CEVs; are they real?

[From Bill Powers (940829.1700 MDT)]

Clark McPhail -- certainly, have Long call me and we'll work out the


Martin Taylor (940830.1130) --

It seems strange to call a complex environmental "variable" a function.

The CEV and its value are physical quantities.

The values of the individual variables are physical quantities, but the
function itself does not exist in that collection of variables. It
exists ONLY in the PIF.

Look at my 1973 Science article, Fig. 1 (I think). Fig. 1 shows a
collection of Vs in the environment, each connected to the input
function by an arrow and each affected by arrows from the control
system's output. There is a single perceptual signal, but there is
nothing in the environment that corresponds to it. All that exists in
the environment is a collection of physical variables. This is what the
diagram was supposed to convey to the reader who went beyond the simple
content of the article.

I know that you want there to be some objective correlate of the
controlled perceptual signal so that you can tie evolution to the
development of the control systems. But there is no objective correlate.
Controlling a function of that set of Vs, we can assume, has some
beneficial effect for the organism, but it need not have anything to do
with the way the organism perceives the Vs. There is no entity in the
environment corresponding to the perceptual signal, nor need there be.
It may be that only one of the Vs is critical and that the rest are just
along for the ride. It is still a physical system, but how we
understand what is controlled is quite beside the point for the control
system itself. The only "CEV" exists in the mind of another observer
looking at the same environment.

The CEV is where interactions in the world occur, even though nobody,
nobody at all, knows at any moment what any particular CEV might be.

Well, I disagree with this reification. All a control system has to do
is look at the same collection of physical variables differently, and
they constitute a different CEV. This has nothing to do with the actual
state of the environment; it is a matter of perception alone.

One fundamental assumption of PCT is that there exists an outer "real"
world, even though we access it only through our PIFs. The values of
perceptual signals depend on what happens in the outer world. They are
not themselves the outer world events.

All that is true, but it does not make CEVs "real" nor does it mean that
for every perceptual signal there is some physically meaningful
correlate in the environment. We can get along entirely without that
postulate, as far as I can see.

If there is no "real" world, with real values that multiple factors can
affect, then there is no point to thinking about control systems that
work through it. Under those circumstances, there would be no need to
consider the notion of a CEV. And no PCT.

True but irrelevant. I assume that there are real physical variables in
a real environment: the Vs. But it does not follow from this that a
perceptual signal corresponds to any entity in the environment.

Bill P.