Date: Tue, 21 Aug 2001 21:53:43 -0400
From: howard engelskirchen <lhengels-AT-igc.org>
Subject: Re: BHA: What's wrong with naturalism?


Thanks for the post and references.

As one who couldn't attend the IACR congress, I'd be very interested in a
review of the focus of attention on causality and some summary of main
points of discussion.

I think attention to causality is excellent.

There is much I agree with in your post, but I am not really clear on what
you mean by structural causality.  Also, I am not clear when you say that
we must accept the three kinds of causal powers whether you mean that they
all do the same thing.  Do structures roll a ball uphill?  

The trouble is that since the Renaissance we have dropped final cause as
mystic stuff, assumed material cause, dropped formal cause and reduced
everything to efficient cause.  Yet there may be a kernal of truth here we
cannot avoid living with.  When we speak of the actual transformation of
material things, in some respect this takes always an efficient cause.
Heat is an efficient cause and molecules go faster.  That is not the
problem.  The problem is to suppose explanation at that point is over.
Otherwise stated:  give me an example of a structural cause that changes
the material world that does not also involve some way or other an
efficient cause.  Probably we need to speak of causal ensembles where we
recover not other types of cause acting as if they were doing the kind
thing cause does -- ie they also are causes that get things done -- but
other types of cause act together with efficient cause to complete an
explanation (as they must also in the natural sciences).  Or again,
methodological individualism is not a problem because it focuses attention
on the fact that only individuals ever act in the world; it is a problem
because it ignores the social relations presupposed by and within which
they act.

Howard




At 12:08 PM 8/21/01 +0200, you wrote:
>Just returned from the IACR congress in Roskilde, Denmark, I want to
>further discuss the issue of causal models.
>
>Many sociologists, especially postmodernists, are strongly opposed to any
>naturalist argument and reject any talk of causal models. They want
>hermeneutics instead. In my view, their hermeneutic explanations are either
>causal explanations in disguise, or they are not testable scientific
>statements (following the criteria of Popper and Lakatos).
>
>Critical realists accept causality, but some only accept certain kinds of
>causality.
>
>You may distinguish three kinds of causal powers:
>1. Efficient causes, i.e. people who act.
>2. Structural causes, i.e. social structures that constrain or enable.
>3. Material causes, i.e. things or environment that influence social events.
>
>Some realists accept only efficient causes, arguing that only people can do
>things.
>In my opinion, all types of causes must be accepted for the following
reasons:
>a. No additional insight is gained by explaining a material or structural
>cause by reformulating it in terms of the agent (person) that it works
>through.
>b. A society can show emergent phenomena that are best explained in terms
>of structural causes. Explaining such social phenomena with reference to
>individual actors appears to be reductionist, and it is hard to imagine a
>theory of collective action that doesn't include structure.
>c. I am doing research on unintended and undesired social phenomena. My
>research shows that many unintended developments can only be explained by
>reference to structural causes (see references below).
>d. In the natural sciences, most causes are material causes. A rejection of
>material causes will prevent any interdisciplinary research that involves
>natural sciences or other sciences where material causes are common.
>
>The problem with many philosophers of the social sciences is that they are
>not doing applied research. A philosophy or methodology must be applied to
>see if it works.
>
>Many philosophers and sociologists believe that the social sciences must be
>based on a philosophy and methodology that is radically different from most
>other sciences. I strongly disagree. They are fighting with fundamental
>methodological problems that other sciences have solved long ago. What is
>so special about sociology that it can't use the solutions that other
>sciences have found?
>
>It is argued that social systems are very complex. True, but many other
>sciences are dealing successfully with equally complex systems.
>
>It is argued that a society is an open system where you cannot isolate a
>simple causal mechanism. True, but many other sciences have exactly the
>same problem. Meteorology, for example. Sociologists claim that you can't
>study mechanisms and make predictions in an open complex system. That's
>what meteorologists are doing every day. Of course their predictions
>sometimes fail. But more often their predictions are close to what actually
>happens. Their predictions are certainly better than if they were rolling
>dice. So they must have got something.
>
>Realists are fighting with the problems of defining tendencies, which
>sometimes are actualized, sometimes not because of countertendencies, etc.
>The natural sciences have no problems expressing such statements in terms
>of Boolean algebra or functions of more than one variable.
>
>Sociologists say that you can't make predictions because of uncertainties.
>That's what the theory of probability was made for.
>
>Sociologists say that you can't make causal models because people have a
>free will and have the freedom to do unexpected things. But economists are
>doing this nevertheless. Animals also have the freedom to do unexpected
>things, but this doesn't prevent the biological sciences from making
>successful models. Individual animal behavior is described in the science
>of ethology. Collective animal behavior is described in sociobiology. And
>the complex system of interaction between different species is described in
>ecology. (Sociobiology has got a bad name because it's founder E.O. Wilson
>was a scientific imperialist who believed that genes could explain
>everything. But others, not making this reductionist fallacy have made
>successful models).
>
>Sociologists have problems of how to deal with poorly delineated
>categories. Such problems are handled by the theory of fuzzy sets or by
>cluster analysis.
>
>The problems of small events having a large effect is dealt with in chaos
>theory.
>
>Of course it is extremely difficult to use mathematical models in sociology
>because many parameters are difficult to quantify and measure, but other
>sciences have similar problems. In most cases, we would be happy with
>qualitative statements anyway. This doesn't make sociology qualitatively
>different from other sciences, and certainly it doesn't justify a
>philosophy for sociology that is radically different from other sciences
>and prevents interdisciplinary research.
>
>Some scientists have gone to the opposite extreme and set up complicated
>mathematical formulas for culture (see ref.). They don't get very far
>because they have no numbers to put into their equations. This is of course
>not very fruitful, but at least it is more creative than to just give up,
>as postmodernists do, and claim that causal models are impossible.
>
>
>References:
>-----------
>Papers from the conference: www.ssc.ruc.dk/iacr/programme.htm
>
>My paper, discussing and applying my method to the study of mass media:
>www.ssc.ruc.dk/iacr/papers/Fog.pdf
>
>Paper by Berth Danermark, discussing methodologies for interdisciplinary
>research:
>www.ssc.ruc.dk/iacr/papers/Danermark.pdf
>
>Paper by Bob Carter and Caroline New discussing different types of causality:
>www.ssc.ruc.dk/iacr/papers/NewCarter.pdf
>
>Article with similar arguments about structural causes:
>P Lewis: Realism, Causality and the Problem of Social Structure. Journal
>for the theory of social behaviour, vol 30 no 3, 2000, p. 249.
>
>For an example of extreme mathematical modelling of culture, see:
>C. S. Findlay, et al: Behavioral evolution and biocultural games.
>Proceedings of the National Academy of Science of the USA, vol. 86, no. 2,
>p. 568, 1989.
>
>My website with theory and applications: www.agner.org/cultsel
>
>
>     --- from list bhaskar-AT-lists.village.virginia.edu ---
>



     --- from list bhaskar-AT-lists.village.virginia.edu ---