*****  To join INSNA, visit  *****

Agreed, "social" should be defined much more broadly than "knowing".  A 
better definition of social network analysis might put the emphasis on 
"relational" analysis.  From this perspective, approaches that define 
groups by shared attributes (e.g., sex, race) would not be a form of 
social network analysis, but approaches that define groups in terms of 
shared ties would be.

A nice review of the definition and measurement of relational ties is the 
Methodological Appendix of Claude Fisher's "To Dwell Among Friends".

re: Georg Simmel

What defines being social for Simmel is the super-individual quality of a 
collective (Simmel 1950 (1908), p. 123). A group takes on this unique 
characteristic when its existence cannot be linked to the loss of 
particular members. When there are only two people involved, there is only 
one tie, and the group can be dissolved with the loss of either person, so 
for Simmel, the minimal social group is three persons, also the first 
opportunity for structural variation in the ties -- a "two-path" or a 
completed triangle. In his work on Tertius Gaudens (the "third who 
enjoys"), (thanks to Jim Moody for this paragraph).

On Tue, 15 Apr 2008, Blyden Potts wrote:

> *****  To join INSNA, visit  *****
> I would not define a social relationship so narrowly as "knowing" one
> another. To offer a most obvious example, would you contend that anonymous
> donations are not a social relation merely because the persons involved do
> not know one another? Indeed, it would seem that the portion of any one
> person's social network constituted by those alters whom they know is but a
> fraction of their overall set of social relations, though just what fraction
> that may be depends on how one defines "know". If I recognize a clerk at the
> grocery store as the same one I've interacted with so many times in the
> past, but know virtually nothing else about her, do I know her?
> There is no quality of kinship or friendship or neighboring that defines
> them as a collective type (i.e. "social relation") in exclusion of economic
> or political or other types of social relationships, except the absence of
> whatever additional criteria define those other relations as specifically
> economic, political, etc. They are not, for example, defined by the
> "personal" nature of the relation as opposed to those that are "impersonal".
> A definition of "social relation" as those relations that are not economic,
> political, etc. is either a residual construction defined by filtering out
> the other types of social relations or else an arbitrary aggregation of
> specific types (e.g. kinship, friendship, neighboring, etc). So defined, it
> lacks an elegant logical basis as its own concept.
> By relation I mean that that (at least) one object influences or shares
> something in common with the other. By social I mean that which involves
> more than one agent, as opposed to inert physical or symbolic objects. Thus,
> by social relationship I mean a circumstance in which an agent influences or
> shares something in common with one or more other agents. Understanding
> "social relation" in this way provides the elegant basis and makes clear
> that economic, political, etc. relations are particular subsets or forms of
> social relations.
> A bureaucracy is a complex set of social relationships. The existence of
> "impersonal rules" is secondary, a guide for those social relationships, not
> a definitional criterion independent of those social relations.
> Again, the insight of social network analysis is in understand social
> relations as sets rather than in isolation from each other, in focusing on
> the patterning of the sets (e.g. shape, etc), and finding ways to analyze
> the relations as sets (i.e. in symbolic relation to one another), not in
> identifying some new "form" of social relation.
> Blyden Potts
> -----Original Message-----
> From: Social Networks Discussion Forum [mailto:[log in to unmask]] On
> Behalf Of Jeffrey Broadbent
> Sent: Tuesday, April 15, 2008 9:33 AM
> To: [log in to unmask]
> Subject: social organization = social networks?
> *****  To join INSNA, visit  *****
> I would like to take issue with Blyden Potts' definition of social
> organization as entirely constituted by social networks.  There are many
> types of interpersonal and inter-group relationships, only some of which are
> "social" (defined as "knowing" the other).  For instance, a bureaucracy can
> involve the application of impersonal rules without knowing the role
> inhabitants.  A war can kill people without impersonally.  There, it is
> accurate to say that social networks in the specific sense of social
> relationships most prominently defined by kinship relations is one
> fundamental form of social organization.
> Jeffrey Broadbent
> Abe Fellow
> Visiting Researcher, Faculty of Law, Keio University
> Associate Professor, Department of Sociology
> and Institute of Global Studies
> 909 Social Science Building
> University of Minnesota
> 267  19th Avenue South
> Minneapolis, Minnesota
> USA 55455
> Tel. 612-624-1828
> Fax. 612- 624-7020
> Email: [log in to unmask]
> Webpage:
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers ( To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.

  Professor of Sociology and Statistics
  Director, UWCFAR Sociobehavioral and Prevention Research Core
  Box 354322
  University of Washington
  Seattle, WA 98195-4322

  Office:	(206) 685-3402
  Dept Office: 	(206) 543-5882, 543-7237
  Fax: 		(206) 685-7419

[log in to unmask]

SOCNET is a service of INSNA, the professional association for social
network researchers ( To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.