(May 31, 2023 at 5:20 am)BrianSoddingBoru4 Wrote:(May 30, 2023 at 6:32 pm)FlatAssembler Wrote: And the methodology of comparative linguistics papers isn't even poorer? Virtually all comparative linguistics papers assume in their methodology that sound changes are regular, which is an assumption that's impossible to prove, and it's not obvious that it's true. As for the improper sampling... Well, like I've said, I think you could reject the entire linguistic typology with such rhetorics. Perhaps the pattern that most of the languages have a subject-first word-order (SVO or SOV) has to do with almost all languages coming from some unreconstructed proto-language with such a word order. So, that pattern could be an illusion, rather than a cross-linguistic tendency. Perhaps the fact that in most languages around the world the first-person and the second-person pronoun tend to start either with 'm' and 't', respectively, or with 'n' and 'm', respectively, has to do with nearly all languages around the world descending from two proto-languages with such pronouns, and we therefore have a biased sample and that pattern is an illusion. You can reject most of the facts from linguistic typology with "Perhaps we have a biased sample.".
But that's in no way comparable to what Kleck did. Sometimes, unproven (even unprovable) assumptions are necessary in historical research - ya gotta start somewhere.
But Kleck wasn't doing historical research or historical investigation. He wasn't constrained to make assumptions that may or may not be true. He deliberately chose parameters that would skew his research towards the results he wanted and relied primarily on subjective, pre-disposed witness statements to arrive at his conclusions. Any criminologist - except, apparently Gary Kleck - will tell you that witness statements are notoriously unreliable.
That's not pseudoscience - it's just bad science.
Boru
Well, like I've explained, as far as I understand the methodology of social sciences (and I have published a few papers in social sciences, so I know something about how social sciences work), if something is the only study about something (like the Gary Kleck's study is the only study that tries to estimate how many people are saved by guns each year in the US), we should generally accept its conclusions. There are exceptions, like when the conclusions are highly scientifically implausible (like the Thai study concluding that Moderna vaccine causes myocarditis in adolescents more often than COVID itself does) or when the methodology contains a fatal flaw (such as incorrectly calculating the p-value or not calculating it at all). Generally, if the study is not such that it shouldn't have passed the peer-review, and it is the only one, you should accept it. Complaining how unreliable witness statements are and discarding them, even if there is no better evidence, is not very scientific, in fact, it has more to do with the methodology of The Mad Revisionist than with science.
