[Air-l] conceptual lexicon

[Ellis Godard]

I agree with chunks of what Jeremy wrote. But...
> definitions.   Network... looks like it lets you escape the issues  
> with 'community', but you always have to wonder then what you are  
> really talking about when you are talking about a network of  
> people.   You hopefully have defined some way that the people are  
> connected, thus the network, and then the question always becomes...  
> is the way that people are connected 'real' or an artifact of the  
> research.  frequently, I find that when people say there is a  
> network, I could disagree by saying the way that you operationalize  
> the network is insufficiently rich and lacks the capacity to really  
> indicate connection and all you really have mapped is 'x person has  
> talked to y person' or 'x person reports that they share interest z  
> with y person'.   In the end, the theory has to map onto the  
> empirical evidence, and not extend it.

I agree with the last sentence, and everything before it, but not the
implied connection. Operationalizing, observing, and measuring networks with
paths such as 'x talks to y' is perfectly sufficient if the exogenous
variable is a network of talk. The problem, as with "community", is
extrapolating from 'x talks to y' that the network involves more than (or
only) that, rather than extending only from that empirical evidence.

Somewhat related, networks (and communities) are too often and too hastily
toggled between the independent and dependent variation. Research on the
*consequences* of tight talk networks may say little or nothing about the
creation of or conditions for those networks. Lay talk, of course, makes
such jumps grossly and inanely. But scholarly work, too, could be more
articulate in delineating the empirical extent (and theoretical role) of
networks.

> here's a thought if you want to really deal with social 
> theory, throw  
> out 'methodological individualism' and figure out a way of actually  
> analyzing the collective.  once that is done, then the social 
> will be  
> analyzed in a way that is far more profound and perhaps more real. 

Again, I agree with these two sentences, but not the implication. We should
indeed focus on the "real", and move beyond invididualism. But that doesn't
mean that "analyzing the collective" requires aggregate measures.
Descriptions and observations should proceed from case instances. We can
thus analyze a collective (whatever it's called - network, community,
nation, Frank, etc.) as the multitude of observed interactions (is the
network of talk dense or sparse, tight or loose?) without extrapolating that
the collective is something more or other than those observed interactions.
Analyzing a collective, devoid of descriptions of case instances, is less
profound (and much farther from "real") than individualism.

-eg