Bayesian Statistics engages in Pre-9/11 Thinking....

Going through Jeff Gill's Bayesian text book when I came across the line:

From the Bayesian perspective, there are two types of quantities: known and unknown. (43)

This seems to be from the pre-9/11 school of thought as Rumsfeld* has us living in a more complicated world:

I guess technically I should exclude unknown knowns - that is, those things we did not know that we already knew.  I also have written the words "know" and "known" far too many times that it no longer appears to be a real word anymore.

*"Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns -- the ones we don't know we don't know." ~ Donald Rumsfeld, quote recently appropriated from one of Aparna Kher's dissertation chapters.

Comments

[Drew Conway]

Very clever!

[Julie VanDusky]

I think an unknown known is either (1) something we know that we don't realize is important in solving our problem, or (2) a secret that key people know that they don't share with other important people... like Area 52. Conspiracy theorists think they know the unknown knowns.

[Joshua Livingston]

I know it is amusing, but I've never seen how Rumsfeld is being academically obtuse here. After all, there are a number of paradoxes associated with whether we know that we know when we know, and one of the main disagreements between internalist/externalist epistemologists orient around KK issues.

[Cynthia Van Maanen]

Just an update: MSNBC's coverage of the second Presidential debate sums up the last question of the night like this: "What don't you know and how will you learn it?" I think this might effectively introduce a new classification: knowable versus unknowable knowns and unknowns. The debate question clearly implies that there are some things that the candidates don't know, they may or may not know they don't know, but could know, as opposed to those things they don't know, may or may not know they don't know, and won't be able to know.