There are two methodological layers to So and Long’s essay, though to use this language is to imply a hierarchy which is not necessarily correct. To clarify, I will propose that the investigation of network analysis and the application of network theory to writers and texts are both equally important parts of the whole. To lay one over the other is create a more complete image of both. It might then be more fitting to consider both network analysis and the reading of 20th century poetry So and Long accomplish using it to be transparencies that each contain half the picture. This analogy may seem laborious, but there is no better way to demonstrate the method of the essay than in crafting it. “Network Analysis and the Sociology of Modernism” is a work designed to refute hierarchies and establish the importance of a more rhizomatic approach to literary analysis. It is an essay all about looking at the whole picture.
Network analysis itself is a big part of the method here and is therefore a fine starting point for understanding what is going in this piece. So and Long begin their survey of the method by illuminating its heritage in sociological research. They write: [W]e engage with the empirical in order to perform “distant readings” that put mathematical interpretations of social structure into dialogue with thick historical description and close reading of cultural material, resisting any kind of strict bifurcation of distant empirical explanation and close hermeneutic interpretation” (155). Even within one layer there is a great deal going on. First, the essay makes the usual appeal to the divide between the empirical “distant” and the analytic “close.” This is almost a standard invocation at this point, much like a plea to the Greek pantheon at the beginning of an epic poem. This divide is, as always, sternly rejected in a scheme that promises to integrate both approaches for the common enrichment of both schools of thought. It is important to see this happening but also to look beyond it. To elaborate, So and Long have here placed quantification into the humanistic realm by associating with social and cultural analyses (specifically the labor of sociology). This creates a “we are not so different you and I” ambiance that allows the head-splittingly numerical portion that follows this claim to flow more smoothly. I mark this choice because there is more than one way to go at this juncture. One can point to the benefits of mathematical theory or computer science or algorithmic reading as a way into a better understanding of humanistic objects. But So and Long’s choice to point to Sociology as the originator of their method (despite the fact that there is a case to be made for three I list and more) is one that makes quantification a lot easier to swallow.
Of course, this does not ameliorate the need to describe what exactly network analysis is to a readership of humanists. The definition offered in the “Explanation of Method” section of the essay is as succinct and plain as one could ask for: “Formal social network analysis stems from the basic idea that relations between objects can be abstractly represented as a set of points (or nodes) connected by lines (or edges). These lines indicate some kind of relationship between the nodes they connect” (156). What happens from here is a dedicated analysis of early 20th century poets and their social networks that makes use of this basic theory to draw conclusions about the importance of being a part of these networks. There are two things that are important to note here. The first is from the moment this definition is established and throughout the next section, So and Long remove network theory from the realm of abstraction by couching their discussions of it constantly in the language of poems and poets. This method ensures that their readers are thinking about the objects that they always have (like the poems of Ezra Pound) and not some informatics theory they only just heard about two pages ago. The second thing to note is that this work is always questioning the importance of a few exceptional authors. It is a common move in literary history to offer up exemplars that rise above the rest of their contemporaries to distinguish themselves as the greatest artists of their time. Literary scholars use these exemplars as guideposts for launching inquiries and crafting theory. This essay demonstrates (as many other disciplines and critical approaches have done) that de-centralizing literature and removing the established hierarchies of its authors and texts can provide a more complete view of all the authors and texts that give rise to those previously valued. This is not simply a Moretti-esque employment of quantification to define exactly what “high modernism” is using large swaths of data. Rather, it is a deep investigation of the people and places that constitute the movement. The promise of the first section that no priority be given to close or distant reading is fulfilled in the analysis of the poetic network. The discussion of brokerage and closure works as both a potent advertisement for the method’s effectiveness and as a compromise between close and distant reading. To elaborate, the closure that defines the more successful nodes of the network is only discernible by attending those nodes closely. However, the brokers that bring those enclosed nodes their relational success are important actors only in a larger scheme.
It is fitting to conclude with counterfactuals because they do much to demonstrate the benefit of the method but could not have done without the work of the essay. Networks are effective analytical tools, more effective perhaps than close reading or archival work, because one can play with their constitutions and introduce counterfactual circumstances. It is possible to view a hypothetical network in which one poet or another is removed and see their importance through their absence. The same cannot be said by, for example, a reading of The Tempest that subtracts all of Prospero’s lines. Thus So and Long make plain the benefits of network analysis while simultaneously drawing new conclusions. It is an elegant encapsulation of both the method and their thesis.