I see where you are going with this, and I, too, began to think of the breaking down of institutional borders throughout Watt's work. One question that is nagging at me about this is who determines group affiliation. In Chapter 4 Watt is clear that he is positing random affiliation based on shared interests, and that he recognizes that this is an assumption that does not reflect the social in all its complexities. However all of his assertions are about agency and structure, which he does a great job moving us through BTW.

My question is more about how we end up in groups that aren't always by our choosing. What happens when we define others, and then those others have to take on that affiliation or reject it. This often happens in both the social, and I see it happen in disciplinarily as well. As a matter of fact, see how Watts characterizes physicists. Now do his stories and definitions define that discipline? Or do we only define ourselves disciplinarity through our participation in it? I seem to think that the first claim is valid because of all the systems in place to protect disciplines, single authored books, discreet histories, etc don't seem to follow the real stories we are getting here. I don't know if this extends or just mucks up your comment, but it got me thinking.

Hmm.. Either? It's not just our particpation, but the eloquence of our participation, stories, definitions, etc., and, how powerful our motives or level/degree of connection it will maintain. Even two people of the same discipline could approach the same subject/issue differently, but are they necessarily opposed?

In the chain of signifiers, the signifier leads to another signifier, leads to another signifier...and we have only half truths; whispers, or a faulty connection. Do we end up in particular groups because we are tempted by desire? to repeat a moment of jouissance?

Your comments, Jen, make me think about what constitutes network membership. How is a network membered? Or how are networks membered differently? And I guess this goes to your thoughts, too, Marcia. Does it matter when/whether we define ourselves disciplinarily rather than being defined our projects or our institution/program pedigree (!) (networkic identifiers we might never absolve)?

Forgive me for leaving clumsy comments; I have a cold--bacteria fashioning a small world of miserable congestion in my cloudy head.

In a sense what your are talking about echoes What Louise (in a seminar class) was talking about yesterday---starting from a contextualist view of the world. I think our projects, Derek, and desires, Marcia, (which sometimes necessitate program affiliations and pedigrees) are the unit/event of analysis. I think networks are membered and de-membered and re-membered as curiousity/interest/inquiry requires and demands--i'm still musing about the randomness this might entail...but this does make me wonder how all this could be used to describe the vastly diverse and numerous composition projects that get floated in the discipline and still can be called comp, how that establishes or degenerates status and power (where the $$$ comes from in an institution), how systems (disciplines in this case?) become stable, when what seems to really wants to happen is for things to move outward and fly off into new foci.

I'm wondering, Derek, if the distinction of membership is something you can only measure from outside of the network itself. At the time of the network, absent a formally constructed one like this class, forms, do the participants identify themselves as "network members"? Maybe so, but it seems more likely that the network is observed, either temporaly or spatialy, from outside, and then the members are identified.

And if the network participants are able to self-define in that manner, is that a different kind of network, and does it function differently?

I'd say we do need to allow for differentiating between networks, Chris. But the small world concept gives us a fair amount of flexibility in that regard, relagating us neither to a macro analysis or a micro analysis. Does that make sense? My first thought in response to the question of network membering is that we have to narrow the terms defining/describing the cluster first. From there, we can determine much more about how inclusion, linkage and ties are distinguished from the absence of those same things. And it would work quite differently in a social network than it would when membership is a direct answer to an organizational need, yes?

I like this question, or distinction among networks that involve self-selection (e.g. applying to a graduate program in a particular discipline) and those that are perhaps less conscious (e.g. you may pick a neighborhood to move into, but you don't usually choose your neighbors) and those that are contingent/temporary (e.g. matching your walking pace to the people around you in the mall, going to see a movie or performance). We might also add networks that we are drawn into against our will (trackback spam, for example).

This makes me think that we need to be careful about the term "membership," which often implies conscious, formal selection, which might not always be the case. It's hard for me to think of susceptibles, for example, as members per se of a given viral network...

At the same time, Chris, I think that you're right to suggest that measurement is a challenge from within a given network, particularly if that measurement is a global one--remember that Watts speaks of our limited, local horizon. Or to put it more concretely, I have no clue how famous I am or how much influence I have within the discipline, nor would I really know how to go about measuring that.

cgb