Firstly, giving reasons for your own choices is something a bit different from explaining events over which you had no control. I'd rather concentrate on the latter cases.

On the other hand, if you're trying to teach/explain why something happened in a non-contentious atmosphere, then it may be reasonable to omit facts that are unimportant to maintain coherency and avoid getting bogged down in clutter that doesn't matter to the overarching point.

Sure. I do think that Diamond should have provided pros and cons (and Pinker even more so). However, this discussion has been very useful to me in that I realize that others have different intuitions.

Suppose that there is a convention which says that it's ok to omit countervailing factors (which I'm not sure there is). In that case, Diamond and Pinker are at least to some extent forgiven - they are merely following a convention. However, then I'd say that the convention should be abandoned, because I, as a reader, do want to get the full picture, and not a cherry-picked selection of factors.

--

Even in the example with the teams coming very close to a tie, the response to "why did they win" may have been correct, in that they had all of those factors in their favor and that those were enough to win (barely). I agree that in this case the guy was biased, but on the other hand they didn't ask him "what factors were involved and why did they favor Man City (somewhat)?"

Sure. The question kind of suggests an answer along these lines - which support the notion that there is a convention according to which we're expected only to give pro-reasons. But in this case, this is clearly absurd. Why would I want to hear about Man City's fifth best characteristic (relative to Man U) over Man U's best characteristic (relative to Man C)?

--

Generally, I also think that people have a bias against admitting bias, especially in themselves, but also in others. Philosophers and others have defended the hypothesis that humans are basically rational to absurdity, to my mind, despite the overwhelming evidence to the contrary. (See this excellent post on this, for instance. But the same goes not only for the notion of whether humans are rational in general, but also for whether any particular thought-process was rational or biased. I think that, given what we know about the ubiquity of human biases (as shown in experiments), we should be very open to the possibility that even great books like Diamond's and Pinker's are filled with inferences which are not rational but rather due to biases. This goes particularly for inferences and reasonings in messy and inexact fields. I would be very surprised if someone was able to escape to form very general historical theories such as these without falling prey to at least some biases on a major scale.

That's about all I'll say for one response - I have a bad tendency of rambling on when I've already made the points that I really wanted to make.

Me too - my posts tend to be too long for LW. I think it's a good idea to try to be a bit shorter and more succinct for both of us.