Re: UNCERTAINTY

["Alan G. Isaac" <aisaac@xxxxxxxxxxxx>]

Paul,

I believe you are simply offering idiosyncratic
*definitions* of epistemological and ontological
uncertainty rather than looking for the meaning
embodied in ordinary use of the words.
Specifically, it is extremely odd that you seem
to suggest ontological and epistemological
uncertainty as *alternatives*, when a more
natural claim for you would be that it is
ontological uncertainty rather than computational
limitations which underlie the epistemological
uncertainty that humans encounter in their
every day actions.

Most of the discussion in the past few days has
accepted that we have epistemological uncertainty
when the ways in which we learn about the world
leave us uncertain about the consequences
of our actions.  This might be because there is
no way for even God to know the future (ontological
uncertainty) or because of the obvious and permanent
limitations we face in computational power.
In contrast, ontological uncertainty
has been used rather naturally to refer to the
metaphysical claim that future states of the world
are not implied by natural law along with the current
state of the world.  While it is never fully clear
what people mean by ontological uncertainty
---for the simple reason that the exploration of
metaphysical claims generally overtaxes human reason
(see Kant)---claims on pkt about ontological uncertainty
are usually laced with some kind of assertion of
(non-compatibilist) free will.

If we ignore the general problem that when we speak
about metaphysical things we are usually asking
words to bear more weight than they can,
I will grant that the metaphysical (i.e., religious)
insistence that (non-compatibilist) free will
characterizes human action can serve as a background
for an insistence on epistemological uncertainty.
That is, aside from revealed truth, ontological
uncertainty is sufficient for epistemological uncertainty.
But it is not necessary.  Epistemological uncertainty,
as defined above, is readily discovered in ordinary
human experience, and this discovery does not require
specific metaphysical precommitments.

So although everyone is allowed to act like
the Red Queen and redefine the entire English lexicon,
I find your proposed definitions idiosyncratic and misleading.

In addition, you then offer an invalid deduction from
epistemological uncertainty to policy conclusions.
Even if we were to illegitimately restrict the models
encompassed by epistemological uncertainty to
neoclassical models containing only ergodic risk,
there can be perfectly good reasons for activist policy
in a large subset of the models. The justification for
activist policy is the same in these models as it is
from the PK perspective: it yields better outcomes.
The PK critique should remain that they have the wrong
model, not that they have the wrong metaphysics.

PK economics is done no favors if we pretend its insights
are available only to those who adopt controversial
metaphysical stances as a priori commitments.
Fortunately, that is not necessary.

At the end of your post, you suggest a shift in vocabulary.
You propose that speaking of a ``transmutable vs nontransmutable
future'' is ``perhaps a better phrasing than ontological vs
epistomlogical uncertainty.''  I would certainly agree that
this vocabulary better captures your core concern, which is
very badly captured by the other ``phrasing'', but it
also makes it clearer how tangential this issue is to
policy concerns. Even if we adopted an extreme materialist
metaphysics, the actions of policy makers (including their
discussions and debates) are an indispensible constituent
of economic outcomes.  You might argue that in this setting
policy ``doesn't matter,'' but only in the crude and still
debated sense that in such a world it ``doesn't matter'' if
you kill yourself.  These are inessential (for *economics*)
matters of religion, and it hurts PK economics to treat them
as anything else.

Alan






On Sun, 17 Dec 2000, Paul Davidson wrote:
> Epistomological uncertainty means that human computing power is limited and
> hence the preprogrammed future cannot be significantly predicted by
> humans-- despite the predetermined law of motion.  In Austrian theory --
> uncertainty is epistomological --with no human capable of predicting the
> future -- but the great computer in the sky-- the market can!! [Kight
> believed in epistomological uncertainty -- but as I show in my article on
> "reality and Economic Theory" Knight presumed an unchanging
> (nontransmutablre) economic world.

> In a world of epistomolgical uncertainty where the "real" world is
> preprogrammed from the moment of the "big bang", then humansmay make errors
> in the short run, but in Milton Friedman's long run, those who make
> decisions "as if" they "knew" the future (i.e., knew the economic laws of
> motion that govern the path of the economy over time), will survive and
> thrive.

>   "Epistomological uncertainty implies a "survival of the fittest
> Social  Darwinism" view.  The policy implication is that the market knows
> best -- and hence laissez-faire!!  (George W and his advisors would accept
> epistomological uncertainty.)  In an epistmological uncertain
> world,  humans can only discover the future they cannot create it --
> nothing that humans can do can change the trend path of the economy --
> anymore than Congress can  repeal the law of gravity.  In an
> epistomological uncertain world markets are efficient -- at least in the
> long run.

> In a world of  ontological uncertainty, there is no predetermined real
> world future out there.  Humans can create the future by their "animal
> spirits" ,  There is a role for goverment   to stabilize the economy .The
> economy is transmutable -- for better or for worse and can be pushed
> towards a better or worse future by human actions.

> Ultimately, if you believe in an epistemological uncertainty the only
> reason d'etre for supporting an active government role in the economy is
> that you are impatient and cannot wait for the long run to right things--
> but then if the future is  only epistomologicallly uncertain, then why
> should government bureaucrats (as Ronald Reagan continually asked) know
> better how to spend your money than you do [to maximize utlity]?

> on the other hand if you believe in an ontological uncertainty ---so that
> there is no nontransmutable preprogrammed future path for the economy, then
> there is a role for government to build INSTITUTIONS that will act as
> balancing wheel to offset any human action that pushes the economy in a
> "bad" direction.

> Ultimately this esoteric question of transmutable vs nontransmutable future
> (perhaps a better phrasing than ontological vs epistomlogical uncertainty)
> separates the men from the boys in terms of whether there is an active role
> for government policy in economic matters.

> paul