***** To join INSNA, visit http://www.insna.org *****
I am new to this list, so forgive me (and let me know) if this is off
topic, but the current discussion has brought up an issue that I'm
currently trying to understand. I'm curious about the use of metaphor
and models in research and how they lead to understanding. A strict
epistemological interpretation might hold that even models of atoms
are a metaphor on some level. First, we can never know if our models
of what is out there actually correspond to what exists. This is
brought home by studying the history of science, say physics, where
the model of an atom has changed over time as new evidence has
accumulated, or biology, where the idea of a species has changed
radically over time. Second, however, the models that we use are
always on some level a symbolic representation which we use to
understand and predict the world we live in. Thus, we can still use
Newtonian mechanics to predict aspects of the world, even though we
know it has problems. Yet these models all seem to be in some sense
"strict" in their representation of what exists in the physical world.
The models that were being discussed by Dr. Lanham and others, on the
other hand, seemed to me as more evocative of certain intuitions which
we as humans have about the world and how it works, in order to create
a conceptual framework that may be useful in teasing out processes and
relations that otherwise might go unnoticed. These intuitive models
may be useful for prediction as well, but I can't escape the feeling
that they are qualitatively different in some sense from the models
used in physics.
With the "Kinetic SNA" model, for example, the "melting" of "bonds"
seems to be a very abstract metaphor for a whole series of complex
interactions that may be happening. And, I believe this is true for
more standard applications of SNA: for example, when a researcher
considers the set of communications that one person makes that are
directed to another as a single directed edge between two nodes in a
network--i.e., using ANT terminology, a series of complex events have
been black-boxed or fixed within the model into a single
representation. This feels very different from the physicist's
black-boxing of a phenomenon of a quantum of negative electric charge
into an electron, or even the black-boxing that takes place when
talking about the nucleus of an atom, which has similar complexities
(protons, neutrons, maybe even gravitons? not to mention quarks...).
My question is, if it exists, what is this difference? Is it a matter
of precision of predictions? Does it feel different because the
black boxes are more easily decomposed into their component parts
(also black boxes)? Is it because the component parts have a different
qualitative feel? Or is it as some have put forth (Jonathan Grudin I
believe is one) that we just don't have the quality of data about the
"social" phenomena we study which physicists and biologists have for
their fields of study? If the latter is the case, then what are the
features or affordances of the descriptive accounts of the world which
were made by naturalists which allowed Darwin to be able to develop
his theory of evolution? Or allowed chemists to develop the periodic
table of elements? And do the descriptions of human behavior we are
currently creating have those same features/affordances? For example,
long before Tycho Brahe and Kepler, astronomers knew what kind of data
they needed to make more accurate predictions, they just didn't have
the equipment (telescopes) they needed in order to collect the data
Now, I understand that the degree to which a model corresponds
"strictly" to existing phenomena is probably more a gradient than a
dichotomy, but what are the variables which describe the gradient?
When we talk about species of seagulls along the arctic circle where
gulls in North America can't breed with gulls in Europe, but any gull
population can breed with the gull population immediately next to it,
our variables are the individual birds, their genetic makeup, their
geographic location, etc.
Or is this even a relevant question? Does it matter whether a model
corresponds in any way to "reality" as long as it can successfully
make predictions? My intuitions tell me yes, because many scientists
will reject certain models such as a model that explains gravity by
postulating the existence of invisible apes who are continuously
pushing objects together, and hanging onto one-another with their
feet. It may be true that what we accept as being a good model and
what we reject as being a bad one may be culturally determined, but
usually cultures have good reasons for the rules that they make, even
if those reasons are not explicit or widely recognized within the
culture (except on an intuitive basis). So what are the good reasons
we have for promoting Occam's Razor?
Graduate School of Library and Information Science
University of Illinois at Urbana-Champaign
On 11/10/06, Ryan Lanham <[log in to unmask]> wrote:
> ***** To join INSNA, visit http://www.insna.org *****
> -----Original Message-----
> On Behalf Of Paul B. Hartzog
> Sent: Thursday, November 09, 2006 2:27 PM
> >Saying that social systems do not live is a huge ontological leap.
> >Serious complexity scholars debate various kinds of superorganism
> >social theories.
