What is wrong with “Traditional Rationality”?

In sev­eral places in the se­quences, Eliezer writes con­de­scend­ingly about “Tra­di­tional Ra­tion­al­ity”. The im­pres­sion given is that Tra­di­tional Ra­tion­al­ity was OK in its day, but that to­day we have bet­ter va­ri­eties of ra­tio­nal­ity available.

That is fine, ex­cept that it is un­clear to me just what the tra­di­tional kind of ra­tio­nal­ity in­cluded, and it is also un­clear just what it failed to in­clude. In one es­say, Eliezer seems to be say­ing that Tra­di­tional Ra­tion­al­ity was too con­cerned with pro­cess, whereas it should have been con­cerned with win­ning. In other pas­sages, it seems that the miss­ing in­gre­di­ent in the tra­di­tional ver­sion was Bayesi­anism (a la Jaynes). Or some­times, the miss­ing in­gre­di­ent seems to be an un­der­stand­ing of bi­ases (a la Kah­ne­man and Tver­sky).

In this es­say, Eliezer laments that be­ing a tra­di­tional ra­tio­nal­ist was not enough to keep him from de­vis­ing a Mys­te­ri­ous An­swer to a mys­te­ri­ous ques­tion. That puz­zles me be­cause I would have thought that tra­di­tional ideas from Peirce, Pop­per, and Korzyb­ski would have been suffi­cient to avoid that er­ror. So ap­par­ently I fail to un­der­stand ei­ther what a Mys­te­ri­ous An­swer is or just how weak the tra­di­tional form of ra­tio­nal­ity ac­tu­ally is.

Can any­one help to clar­ify this? By “Tra­di­tional Ra­tion­al­ity”, does Eliezer mean to des­ig­nate a par­tic­u­lar col­lec­tion of ideas, or does he use it more loosely to in­di­cate any think­ing that is not quite up to his level?