the historian is asking. Documents do not speak for themselves, historians do the "speaking". One should not ask all the same questions when evaluating a novel and a statistical document: we may perhaps pause before using a novel as a source of statistical data. On the other hand, the probative value of public records relies on the historian knowing what assumptions prevailed at the time and what meaning was attached to a certain corpus of "facts". What comes to mind is how would the balance sheets of banking houses and merchant shipping lines be understood today if one did not know the "story" of the Middle Passage or the Opium War. Contemporary novels certainly offer means of assessing the cultural assumptions prevailing at any given time and allow one to reconstruct potential narrativation of facts.
A second example occurs to me: what is the Gulag Archipelago? Is it a historical novel or is it a documentary? It is certainly not an official historical document.
Mit freundlichen Grüßen/ Sincerely/ Cordialement/ Atenciosemente/ Cordiali saluti
Dr. Patrick Wilkinson
Institute for Advanced Cultural Studies Europe
It is an older discussion: to describe or explain? I think Lukacs also contributed some insights here. However, I believe the debate about narration or description is somewhat akin to the distinction between counting and arithmetic. In order to count one has to assert that there is at least a putative number of defining attributes which justify the assertion that two items belong to the same category. That means there is always an implicit explanation (subsumption of attributes to define an entity) behind every description.
Historical research cannot escape this vicious circle any more than other sciences. Hence it is not the distinctive character of the method and the distinction between narration or description which settles the character of historical research. It is as someone has already said the subject "humanity" or "human history" that makes history a particular way of knowing. Beginning with the counting of the foremothers and fathers and showing what time and space (period and place) mean(t) to those from whom we derive our identities or culture.
Now there are those who understand humanity as driven by
forces or ideas like "progress" or "salvation". A careful
investigation shows that a certain kind of "humanity" is linked to
this way of "counting" the lives that have gone before us. Now
historians have shown that there are other kinds of "humanity" and
different attributes to be counted when the human is not solely or
ultimately defined by one narrative conclusion. This opens the
question as to whether "immediate history" is not really