As a medievalist, I'm sure you're well aware how many of the latest advances in medieval history have come precisely because we have been able to scry into sources like saint's lives for their literary reconstructions of their world, i.e. it matters less what "actually happened" than what people believe "should be represented." So postmodern literary analysis has its usefulness in constructing the "plausible" world of the past, but never the "factual/actual."

Hayden White himself illustrated the dilemma created by his own analysis when he pointed out that from a moral perspective it is unacceptable to deny the reality of the human horrors of the past, yet it is impossible in a historical narrative to establish objectively that it happened. What does one do with this knowledge? Can anything be real or true?

I myself come from a journalism background (also a constructed literary form), and I find the arguments over the science or the "real past" of history analogous over the arguments over "objectivity" in journalism. It's well known that "ultimate" objectivity does not exist, but one should consider the ALLUSION (perhaps its not a coincidence that this word sounds like "illusion") to the idea of objectivity is important. It is, of course, a platonic solution to a postmodern problem, but one taken because its opposite is straight political partisanship (e.g. "@#$% objectivity, I'm watching Fox News!!") which results in the segregation of ideas from each other rather than their engagement with each other, or paralyzing despair (e.g. "nothing is real! what's the point??").

The allusion to "a past that really happened" is an important distinction in the separation from a fiction "entirely constructed." Think of them as two fuzzy poles of a magnet. Neither will ever be perfectly ideal (sorry Plato), because we all know that fiction is equally constructed from examples of the real world, and the real world is constructed from its fictional tropes as well. After all, isn't this why Foucault preferred to refer to it all as "discourse?"

I mean this is the epiphany I gleaned after taking some writing workshops on non-fiction as well as fiction. It's ALL discourse, as the best fiction tends to not be "entirely constructed" but in fact, a "predictive social model" based on a "real world" that results in a world-consistent plausible conclusion. In which case, how are "historical models" any different from these "constructed" fictions?

This is why the allusion to a "real past that happened" is important. It's just that, like a lot of things in life, you have to take the existence of a real past on faith. I mean frankly, we have to take on metaphysical faith that the world exists at all, and that the inductive observation that underwrites all of modern natural science is a permanent fact of the universe.

I think ultimately, the community itself decides the "truths" it wants to hold dear. I mean this is how the natural sciences operates, with its "paradigms" (although comparing Thomas Kuhn and White can get even more complicated). Paradigms are whatever bodies of thought best explain the current batch of evidence. They will always be overturned when enough logical anomalies appear as to disrupt the efficiency of the paradigm.

(Although with that said, I'm also aware that this only operates in the natural sciences because there is a willingness to accept common epistemology that the social sciences don't, but that's a separate argument).

Its the ultimate simultaneous balance of subjective and objective. "We acknowledge that the world is infinitely subjective. But within our world, we creat our own norms, and our own objectivity."

And frankly, that is good enough for me.

I'm of two minds. On the one hand, I came to grad school already pumped full of White et al, and it wasn't so much a revelation to me as a basic principle underlying historical work. On a practical level, it has helped me to worry less about distorting my findings as I arrange them in a narrative because I'm already primed to take narrative seriously and see it even in the supposed discrete factual data that I'd hypothetically be "distorting."

At the same time, it's hard not to see some serious weaknesses in his work, particularly in Metahistory itself (if we choose to take it as a programmatic statement about history as a discipline, which I think White wants us to do). The basic argument revolves around breaking down supposed distinctions between history and philosophy of history, but the nineteenth-century authors he chooses were engaged in a rather different task from the historians that are working today. The kind of product a contemporary academic historian is expected to generate is one that avoids broad-strokes generalizations and is often tightly wedded to a specific source base and methodological approach, while answering questions that are generally more interesting to other historians than to lay readers. These things were not true of the works and authors White is analyzing in Metahistory. I would say that there is a narrative pattern that is common to generations in historiography, with pre-60s work being generally progressivist, 60s-80s work pitting the little guy (the working class, premodern urban culture, etc.) against the system (whether "progressive" or "conservative"), usually leading up to the former's defeat, and so on. But I think it's harder to pin something like that down for the stuff that's being published today. (Of course, everything will become clearer in hindsight.)