[H]istorical narratives 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 the sciences(*)
This should be qualified somewhat, but can be stated perhaps even more simply: History is written to persaude, historical fiction is written to entertain. All else is form following function.
Lester Grabbe cites this passage in his Ancient Israel,(**) and follows this with several pages of discussion of postmodern criticism of historiography. Despite a too sympathetic nod to Evans' In Defense of History (1997), the discussion is at least reasonable.
It is perhaps worth noting that both the discussion and the sympathetic nod are symptomatic of a problem in history generally: Historians do not care about epistemology. Evans' book was well received by practising historians, who need any port in a storm to avoid the elephant in the room, but panned by philosophers of history. See, for example, Easthope's disgust with the volume, available online.
Grabbe himself notes this shortcoming:
In the end, however, one can only agree with the observation that 'the majority of professional historians ... as usual, appear to ignore theoretical issues and would prefer to be left undisturbed to get on with their work while no doubt hoping the postmodernist challenge will eventually go away' (Zagorin 1999: 2). This certainly fits the attitudes of most historians I know in my own university who seem to have little interest in the debates on theory. (p28)
After (not surprisingly) rejecting the postmodern paradigm (and, it should be noted, paying particular attention to the problem of subjectivity), Grabbe moves on to discussing some basic principles of historical inquiry. Most of his tenets are not only reasonable, but quite good, and I might have more to say about them in a later post. His sixth principle, however, should probably be split into two. Lest I rob it of context, I'll cite it in its entirety, with the point I think the separation should be made set off in red:
All reconstructions have to be argued for. There can be no default position. You cannot just follow the text unless it can be disproved (sometimes expressed in the nonsensical phrase, 'innocent until proved guilty' - as if the text was a defendant in court; if there is a forensic analogy, the text is a witness whose veracity must be probed and tested). The only valid arguments are historical ones. Ideology, utility, theology, morality, politics, authority - none of these has a place in judging how to reconstruct an event. The only argumentation allowed is that based on historical principles. Naturally, subjectivity is inevitable in the process, and all historians are human and have their weaknesses and blindspots. This is why each must argue for their viewpoint and then subject the result to the judgement of peers, who are also human and subjective.(p36)
Looking at the red portion, did you spot the problem? If you were thinking in terms of theory you may have. If you were more interested in just getting on with "doing history," you almost certainly did not.
I checked several reviews of Grabbe's book. Nobody mentions what is, to me, an obvious oversight. One which makes it clear that, despite his citation of both, Grabbe actually doesn't understand the criticisms of White or Jenkins as pertains to subjectivity. More importantly, despite explicitly criticizing his peers, Grabbe too only gives a cursory look at theory.
To explain, we'll take a look elsewhere for a moment. Imagine I gave you a paper with such a radically feminist view that we would, today, consider it flagrant revisionism. Then imagine I told you that this paper had been very well received in peer-review, despite what seem to be obvious problems.
Now I ask you to name the decade it was published in.
You would almost certainly guess the late 1960s, or more probably the 1970s. And you would almost certainly be right. This is a fairly flagrant example of what is generally a far subtler problem.
The implication in Grabbe's principle is that peer-review, conducted by reviewers who are "human and subjective," will ultimately create a balance. A sort of levelling off of the blind spots. And it might; for the blind spots of the individual.
But those aren't the only blind spots. There are also collective blind spots. Blind spots borne of context, a context that is extremely likely to be shared by the huge majority of peer reviewers. And those cannot be balanced, cannot be identified, cannot even be seen until that context is gone. That might be a span of a few years, such as our example. It might be a span of decades, even centuries.
We might suggest that a historiography that allows for this--that requires both peer-review and survival of multiple contexts (as demonstrated perhaps by longevity), but this would drop our list of established historical "truths" to somewhere in the neighbourhood of zero, and would suggest that what we produce today cannot be considered right or wrong by anyone reading it now.
Even this optimistic suggestion, in other words, says that Hayden White was at least functionally right after all.
(*)White, H. (1978). Tropics of Discourse: Essays in cultural criticism. Baltimore: The John Hopkins University Press, p82.
(**)Grabbe, L. (2007). Ancient israel. New York: T&T Clark, p27