Since I occasionally make cryptic, sniping remarks about Mearsheimer's The Tragedy of Great Power Politics, I should make clear that there are at least a few things in the book that I find illuminating. One of them is his discussion of the situation that Japan faced in 1941. (Btw, the prompt for this post is a discussion about Pearl Harbor on an LGM thread that started off being about the Russo-Japanese war.)
He notes that in mid-1941 the U.S. applied a "full-scale embargo" against Japan, "emphasizing...that it could avoid economic strangulation only by abandoning China, Indochina, and maybe Manchuria." (p.223) The U.S. was determined that Japan should not dominate Asia or be in a position to strike the USSR, then on the ropes against Hitler's invasion. This left Japan with two bad, from its perspective, choices: "cave in to American pressure and accept a significant diminution of its power, or go to war against the United States, even though an American victory was widely agreed to be the likely outcome." (ibid.) The Japanese chose the latter course, he goes on to say, as the less bad of two very bad alternatives. That does not mean the decision was irrational, though it was an extremely "risky gamble" (ibid., 224).
Where I would part company (or so I assume) with Mearsheimer is in seeing the entire Japanese militarist-imperialist enterprise as irrational -- and also immoral and criminal -- from the outset. However, given that Japan's leaders at the time were committed to that enterprise and given the particular circumstances that they faced, their decision to go to war with the U.S. was not 'irrational'. Whether the specific decision to attack Pearl Harbor was a mistake is really a secondary question; the more basic question has to do with the decision to launch war against the U.S.
ETA: And it's the more basic questions that tend to get shunted aside or overlooked when discussions focus narrowly on specific strategic decisions.
2nd update: Googling "sagan origins of the pacific war" brings up results, including Scott Sagan's 1988 article that M. cites (although my browser didn't like the pdf) and a 2010 paper at academia.edu by a UCLA grad student ("Revisiting the Origins of the Pacific War") that appears, on a quick glance, to take M.'s view, more or less. (But I've already conceded in the comments that TBA could be right.)
3rd update: For a somewhat different take on this, see John Mueller, Retreat from Doomsday (1989), pp.229-30.