Complexity and Social Networks Blog

... thus, with the formulation described, we have both stocks - the network structure - and flows - the 'relational flows' described. This is surely of the essence, if we are to have both elements that human cognition seems to require for drawing a complete picture - structure and flow.

BTW, one clarification on my posting. What I mean by behavioral relational flows is just regarding how relational behavior changes over time, not what "flows" over a relationship (e.g., information, exchange, etc).

Yes, and that puts front and center a choice, as we think about SNA as a tool that may offer dynamic analysis (beyond comparative statics): In such dynamics, what will be the flows? the now-traditional information and exchange? or ‘relational flows’?

My take on relational flows: the state of a given relationship at snapshots across time. How close am I?

With that definition, I am tempted to relational flows as the ‘flow’ for analysis. Or, do ‘good’ dynamics need to fit in both sorts of flow?

Then the next question could be, What limits and opportunities, conceptually and analytically, would a given formulation present?

I'll use David L.'s posting as an excuse to plug this work a bit more. As a student of Harrison White's I was constantly confronted with the idea that structure blocks action--that one's position in an enduring structure of obligations and expectations thwarts attempts at behavior which does more than reproduce that structure. Being contrary I went on to turn this on its head, and to write two papers on how action blocks structure--or, to be more precise, how constraints on action complicate the translation of structural propensities (e.g. to socialize with particular people) into tie enactments (actually socializing). One is the paper David cited, where scheduling constraints (if you're here you can't be somewhere else) intercede between a latent network of propensities to spend time with particular people and the enacted network of encounters, along with diffusion based on that. The other is

Gibson, David R. 2005. "Taking Turns and Talking Ties: Network Structure and Conversational Sequences." American Journal of Sociology 110:1561-97

in which the action constraints are those rooted in conversation, particularly the one-speaker rule and sequential constraints related to the reactive nature of conversational talk.

A third paper explores the same idea in a very different way. A thought can be imagined as a network of discrete ideas, where ties signify implication. Thoughts are turned into verbal action (speech) through the process of linearization, involving grammatical rules and psychological constraints and conversational norms. Given enough time these constraints might not prevent someone from speaking their mind but in conversation other people can interrupt, in consequence of which one’s heard utterance may be a distorted version of what was intended. The paper, which technically should be in print by tomorrow but presumably won’t be, is:

Gibson, David R. “Opportunistic Interruptions: Interactional Vulnerabilities Deriving from Linearization.� Forthcoming, Social Psychology Quarterly (December 2005 I think but otherwise look for it in March).