Economics is a Science was Re: OT: Most Shameful Moments in American History
Luke Kaven
wrote:
>mochas...@notsohotmail.com (mochaspresso) wrote
>
>> On 21 Jan 2003 04:36:52 GMT, "Die For Oil, Sucker"
>> <jas...@yahoo.com> wrote:
>>
>> Perhaps I'm sticking my nose where I don't belong.....but isn't
>> economics considered a Social Science? It's not a pure / physical
>> science in the same vain as chemistry and biology.
>>
>Correct you are, but no one is claiming it is a physical science.
>
>> Mochaspresso
The members of this debate appear to be conflating two important
aspects of inquiry -- metaphysics and empirical observation. In order
to understand science, one has to understand how these work (or fail
to work) together.
Maths is a mostly a branch of metaphysics, just as logic is. The law
of identity is not something one can derive from observation *alone*.
Those who claim it is a science are not entirely mistaken, though it
has less of an empirical component than other sciences, such as
chemistry and biology. Though chemistry and biology both make use of
metaphysics. Evolution is a mostly metaphysical theory.
For those who don't understand what metaphysics is, it is just the
study of unobservables -- all abstracta, including reasoning about
time and causation belong to it. [You observe neither time, nor
causation.] Scientists do metaphysics routinely along with empirical
observation. The difficulty in doing metaphysics is that all one has
to rely upon is the social web of reasoning over history. In the
early 20th c., there were aggressive attempts to eliminate metaphysics
from the sciences (logical positivism, logical empiricism, and some --
but not all -- varieties of behaviorism). But attempts thus far have
failed, and the practice fell into disfavor with the arrival of
Kripke's modal arguments about a posteriori necessity in the early
1970s, which refuted a number of semantic arguments underpinning the
logical positivist/empiricst arguments.
Economics is more akin to the social sciences. It has a higher
concentration of metaphysics, though it also draws upon empirical
observation.
These questions are a matter of degree. Metaphysics and empiricism
are continuous.
Luke
Uncle Al
>
> On 21 Jan 2003 05:47:35 GMT, "Die For Oil, Sucker" <jas...@yahoo.com>
> wrote:
> >mochas...@notsohotmail.com (mochaspresso) wrote
> >
> >> On 21 Jan 2003 04:36:52 GMT, "Die For Oil, Sucker"
> >> <jas...@yahoo.com> wrote:
> Economics is more akin to the social sciences. It has a higher
> concentration of metaphysics, though it also draws upon empirical
> observation.
[snip]
Let's cut the crap. Given 200 years of financial records, a sack of
Nobel Laurates, multi-$billion brokerage houses supporting the most
complete and esoteric mathematical modelings of which the human mind
is capable, plus hardware... tell Uncle Al whether the Dow Jones
Industrial Average will have risen or fallen at the end of the next
market session.
If economics is a science, predicting 40 sessions in advance should
raesonably give you a 0.900 batting average. If economics is
flummery, you will snug .500. If economics is crap, you won't even
get to 0.500.
Given the continuing collapse of the stock market dating from Bush the
Lesser's election and the fact nobody got rich massively selling short
- NOBODY - Uncle Al posits that economics is crap. Anybody who still
doubts is invited to peruse the reuslts of social engineering and
income redistribution - a massive hungry army of Black and Brown
reprodcutive warriors - and the Boys From Chicago led by Nobel
Laureate economist Milton Friedman who destroyed Chile. Folks are
still pissed of at Pinochet 30 years later.
The idea that a large number of variables can be mutually reconciled
is not only repeatedly proven insane in the real world, it is
trivially proven to be mathemtically impossible ab initio,
http://www.mazepath.com/uncleal/comprom.htm
If you want to get from here to there, lots of folks must be massively
screwed without a second thought. Gordon Gecko knew how the stock
market is really played.
http://www.mazepath.com/uncleal/eotvos.htm
(Do something naughty to physics)
--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
"Quis custodiet ipsos custodes?" The Net!