> For me, dualities of life and non-life create boundaries. Usually those
> boundary objects do not add much to a discussion. They exist to
> reinforce a given ontology rather than attempting to increase commons.
> The difference between an organization and a network rests in the
> propensity of networks to be boundary-less. Organizations emphasize
> identity. Identity is a vehicle of ontology. You strengthen identity
> by reinforcing boundary objects at the edges of ontologies. Recall that
> boundary objects are contested items that are mutually recognized at
> some functional level by actants who apply differing ontologies to a
> situation. Thus, the Golan Heights is a boundary object between Syrian
> and Israeli ontologies. Both know what it is at one level, but disagree
> how its relationships are structured at another.
> Social movements, it seems to me, are attempts are transferring action
> into identity. They form ontologies. The work of researchers is also
> to knit ontologies. It is the extension of relationships, facts, and
> ideas that can be applied to situations without raising a level of
> cognitive dissonance.
> An e-journal is a new attempt to circumvent boundary objects associated
> with print production, editorial expense, and so forth.
> It seems to me that in information terms, a boundary is a field where
> relationship potentials of certain types are neutralized or
> counter-reacted. In computer networks, boundaries are routers or cable
> wrappings, etc. In semantic terms, a boundary is an inflexible
> interpretation of a symbol. The cross means THIS. The Star of David
> means THAT. Artists often create dissonance by playing with strong
> ontologies that are not readily conscious in human interactions.
> Dissonance serves to value or devalue an ontology. How much dissonance
> occurs in situations is the stock price of an ontology. Less
> dissonance, the more likely I am to keep applying a given ontology. The
> more important an identity ontology, the less likely I am to enter into
> situations where I might feel dissonance.
> My argument is that post-Weberian organizations are generally tending
> toward lower identity ontologies. That is because flexibility and
> innovation is becoming biologically more important than stability and
> control. In other words, bureaucracies are giving way to
> actor-networks. I study this in relationship to Community Foundations
> (I have a working paper if anyone is interested I am giving next week at
> I believe that it is a general trend in all highly interactive (i.e.
> trans-boundary) environments. I borrow this idea from ecology in
> relationship to species diversity (e.g. how corridor ecology works...)
> I don't mean this to be shameless self-promotion so much as an attempt
> to move the conversations a bit away from continual reinforcement of
> ontologies toward attempts to find commons (Marilyn Strathern's 2004
> Commons and Borderlands on that point.)
> What do I read on this?
> Anthropology of globalizations (e.g. Appadurai, Aihwa Ong, Charles
> Tilly, Mol, etc.)
> Ecology and systems biology...especially Island Press-like Ecology
> policy--e.g. Holling and Gunderson's version of Panarchy, Corridor
> Ecology, Island Biodiversity research, etc.
> Actor-network-theory of categories and borders (e.g. Marilyn Strathern,
> Bowker and Star, Latour, etc.) I am particularly interested in the
> notion of decentering humans in networks.
> SNA--especially of the ilk that follows on Harrison White's insights. I
> am particular interested in SNA as a discussion of crossing boundaries,
> borders, cultural lines, civic elites, etc.
> Experimental and evolutionary economics--because they are attempting to
> get at motivations--which is necessary to understand identity formation
> values and the value of leaving identities and moving toward new
> territories (both real and intellectual).
> International affairs and social theories related to post-Westphalian
> policy matters--the decline of nations and nation-states due to
> migrations, etc.
> Revolutions, diffusions and innovations of all sorts--both agent-based
> models and cultural ones.
> I personally think all this could be organized (problematic) under
> post-Weberian organizational theory, but it is more important to
> maintain commons than identity--my very point.
> I am always interested in where these discussions are occurring because
> it is very hard to know how to not be tied to discipline on these issues
> which, given the above, is obviously highly problematic. Any help,
> pointers, side-comments or co-travelers appreciated.
> 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.
Check out the unofficial GSLIS Wiki:
Tell me what you think, if you find it useful, or if you have any
ideas for how to organize it better. And if you feel comfortable
doing so, I heartily encourage you to contribute content!
This GSLIS is the Graduate School of Library and Information Science
at the University of Illinois at Urbana Champaign.
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.