Opiniatrety

Evidentiality Forever

I was just discussing the game Persian Monarchs in the works of P.G. Wodehouse. I remembered its appearance in "The Smile that Wins," but someone else pointed out that it also plays a role in Uncle Fred in the Springtime, which I'd forgotten. Then I was asked a question about a line involving Persian Monarchs, and said:

(1) That sounds like it must be from Uncle Fred in the Springtime.

Note two evidential markers here: 'sounds like' and 'must'. It might be thought that these are redundant, since 'sounds like' already implies a certain indirectness of my evidence. If I remembered that particular line from UFitS, I wouldn't have said 'sounds like'. But, in this case, what I was relying on was the fact that it didn't sound like it was from "The Smile that Wins." (I don't have TSTW memorized, so I have to go on 'sounds like'.) If it's not in TSTW, it must be in UFitS, since that's the only other work (so far as we know) that mentions Persian Monarchs. But I don't have direct evidence even that it sounds like it's from UFitS; I had to infer it. Hence the double evidential marking.

Of course, my intuitions on this are completely polluted. For the record, in the conversation I first typed it as (1), then changed 'must be' to 'is', then changed it back to 'must be' because I feared I'd be claiming too much familiarity with the role of Persian Monarchs in UFitS. And because I wanted to write this post.

Posted by Matt Weiner at August 4, 2006 02:35 PM