Date: Tue,  2 Apr 2002 01:24:23 +0200
Subject: RE: BHA: Ellis, Scientific Essentialism


Dear Ruth (and others),

Finally, I have regular computer access. I am sorry for not replying to your 
mails earlier, but as I have had to use computers in overcrowded rooms at my 
university I have not been able to comment your points until now.

I am now finishing a review of Ellis' book for the JCR. As some of your points 
will be treated there I find it more convenient to send you the review instead 
of repeating the general points here (If this is acceptable for Mervyn).

Some short comments on your mail:

1) Properties
Categorical and Dispositional are to my knowledge all types of properties that 
Ellis accepts and think legitime. I find it hard to think of any other 
possible type of real metaphysical type of property-universals than those two. 
Some properties may be hard to place however, most real properties will be 
dispositional however. 

2) Dispositional Essentialism
Ellis follows Shoemaker in constructing causal, dispositional or modal 
properties as the essences of substantive kinds. This amounts to Ellis' 
Dispositional Essentialism. It is what a thing can do that determines what a 
thing is. The theory of process-kinds and their essences I find mildly 
confusing however. See my review for details. In short my point is that we 
will, as dispositional essentialists, say that substantive-kinds are in some 
way primary entities while properties and especially events and processes are 
ontologically dependent on substances for existing. Ellis is quite explicit in 
treating tropes as ontological dependent on substances but is more ambiguous 
on processes and events. I think that the latter position bear some 
resemblance to Bhaskars position in RTS. In any case what is missing in both 
RTS and SE is a "sparse" theory of processes/events. Such an theory is quite 
close to Geach's distinction between a real change and a "mere" Cambridge 
change (I hope you are familiar with this distinction). But wishing for such a 
theory is quite different than arguing for it ...

3)Events
Ellis has the same concept for events and processes. He is calling 
them "natural kinds of events or processes" It's nothing mystical about them.

4) Stochastic Processes
Some laws of nature are not causal laws at all. Ellis argues that these laws 
arise from the global kind, that is the universe. Some fundamental laws such 
as the conservation laws and the laws of thermodynamics plus GTR my be 
construed as fundamental ways of being for the universe. I will lightly touch 
on this in my review.

Best Regards,

Ronny

*****************************************************************************
Ruth Groff <rgroff-AT-yorku.ca>:

Hi Ronny, all, 

Ronny, I am hoping that you will have computer access soon! 

I have gone back over the first three chapters of Ellis with a fine-tooth 
comb!  I am writing for two reasons: first, I want to make sure that I've got 
it; second, I have a few questions. 

As I understand him, Ellis sets out the following over-lapping categories:   

First of all, there are three different types of natural kinds: substance-
kinds, process-kinds and property-kinds.  One type of PROCESS-kind is causal 
process-kinds.  Within the category of  natural kinds of PROPERTIES, 
meanwhile, there are categorical properties and dispositional properties.  
[Question #1: does this exhaust the types of property-kinds that there are, or 
are "categorical" and "dispositional" just two of many?]  Finally, within the 
sub-category of dispositional properties there are causal properties (or 
causal powers) and there are stochastic properties (or tendencies). 

Second,  Ellis says that dispositional properties constitute the real essence 
of causal process-kinds.  [I assume that technically it is causal properties 
that are the essences of causal processes and stochastic properties that are 
the essences of stochastic processes.]  Ellis also suggests that causal 
properties are the real essences of at least some substance-kinds -- e.g., 
electrons. 

Third, in characterizing causal processes, Ellis introduces the notion of 
kinds of EVENT.  He says that causal properties, or powers, relate certain 
event-kinds (viz., those that trigger the display of the causal power in 
question) to other event-kinds (viz., those which constitute the manifestation 
of the power in question).  Thus, causality, he suggests, is best understood 
as a relationship between KINDS of events, not between events themselves.     

Questions:   

Have I got the above right? 

If dispositional properties are the real essences of some process-kinds and of 
(some?) substance kinds, does this not suggest that property-kinds (of which 
dispositional properties are a type) are ontologically more basic than either 
substance or process kinds?   

Where in the tri-part typology of kinds (substance, process, properties) do 
event-kinds fit?  He introduces them in the text more or less out of the blue. 

Why does Ellis characterize causality as a type of relationship between event-
kinds -- and ultimately, it seems to me, as a type of process-kind -- rather 
than as the expression of dispositional properties?  In Ellis' terms, would 
people agree that in RTS causality is conceptualized more in terms of the 
properties, as something like "the capacity of powers to effect change?" 

Thanks in advance to anyone who has looked at Ellis and wants to comment.   

r. 





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