Re: Why four categories of causation?

[Jannie Hofmeyr <***>]

Hi Steve

You quoted Robert Rosen as writing: "What Aristotle was essentially doing
in his discussion of the categories of causation was giving names to, and
thereby distinguishing between, certain kinds of relations between events.
As we noted in Section V, this is precisely the province of causality, one
of the twin pillars supporting our belief in natural law". Theoretical
Biology and Complexity p. 188.

I am surprised that, despite reading this paper many times, I had not
picked up on RR's use of "events" for both cause and effect in the above
quote. Aristotle's four "causes" are different (and exhaustive, according
to Aristotle) answers to the question "why something?", where something
could be an event, but is more often an object or substance (both natural
or artefactual). These answers then explain the object. Note that the
Aristotelean causes themselves are not events, as are they are in the
modern (post-Humean) sense.

I have found http://faculty.washington.edu/smcohen/320/4causes.htm a very
useful exposition. In particular, I have found the suggestion that we
shift from using the noun "cause" to the verb "make" useful, as it fits
perfectly with RR's use of Aristotelean causes for explaining fabrication.
If used in this sense, "causality" is an extremely useful concept.

SJ Q1) In everything I've read Rosen always takes Aristotelian taxonomy of
causes as given. I've never seen an argument that would justify this
particular choice. Should we consider Aristotelian taxonomy of causes a
foundational axiom in the Rosennean view? Does it have the same status as
the concept of Natural Law itself?

JH: Since the four Aristotelean causes cannot be derived from anything
else, I suppose they have axiomatic status.

SJ Q2) Are there other taxonomies of causes?

JH: To my knowledge nobody has come up with a different set, but if
alternatives exist I would love to hear about them.

SJ Q3) Is "Efficient Cause" a rigorously defined concept? Rosen (and
others) in describing this concept typically give the Aristotelian example
of the "house" with bricks as material cause, construction workers as
efficient cause, blueprint as formal cause, and intent to dwell as the
final cause. I've never seen any more rigorous definition besides
invocation of this admittedly useful metaphor. What is the definition of
efficient cause? Can anyone provide a defintion that does not invoke the
house metaphor? Efficient cause is key to Rosen's argument in Life Itself
so I think it's important to have a consensus on what it means.

JH: See http://faculty.washington.edu/smcohen/320/4causes.htm
Efficient cause is one sense of "aition", namely the one that "produces"
(or "makes") an object or event. In the formalised version of Rosen any
object or event b is in the range B of a map

f: A x I ---> B

where, for a particular element,

(a,i) |---> f(a,i) = b

A is the material cause, I the formal cause, f is the efficient cause.
This is a very powerful (and still underappreciated) and very precise way
of defining three of the four causes. In LI Rosen of course shows how to
handle the fourth: final cause.

All the best
Jannie