Social & Information Networks

Social & Information Networks: A Brief History

Social networks as a method, have had a pretty significant series of conversions in use, including the conversion of it from a set of theories, methods, into an entire discipline. There are now several PhD programs that either explicitly consider “Network Science” to be a part of complex systems analysis that bridges endogenous actors in social theory with materialist views from physics, and there are programs which actually name themselves “Network Science PhD” programs; both of these tend to avoid notions that they are strictly “social” or “physical” sciences, in a similar way that often iSchools tend to avoid claiming that they are strictly either informational or social.

As examples, Barabási is an evangelist of Northeastern’s Network Science PhD which is also strongly connected to Central European University’s Network Science PhD. There are also those that consider social networks to be strictly a set of methods to deal with endogenous behaviors of individuals and they are not actually social theory. This has a much longer history in sociology than the network science programs that have developed less than a decade ago. Thus, social networks is a relevant topic that developed which has a range of views about it’s epistemological and methodological nature. There are even multiple significant accounts of its development.

Mathematicians and Computer Scientists are quick to argue that Euler’s Königsberg bridge problem is the beginning of network theory, as the basis of networks are mathematically grounded in Graph Theory and Combinatorics. If one comes from sociology, anthropology, or economics, then perhaps they are more likely to align themselves with the idea that mathematical graphs do not correspond precisely with social network theory.

For example Actor-Network Theory, largely accredited to Bruno Latour, is a network structure of society that does not entirely endorse the intensional completeness of a mathematical node or edge under certain circumstances. As another example although sociologists might in Social Network Analysis might have borrowed Graph Theoretic concepts, large portions of the epistemic views and causal problems related to them are exclusively beyond the typical mathematician’s ontology. Many Graph Theorists, for example, ask different kinds of questions related to properties of n-dimensional graphs, or statistical properties with a history dating back to Erdős–Rényi random graphs, which flies in the face of Barabási, Mark Newman, and other network scientists claims of physical law-like structures related to “scale-free”, “log-log”, or whatever flavor of that kind of distribution that is statistically found.1

So, it suffices to say, there are a vast number of differing opinions on how we should proceed in speaking of the study of or the use of networks. But, here in particular, I will view network structures as a methodology to view agents in endogenous systems of decision making.2 Albeit, I have yet to see reason to confine myself to saying that these individuals necessarily must be human. Many social network analysis scientists might say this is problematic, it is not if one is an ANT scholar 3 or one requires actions to be documented in some way for it to be “observable” such as in citation, websphere, hyperlink, or some other kind of informational network analysis that explicitly or implicitly invokes social behavior.4 5 6

Underlying Notion of Identity: An Open Discussion

To me, the notion that non-human nodes in a network cannot exist as a part of endogenous social influence, seems like a strange constraint that has its basis in assuming Cartesian reasoning. I have written a little about this before in a prior blog entry. I would like to expand this a little.

If we consider that by accepting that social networks assume at some basis that our interactions with others are endogenous to the structure of our network then it seems very easy to see how one might claim that my thoughts are somewhat endogenous to my neighbors’ thoughts even if we are not immediately accessing their thoughts. This is more of the realm of Spinozian thinking than Descartes 7, and has been discussed more at length by Gilles Deleuze.8 9

Surely, the notion that thinking being tied to the material individual due to personal experience does not apply here as one could easily assume that the experience is partially explained by material that is not ourselves. In order to speak of individual nodes having unique experiences and knowledges in this kind of endogenous society would require us to speak of clashing bodies which contain minds definitively, and to practically define “human minds” as being entirely embodied in a human.

So to say we have “common experiences” in an exact sense would then be a different node, and furthermore our perceived identities having contradictory experiences are diminishments and developments of moral persuasion as Deleuze would argue. Thus confronting the multiple personalities of “voting with our feet” notion (as I spoke of in the prior blog entry) in an online space may make it more likely for us to consider a more practical heuristic of having multiple minds in online spaces. If this is true, then using documentary practices of observing the “self”, “mind”, or “identity” might be more true to how we socially develop than the denial of “social” identity in documentary practice.

In this way, one might be able to appropriate most notions of classical SNA literature such that it includes other kinds of network analysis, such as many information networks, citation networks, and can even become more granular to the notion of identities of heuristics as loosely explaining how we use theories, methodologies, or memes to develop a notion of consistent identities that individual humans can all embody without being classified on some level as “schizophrenic” or broadly irrational. At the same time, it seems that this would allow for us to speak of nodes at multiple levels such as how Latour might wish us to consider “nodes” without entirely having to abandon other more computational methodologies.

If this is possible, it would certainly create some interesting outcomes in methodological interests and interpretations. At the moment, this is something that I would hope that a social network minded person would not entirely discount as being too progressive of a scientific claim. I believe that the philosophical framework exists to at least begin this discussion seriously with a larger audience.


Image Attribution: The Computer History Museum

  1. Adamic, Lada A. “Zipf, Power-Laws, and Pareto-a Ranking Tutorial.” Xerox Palo Alto Research Center, Palo Alto, CA, Http://Ginger. Hpl. Hp. Com/Shl/Papers/Ranking/Ranking. Html, 2000. 

  2. Ackland, R. (2013). Social media networks. In Web social science: Concepts, data and tools for social scientists in the digital age (pp. 48-77). London: SAGE Publications Ltd. doi: 10.4135/9781446270011.n3 

  3. Latour, Bruno. Reassembling the Social: An Introduction to Actor-Network-Theory. Clarendon Lectures in Management Studies. Oxford ; New York : Oxford University Press, 2005., 2005. 

  4. Wallace, Matthew L., Vincent Larivière, and Yves Gingras. “A Small World of Citations? The Influence of Collaboration Networks on Citation Practices.” PLOS ONE 7, no. 3 (March 7, 2012): e33339. 

  5. Baumgartner, Jason, and Tim A. Waugh. “Visualization of Roget’s Hyperlinked Thesaurus,” 2002. 

  6. Rogers, Richard. Digital Methods. Cambridge, Massachusetts: The MIT Press, 2013. 

  7. Spinoza, Baruch. The Essential Spinoza: Ethics and Related Writings. Hackett Publishing, 2006. 

  8. Deleuze, Gilles. Expressionism in Philosophy: Spinoza. Translated by Martin Joughin. First Paperback Edition edition. New York : Cambridge, Mass: Zone Books, 1992. 

  9. Deleuze, Gilles. Spinoza: Practical Philosophy. Translated by Robert Hurley. First Edition in English edition. San Francisco, Calif: City Lights Publishers, 2001. 

You might also enjoy