David Greenberg’s review of Chris Matthew’s new Kennedy biography is, like everything Greenberg writes, worth reading. It’s a wonderful takedown, I think, because it’s not entirely captivated by the search for the perfect snark (they’re wily, snarks are, and should be hunted at dawn and dusk, when they typically rest).

That said, I wonder about these passages and what follows in a similar vein:

So completely does talking constitute Matthews’s raison d’être that he mistakes it for historical research.

There is nothing wrong, needless to say, with conducting interviews, and they can be invaluable for journalists trying to set down a first draft of history. But the farther one reaches into the past, the less precise and clear memories become; and while they can provide help when existing records are unreliable or deficient, they have to be used with care.

I don’t think Greenberg intended this as an attack on oral history — because, as he intimates, Matthews isn’t an oral historian in any meaningful sense of that title — but I do find myself reading the review as an endorsement of the idea that historians should, whenever possible, be using documentary records and, as Greenberg says outright, leaving interviews to journalists. I wish Greenberg had delved a bit deeper into this point, particularly the relationship between oral history and the fallibility of memory. The idea that written sources have more truth value, or perhaps a greater capacity to illuminate the past, than oral histories, interests me, but I’m pretty sure I’m less convinced by that argument than Greenberg appear to be.

My general rule is that all sources — written, oral, etc. — should be treated like hostile witnesses. Which is to say, I’m skeptical of anything my sources tell me and, as a result, seek context and corroboration whenever I can. I suspect Greenberg would agree with that proposition, so perhaps I’m making something of nothing.