Surely, however, if an earlier source disagrees with a later one, it is more likely that the earlier source is better. NOT that is is NECESSARILY authentic, I stress.This is the same principle as that by which earlier manuscripts are preferred - a scribe is more likely to introduce an error than to correct one, there being very many more possible wrong ways than right ways (one), of copying a text.
Sort of. Historians do prefer sources as close as the event as possible and as unredacted as possible. If you could find an account of the crucifixion written the next day, or even a report written in the same year, that would be BRILLIANT. So, yeah, if you're going to take something as a source at all, indeed you'll prefer the early one. With a bunch of caveats, of course, but generally that's indeed so.
But the problem for HJ is exactly what you yourself stress: an inconvenient early source isn't necessarily saying the truth either. The earliest known version of Cinderella, dating all the way to ancient Egypt, doesn't necessarily point at a historical Cinderella. That's why even for early sources, we want multiple sources, we want sources without a vested interest, etc.
Is it not then rational, having identified a text as earlier, to prefer it; particularly if its contents are more plausible than those in later sources? E.g. insane preacher with brothers, as against son of the Holy Ghost sprung from the womb of a virgin. This preference should apply even when there is no contemporary historical evidence for the existence of the character portrayed. Brainache makes the same point with his progressive "morphing" of John the Baptist, noted by very many commentators. Literary device or not, the principle is a valid one, it seems to me.
It's only rational if you've already established that you can take something as a reliable source. THEN you can use the criterion to guess which are the more reliable sources. But, you know, that's an extra IF there.
Otherwise it's like finding a historical Cinderella in that Egyptian story.
In your Wallace example, we already know that he existed, and we have several corroborating sources for those events. We only use that criterion to guess which sources are more reliable than other sources about some details, but we already established that we can trust that he existed and did certain things.
In the case of HJ, that step is skipped, and replaced with BS sophistry. And it's kinda a crucial step.
The usual one being the argument from personal incredulity (yeah, it's named that even when it's actually an argument from credulity) which basically takes something for true or likely true, just because it's not unbelievable. E.g., "it's not unbelievable that a preacher named Jesus existed, and did X, Y and Z". So what? The veracity of a statement has little bearing on its actual probability to be true.
And another one is sneaking arguments from step 2 as arguments for step 1, although they were never meant to have any relevance there. An argument which is used in step 2 to guess which sources are MORE historical, once you've established they're historical at all in the first place, is snuck in as an argument in step 1 for why one can take them for historical in the first place. And it just doesn't work for that. In fact it fails so badly that it would make Darth Vader and Little Red Riding Hood historical too, if used that way.
Argument ad Google.
.
