Re: Piorot on Madrick - Winslow's blast

[Ted Winslow <egwinslow@xxxxxxxxxx>]

John M. Legge wrote:

> There is a simple polemical issue in asserting that people act
> "irrationally".  This can only mean that they carefully work out what
> they
> really want to do and then do something else.

On the understanding of irrationality that underpins Keynes's treament
of it, there is a form of irrationality that fits this definition.
Unconscious motives can cause speech and action to contradict what was
consciously intended ("parapraxes").  This is not the only form, however.

Again on the understanding that underpins Keynes, irrational beliefs and
actions in general are those that are unconsciously motivated and for
this reason immune to rational critique.

An example is "the idea that the quantity of hoards depends in any way
on what people are doing with their savings, or that there is any
connection between idle balances and the conception (meaningless on my
definitions) of idle savings."

Keynes claims that attempting to clear minds of this idea by pointing to
the "obvious" mistakes involved has only "a limited hope of success"
because "there is a deep-seated obsession associating idle balances, not
with the action of the banks in fixing the supply of cash or with the
attitude of the public towards the comparative attractions of cash and
other assets, but with some aspect of current savings."  (XIV, p. 214)

This claim is confirmed frequently on this list.

> What Arthur and the complexity theorists have demonstrated is that the
> selection of actions in the true best interests of the selector requires
> infinite information collection ability supported by infinite computing
> ability, neither of which are possessed by finite mortals, or indeed,
> anything smaller than the universe itself.  The obvious and frequent
> failure
> of events to follow neoliberal predictions need not be due to people
> acting
> irrationally, since such variant outcomes can arise when people act
> perfectly rationally within the limitations imposed by their finite
> information gathering and computing capability.

Demonstrating that a conclusion - "actions in the true best interests of
the selector requires infinite information collection ability supported
by infinite computing ability, neither of which are possessed by finite
mortals" - follows logically from a set of premises - those of
complexity theory - doesn't demonstrate that the conclusion and the
premises are true.

Keynes claims that it's also very difficult to clear a particular kind
of mind of this obvious mistake i.e. that the belief is "irrational" in
the sense I've just specified.  He describes a  particular instance of
it - Hayek's argument in Prices and Production - as "an extraordinary
example of how, starting with a mistake, a remorseless logician can end
up in Bedlam." (XII, p. 252)

> As far as I am concerned, adding some additional cause of "fundamental
> uncertainty" beyond the simple impenetrability of the future as
> demonstrated
> by complexity theory can be shaved away with Occam's razor.

Again ("as demonstrated by") you are misidentifying consistency with
truth.  You compound this error by mischaracterizing the alternative
explanation of uncertainty provided by the ontological premise of
internal relations - the explanation to which I pointed - as adding an
"additional cause" to those identified by complexity theory when in fact
what it does is derive fundamental uncertainty from ontological premises
inconsistent with the ontological premises of complexity theory.  This
gives "cause" a very different meaning from the one on which you're
insisting.  This is implicit in the argument pointed to in item (5) of
the list in my post.

By the way, you're meaning leads to Hume's problem of induction;
Whitehead's, in terms of "internal relations," doesn't.  This is
additional to the problem of self-contradiction to which I pointed i.e.
on your meaning of "cause" and "determined,"  present actions are
determined in a way that leaves no logical space for the idea of present
_choices_ affecting what will occur in the future.  Even if we ignore
the impossibility, on your premises, of a finite mind reaching perfect
knowledge, "the selection of actions in the true best interests of the
selector" is, on your premises, logically impossible e.g. knowing a
future winning lottery number would be of no use because you would also
know the winner.

> As a practical matter, complexity theory suggests that projecting the
> past
> into the future is in general the "best" estimate of what is likely to
> happen, because while we do not know whether the future will go up or
> down
> or left or right we know that it starts with the present and that
> discontinuities are relatively unusual.  We also know, thanks to
> Lyapunov,
> that the uncertainty surrounding our projections rises exponentially
> with
> time until all possible outcomes approach equal probability at a finite
> time
> into the future.
>
> I don't expect the Dow to exceed 18,000 or fall below 4,000 tomorrow;
> but as
> historical experience has shown, moves of this order can occur within a
> ten
> year period.

Once more, you are confusing consistency with truth.  That fact that
"the conclusion that projecting the past into the future is in general
the 'best' estimate of what is likely to happen" follows from the
premises of complexity theory doesn't make the conclusion and the
premises true.  So you can't reasonably conclude on this basis alone
that this is the "practical" approach to predicting or that we "know,
thanks to Lyapunov, that the uncertainty surrounding our projections
rises exponentially with time until all possible outcomes approach equal
probability at a finite time into the future."

Keynes's alternative ontological premises also explain the uncertainty
of the long run.  They also provide grounds for reasonably believing
that relationships that have been observed in the past not only will not
continue into the future but will change in a predictable way,  grounds
such as those on which Keynes's prediction   that the US will become a
"high-living, high-cost country" is based.  These grounds invalidate
forms of reasoning that make use of the logical concept of the
"variable."

He doesn't, however, make the mistake of claiming that the truth of this
ontology can be established by means of formal logic alone.  He
adumbrates a field of study, "human logic," which has as its concern the
grounding of ontological premises in experience.  For the reasons
Whitehead gives, "human logic" can't be based on deduction.

Even without taking account of the implications of Keynes's ontological
premises, however, it's obvious that what Keynes describes as
"conventional" forecasting practices don't produce "'best' estimates"
e.g. "day-to-day fluctuations in the profits of existing investments,
which are obviously of an ephemeral and non-significant character, tend
to have an altogether excessive, and even an absurd, influence on the
market." (GT, pp. 153-4)

> My biggest problem is in trying to see why this is contentious,

This is true.

Ted