Repeating Decimal
lu...@smallsrecords.com wrote on 2/14/03 1:43 PM:
>> mochas...@notsohotmail.com (mochaspresso) wrote
>>
>>> On 21 Jan 2003 04:36:52 GMT, "Die For Oil, Sucker"
>>> <jas...@yahoo.com> wrote:
>>>
>>> Perhaps I'm sticking my nose where I don't belong.....but isn't
>>> economics considered a Social Science? It's not a pure / physical
>>> science in the same vain as chemistry and biology.
>>>
>> Correct you are, but no one is claiming it is a physical science.
>>
There are experiments being made in economics. Caltech is one center. IMHO,
it is as much a matter of psychology as economics. Nevertheless, the
experimental results are indeed interesting.
Bill
Luke Kaven
wrote:
>Luke Kaven wrote:
>> Economics is more akin to the social sciences. It has a higher
>> concentration of metaphysics, though it also draws upon empirical
>> observation.
>[snip]
>
>Let's cut the crap. Given 200 years of financial records, a sack of
>Nobel Laurates, multi-$billion brokerage houses supporting the most
>complete and esoteric mathematical modelings of which the human mind
>is capable, plus hardware... tell Uncle Al whether the Dow Jones
>Industrial Average will have risen or fallen at the end of the next
>market session.
>
>If economics is a science, predicting 40 sessions in advance should
>raesonably give you a 0.900 batting average. If economics is
>flummery, you will snug .500. If economics is crap, you won't even
>get to 0.500.
>
> Let's cut the crap. Given 200 years of financial records, a sack of
> Nobel Laurates, multi-$billion brokerage houses supporting the most
> complete and esoteric mathematical modelings of which the human mind
> is capable, plus hardware... tell Uncle Al whether the Dow Jones
> Industrial Average will have risen or fallen at the end of the next
> market session.
Uncle,
While I'm the last to defend the current practices of any of the
sciences, especially those in their infancy, I do think you're missed
the point here.
There is no necessity that science be predictive in the particular.
We can't predict how species will evolve over the next billion years,
but that doesn't mean the theory of evolution is false or pointless.
It just means that there are enough stochastic variables to make the
prediction intractable. [Consider the "halting problem" as an
example.] Anyway, evolution is better as a retrodictive theory.
You'd like prediction in the particular from economics, but there is
no reason to believe that such predictions can be made with accuracy
under any circumstances due, at the least, to the number of stochastic
variables and extreme sensitivity to initial conditions. But that
isn't an embarassment for economics, whatever other embarassments it
does suffer. You have to distinguish between the metaphysical problem
and the epistemological problem here.
Anyway, if there were perfect knowledge of the game, then all the
players would have perfect knowledge, and it would be a much different
game, to say the least, and I'm not sure whether anyone would play, or
whether the concept of winners and losers would make any rational
sense. Nobody would play if they knew for certain that they would
lose, and if that were the case, who would be the winners? Is there a
Nash equillibrium in there?
Luke
Glen M. Sizemore
of the
sciences,[...]
GS: Any of them?
[...]especially those in their infancy, I do think you're missed
the point here.
GS: Right...but it is hard to discern when the application
of the term "infancy" becomes an excuse for a dead-
end conceptualization. This is certainly the case with
mainstream psychology (with an emphasis on
"mainstream").
There is no necessity that science be predictive in the
particular.
GS: But nonetheless prediction remains a goal of
science.
We can't predict how species will evolve over the next
billion years,
but that doesn't mean the theory of evolution is false or pointless.
It just means that there are enough stochastic variables
to make the
prediction intractable. [Consider the "halting problem"
as an
example.] Anyway, evolution is better as a retrodictive
theory.
You'd like prediction in the particular from economics,
but there is
no reason to believe that such predictions can be made
with accuracy
under any circumstances due, at the least, to the
number of stochastic
variables and extreme sensitivity to initial conditions.
But that
isn't an embarassment for economics, whatever other
embarassments it
does suffer. You have to distinguish between the
metaphysical problem
and the epistemological problem here.
