***** To join INSNA, visit http://www.insna.org ***** Dear Ryan, We come from a different background. I have been fascinated with the structure in aggregated journal-journal citation relations. These are large networks with fine structures which are reproduced from year to year. Stability is prevailing and change the exception. For example, I have tried to measure areas of change using probabilistic entropy measures: where is the network "hot"? In other domains, there may also be more stability than one thinks if one chooses a more abstract unit of analysis--or, perhaps better, a unit of operation. If one looks at variation in relations, one cannot find the patterns. Otherwise, we agree about the hypothetical nature of the various measures for change, structure, and stability. Best wishes, Loet ________________________________ Loet Leydesdorff Amsterdam School of Communications Research (ASCoR) Kloveniersburgwal 48, 1012 CX Amsterdam Tel.: +31-20- 525 6598; fax: +31-20- 525 3681 [log in to unmask] ; http://www.leydesdorff.net/ > -----Original Message----- > From: Ryan Lanham [mailto:[log in to unmask]] > Sent: Monday, January 08, 2007 9:03 PM > To: 'Loet Leydesdorff'; [log in to unmask] > Subject: RE: Networks and conformity > > Loet: > > You are mixing, in my opinion, a realist and a constructivist > discourse about systems. From my perspective, the systems > remain constructs. The hypotheses are disturbed by the > observations. Thus, it becomes important to develop measures > when the variation exhibits more systemness or not. > However, this presumes specification of the system(s) of > reference. I am not intending to set boundaries. We can > quietly leave that to management and politics. > > Ryan: > > I'll try to summarize (not too succinctly) what I think and > what I think I said: > > 15 Assertions about the Philosophy and Future of SNA > > 1. Networks are a form of classification method. Membership > is constructed (i.e. perspective-based). There are no > natural networks just as philosophers have found no natural > classifications. > > 1A. The primary purpose of SNA is to create discrete bounded sets. > > 2. The underlying premises of SNA in establishing sets are > dichotomies between linkage and actor and between actor and > context. These dichotomies were theorized to enable discrete > mathematics to be applied to networks in the form of > measurements. They also followed cybernetic models that > simplified computing networks. > > 3. Measures (e.g. centeredness) are expressions of relationships > (ontologies) made from a given perspective. Because the > underlying data is constructed, the measures are constructed. > This is not an assertion > that they are useless--only that they are cultural. > > 4. Stability and consistency are the primary problems of > network membership because the hoped-for structure is > non-existent. Systems are not predictable and membership in > classification schemes is problematic. > > > 5. Assertions about lasting stability refer to "structure." > For structure to exist there would need to be stable > classifications. > > 6. The problems experienced with stability of membership (or > relationships) are quite similar to those previously found in > artificial > intelligence. > > 7. The standard response to such problems (hoping to maintain discrete > sets) would be probabilistic memberships. > > 8. The response after this one is typically to look into how > cultures evolve (e.g. the Boyd and > Richardson/Bowles/Bienhocker, Gunderson and Hollingesque > Panarchy and other Santa Fe sorts of works). The result is a > turn to "complexity studies" where cycles and long-term > trends are sought in the evolution of "structures." > > 9. Increasingly there is a crisis between stability > (structure) and change (innovation) where innovation is the > definition of all factors > acting as solvents to social stability. > > 10. This crisis seems to evolve in the same sorts of cultural > relativism problems and clash of structure issues that have > been under consideration in anthropology for a > generation--resulting in innovators like Bruno Latour, Saskia > Sassen and Aihwa Ong amongst others. It also seems to spawn > structural reactionaries like Samuel Huntington and fence > sitters like Francis Fukuyama. > > 11. SNA will tend to move toward network dynamics and deep > classification description in response. Within 3 years, > publications of static social network research will be > considered passť. > > 12. New forms of technology analogous to, say, Feynman > diagrams, will emerge that describe the relationships of > actants to contexts. > > 13. Qualitative methods will enjoy a period of notice as > means of thick description in networks are explored. These > will, as usual, be more broadly rejected as unworkably > complex and subjective but they will create anxieties and > awareness about data sources and constructiveness. > > 14. The future of business SNA will be a race toward mass > customization of the evolutionary properties of specific > networks--e.g. how a consumer's decisions evolve--based on > aging children, personal age, life situation, etc. Or, how a > medical history evolves. The key will be to model people > based not on their simple past decisions but how their past > decisions changed...and when...the goal (as always) will be to more > comprehensively anticipate needs JIT a la TiVo. > > > > _____________________________________________________________________ SOCNET is a service of INSNA, the professional association for social network researchers (http://www.insna.org). To unsubscribe, send an email message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message.