[90601] in Cypherpunks
Re: Environmentalism (was Re: Crimes) (fwd)
daemon@ATHENA.MIT.EDU (Jim Choate)
Thu Nov 20 22:44:20 1997
From: Jim Choate <ravage@ssz.com>
To: cypherpunks@ssz.com (Cypherpunks Distributed Remailer)
Date: Thu, 20 Nov 1997 21:12:44 -0600 (CST)
Reply-To: Jim Choate <ravage@ssz.com>
Forwarded message:
From owner-traveller@Phaser.ShowCase.MPGN.COM Thu Nov 20 21:07:00 1997
Message-Id: <3.0.3.32.19971120170206.007b1100@mail.deltanet.com>
X-Sender: fuz@mail.deltanet.com
X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32)
Date: Thu, 20 Nov 1997 17:02:06 -0800
To: traveller@MPGN.COM
From: Scott Ellsworth <Scott_Ellsworth@alumni.hmc.edu>
Subject: Re: Environmentalism (was Re: Crimes)
In-Reply-To: <199711202104.QAA14932@Mithril.MPGN.COM>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-traveller@Phaser.ShowCase.MPGN.COM
Reply-To: traveller@MPGN.COM
At 04:00 PM 11/20/97 -0500, Rob Flamming wrote:
>> From: bmac@astro.ucla.edu (Bruce Alan Macintosh)
>>>Economics is not a /prescription/ for how humans
>>> /should/ behave; it is a /description/ of how humans /do/ behave.
>> Economics doesn't reflect absolute truths; it makes subjective decisions
>> based on relatively arbitrary criteria, including extremely arbitrary
>> accounting conventions.
> Economics reflects truth in the same way that Physics, Chemistry, and
> other sciences reflect truth. Obviously, an incorrect economic theory
> will not model the truth correctly just as an incorrect physical
> theory will not model the truth correctly.
Please do not take this too badly, but I must strenuously disagree, A
primary difference between Physics, Chemistry, and other hard sciences, and
Economics, along with many of the other social sciences is repeatability
and system independence. Most of the hard sciences have varying degrees of
those, while the social sciences have a hard time inducing them. I draw
this conclusion after earning two degrees in hard science, and working for
the last five years for Economists developing Econometric software, so I am
not arguing from a purely theoretical basis.
Repeatability:
Like cosmology, Economics has a hard time coming up with new data for a
previous situation - once the situation has passed, you have all of the
data you are ever likely to get on that economic climate. Even if it
returns, it is not likely to be a similar situation. This problem plagues
economists, especially since when you start abstracting things in the model
to make it one that might be general enough to apply to more than a single
economic era, the benefits of the models tend to vanish.
This might arguably be solved with better computing power, and a more
comprehensive model of how societies function. I do believe, for instance,
that the Vilani system was far better understood, as they had much less
diversity, and thus could reasonably expect to see similar situation arise.
System independence:
The physical universe cannot change behavior according to our current
theories, while the people who practice and formulate economics can alter
the behavior of the people they are trying to model. For instance, recent
census data has pointed out the new trends in child rearing, especially as
it affects the number of single parents. This information has already
started to affect the debates on welfare and other issues, which means that
a model that perfectly predicted behavior without that information would
have some added difficulty.
I am aware that QM implies a lack of observer independence, but on the
macro scale that most hard sciences take place, observer independence is
still effective. For example, if you are measuring the Hall effect, which
depends on QM, you will get statistically significant experimental results
which match a colleague with different expectations, even if the data has
already been published.
> Making sure that your
> theories are right is what makes science in general so hard, but the
> whole idea of all sciences is that they reflect the truth and
> (eventually) weed out falsehood.
Clearly, clearly, but it is easier to do this in the physical sciences
where you system does not adapt to what you do quite so well, and where
your system can be reset to the initial state so you can look at it in detail.
[accounting conventions mentioned]
> This is an accounting convention. It has as much bearing on Economics
> as an engineering convention has on Physics. Economists, like
> Physicists, need to take care that their theories remain independent
> of such conventions. The discipline of science is intended to ensure
> such independence.
While this is true in the abstract, I believe that economics, by its very
nature, is going to be affected by its conventions more than the hard
sciences usually are. Economists have a great deal of difficulty isolating
factors under investigation, which include the responses to accounting
conventions. I do not mean data isolation with statistical techniques, I
mean actually separating a group of people and their behaviors from the
outside world, so that they can be examined under different events. Much
like psychology, you have a hard time controlling the initial conditions,
and the behaviors are determined in part by those accounting conventions.
Let me say it another way - an engineering convention, such as the sign on
the electron's charge, can be accounted for. Likewise a good physicist can
get an idea of how well their measurement device works, which falls out of
those engineering conventions. The economist, on the other hand, cannot so
easily separate the behavior of people from the accounting convention that
engendered the behavior.
Clearly, economists do try to make their plans usable over the long term,
but the difficulties inherent in their task will cause small amounts of
noise in their models to amplify, such that it is very difficult to make a
long term prediction good enough to convince someone of an economic good
that is a decade away. In other words, before being worried about a crisis
predicted for a decade from now, I am going to examine what the same group
said ten years back, and see if they have any credibility over that time
horizon.
Note: hard scientists also have their share of such things. Any time you
try to make a model of a complex system, and push your models beyond design
constraints, you are going to have some amount of difficulty. Predicting
weather a decade hence is about a tough as predicting the market over the
same time period. Neither is very reliable after next week.
Scott
Scott_Ellsworth@alumni.hmc.edu http://users.deltanet.com/~fuz
"When a great many people are unable to find work, unemployment
results" - Calvin Coolidge, (Stanley Walker, City Editor, p. 131 (1934))
"The barbarian is thwarted at the moat." - Scott Adams