GS: But, again, prediction remains an important goal of
science. We must be careful that the "number of
variables argument" and the "sensitive dependence
argument" don't become mere excuses. For one thing,
we don't have any evidence that all (or any) of the
phenomena of "economics" (and just which phenomena
are we talking about here?) are chaotic, though they
may well be.
Anyway, if there were perfect knowledge of the game,
then all the
players would have perfect knowledge, and it would be
a much different
game, to say the least, and I'm not sure whether anyone
would play, or
whether the concept of winners and losers would make
any rational
sense. Nobody would play if they knew for certain that
they would
lose, and if that were the case, who would be the
winners? Is there a
Nash equillibrium in there?
GS: Don't understand the last part. Anyway, though,
there is a sub-field of behavioral analysis that is often
called "behavioral economics," and it utilizes notions
like demand functions and unit price. You might be
interested in it.
Notice that I'm not necessarily disagreeing with your
analysis, but behavior analysis faces many of the same
problems that you are talking about, yet I would argue
that it is about the only scientifically legitimate kind of
psychology. Anyway, there's a lot more to say here,
but I'll leave it at that.
Luke Kaven <lu...@smallsrecords.com> wrote in message
Luke Kaven
gmsiz...@yahoo.com (Glen M. Sizemore) wrote
> While I'm the last to defend the current practices of any
> of the
> sciences,[...]
>
> GS: Any of them?
What I won't defend and what I'd criticize might be different things.
I'm here to defend science in general, regardless of present practice.
> [...]especially those in their infancy, I do think you're missed
> the point here.
>
> GS: Right...but it is hard to discern when the application
> of the term "infancy" becomes an excuse for a dead-
> end conceptualization. This is certainly the case with
> mainstream psychology (with an emphasis on
> "mainstream").
I think you're right, regardless of whether we'd agree on the
particulars of the shortcomings.
> There is no necessity that science be predictive in the
> particular.
>
> GS: But nonetheless prediction remains a goal of
> science.
Agreed. Where science can be predictive in the particular, it is
clearly useful. I become worried, though, when people assume that
predictive success in the particular is the only measure of the value
of a scientific theory. Even if one can't be predictive in the
particular, one can be predictive in general, and this may have
instrumental value.
I'm not assuming anything about chaos -- though I understand that a
couple of the terms I used are sometimes commonly associated with it.
I feel the idea of a "Chaos Theory" as distinct from analysis of
non-linear systems was always an invention of pop science culture.
I'm not espousing any such thing.
> Anyway, if there were perfect knowledge of the game,
> then all the
> players would have perfect knowledge, and it would be
> a much different
> game, to say the least, and I'm not sure whether anyone
> would play, or
> whether the concept of winners and losers would make
> any rational
> sense. Nobody would play if they knew for certain that
> they would
> lose, and if that were the case, who would be the
> winners? Is there a
> Nash equillibrium in there?
>
> GS: Don't understand the last part. Anyway, though,
> there is a sub-field of behavioral analysis that is often
> called "behavioral economics," and it utilizes notions
> like demand functions and unit price. You might be
> interested in it.
I was trying to point out that perfect knowledge about the outcome of
games changes the nature of the game.
Imagine, ex hypothesi, you have a zero-sum game, where in order for
there to be winners, there must be losers, and where there is perfect
knowledge. Nobody would rationally elect to be a loser, therefore
there would be no winners, therefore, nobody would elect to play the
game.
Granted the stock market may not fit exactly this description. But
when people ask for a science that will predict, viz, the value of the
Dow Jones in a year's time, they seem to be implicitly relying on the
premise (an implied cateris paribus) that *only they* have that
knowledge and that all else remains the same. Whereas I think that
the nature of the game would be radically different.
> Notice that I'm not necessarily disagreeing with your
> analysis, but behavior analysis faces many of the same
> problems that you are talking about, yet I would argue
> that it is about the only scientifically legitimate kind of
> psychology. Anyway, there's a lot more to say here,
> but I'll leave it at that.
Interesting point. I admit, I've never been able to entirely
encompass your neo-Skinnerian theories, so I can't say much about
that. I am glad, though, that someone is taking the time to think
those ideas through more completely.
Luke