Thanks to Barkley and Brad for educating me on Keynes. Quite helpful
comments.

In response to Brad, I'd argue that my argument about ACTORs and OBSERVERs in
Keynes and Knight still stands, however. For Knight, as I understand him,
uncertainty is rooted in the problem of human action (and hence, has two
sides: I have to act, and other people have to act; both of us often have to
act ab nihilo). For Keynes, as I understand him, even after reading your
comments, uncertainty is rooted in the problem that the observer cannot know
things. For Keynes, human action proceeds from observation and the knowledge
built upon it; for Knight, action may or may not proceed from knowledge.
Thus, for Keynes, the problem of knowledge proceeds the problem of action;
hence, the problem of knowledge is the fundamental human problem. For
Knight, the problem of action is the fundamental problem, which knowledge
may or may not assist.

Thus, I was not arguing that Keynes' theory of probability is unconnected to
human action. But he and Knight see the relation between knowledge and
action in fundamentally different ways.

I haven't talked about epistemology and ontology (thanks, Barkley, for
using the right word) for quite some time, and find it difficult to express
the difference any clearer than I have here. I tried to avoid using the term
"pragmatist" in my earlier post, because I don't see Knight fitting that
description entirely, either in its philosophical sense and its common
usage. If there is a pragmatist element in his theory of uncertainty, it
probably comes via William James and Henri Bergson.

Ross Emmett