Mapping and Non-Reductive Critical Practices

Posted by on Nov 12, 2014 in Blog, Modeling/Prototyping, Projects

[Submitted by Amy Robinson]


One of the initial questions which interested me in my early experiences with NewRadial was whether such a radical alteration research practices and argument development would produce a different approach to literary critical practice and interpretation. Given that the approach that we take in research and the development of critical argument shapes the questions that we ask and the results which we produce, NewRadial as a method for scholarly research engages with specific theoretical principles of literary criticism, potentially affording us fertile new perspectives on primary and secondary texts and their underlying relationships. This potential for digital humanities to impact the production of literary studies was evidenced in the concern and skepticism with which early forays into mapping critical analysis of literary texts. For example, Franco Moretti`s distant readings of Hamlet, Our Mutual Friend, and The Story of the Stone, were met with concern and skepticism by more traditional scholars. The primary objection to Moretti’s method and mapping as a mode of literary criticism in general was the reductive effects of data visualization on literary texts. As Lisa Gitelman observes, “few literary critics want to think of the poems or novels they read as “data,” and for good reason” (Gitelman 3). The product of Moretti’s approach, his plot maps, do seem to be distinctly different from the texts which they represent and the critical analysis which  Distant Reading provides seems quite divorced from the engagement with text which is so central to literary studies. However, much of this criticism is a result of reading literary maps as finished products akin to journal articles and monographs rather than as a tool for study which works in tandem with traditional close reading practices. Unlike Moretti’s Distant Reading, NewRadial employs mapping within a workspace, as a mode of research rather than the finished product, but the central concern remains. In reducing a text to nodes in order to establish relational edges, will NewRadial ultimately produce reductive criticism or can the new perspective on text objects which the interface affords users actually enable a more broad consideration of the formal and contextual aspects of a literary text?

Reduction of the literary text in critical analysis is not a concern unique to network analysis. Feminist, Marxist, Psychoanalytic and Historicist readings of texts are often accused of subordinating formal aspects of the text to their own agendas by treating the literary object as a product of its contexts, informed, defined and limited by the patriarchal hierarchy, capitalist system, or historical period the critic is interested in. As Felski observes, “What the literary text does not see, in this line of thought, are the larger circumstances that shape and sustain it and that are drawn into the light by the corrective force of the critic’s own vigilant gaze” (Felski 574). Context, seen as having preceded and produced the literary text, can therefore “correct” the idea that the text is its own authority with regards to its meanings by asserting the text’s own blindness to the manner in which it participates in a larger discourse of which it is only a small part. In other words, the story of a literary work begins with its contexts and ends with its text. A troubling result of this understanding is that much criticism has become focused on rehashing the question of supremacy of text or context. However, if processes of thought and research are indeed informed and shaped by the forms their expression take then it is entirely possible that the creation of this dichotomy is a natural result of a narrative approach to research which forces its elements a linear pattern in which there must a beginning (context) and an end (text). If we were to shift to a network approach to literary studies such as mapping, could we perhaps start a new conversation? One which, rather than debating where the story starts, at text or at context, eschews a linear understanding for a relational one which asks what new meanings are produced when the critic wanders a text’s many relational  edges or paths.

In a secondary research mapping environment like NewRadial, the theoretical implications of the form of argument immediately become clear in the lack of differentiation between primary and secondary source nodes: in appearance, the node that represents either type of source remains the same, and these nodes interact on the same level playing field with one another. Thus the defining characteristic of any given node on a map becomes the relationships that node to the other nodes on that map as they are generated by users. If the NewRadial prototype was regularly used by a knowledge community the impact of community-generated edges would make the theoretical implications much clearer: an exploration of any given text could reveal a vivid and ongoing life of a text object which is not simply a static product of a given time period but continually establishing, strengthening, or challenging relationships between the object, its audiences, and the various data contexts with which it interacts. Furthermore, an approach to secondary scholarship which mapped out existing relationships would literally create areas of uncharted territory for the intrepid literary explorer as well as highlighting areas of contention and debate. Instead of one linear path to meaning with a beginning and an end, networks reveal many paths, many relationships. Instead of one conversation which argues whether the story of a literary text begins with the text itself or with its contexts, a multitude of new conversations as interpretive roads less traveled by are revealed and explored.

As Moretti observed in his plot analysis of Hamlet, “once you make a network of a play, you stop working on the play proper, and work on a model instead” (Moretti 218) and I think this idea of working on a model instead of the text itself is particularly applicable to the exploration of network relationships in digital mapping environments. This became clear to me in my early encounters with NewRadial as I experimented with mapping my arguments instead of developing them through narrative. I realized that in stepping into the map world of NewRadial, I had left the place of the book behind even as I continued to work with my chosen text object. My question then was “what then was I working with?” and Moretti’s work seems to provide an answer. The challenge, then, was to experiment with a model of a literary work in order to determine whether reducing the text to a model within the workspace of NewRadial would ultimately produce a reductive critical analysis or if a network approach could indeed negotiate the questions of text and context in such a manner as to avoid minimizing, distorting, or obscuring the text proper in favor of the model or of the contexts.

At this time I was working with one text which lent itself particularly well to these questions. Charlotte Perkins Gilman’s “The Yellow Wallpaper” is a stylistically complex account of one woman’s descent into madness which also serves as a vehicle for attacking the patriarchal Victorian era medical system and its use of the rest cure to treat “female hysteria.” A piece with such distinctive formal elements as well as an obvious engagement with historical and feminist contexts seemed like the perfect model for an exploration of the relationship between text and context within network arguments. As I began to work the theoretical implications of working with nodal representations text rather than the body of the text itself became apparent. Working with a model of the text which consisted of a collection of nodes representing the paragraphs of the story enabled me to visually isolate parts of the texts which I felt were interconnected, rearranging them in various ways with my secondary sources. This ability to rearrange my primary text with regards to the relationships I was establishing between its nodal model and those of my secondary sources reinforced for me the idea that I was working with a model, something I could change and play with, something I could explore rather than argue with or against. The more connections I made, the more it became clear to me that there were more connections to be made, and the more I broke apart my texts and contexts to peer at them from different perspectives, the more I wandered the paths I created, the more the character of my work changed from researching to support my argument to experimenting with the relationships I had created. The question of context threatening to subvert text or vice versa became for me irrelevant because working with model representations of my data objects to create a relational map did not require me, as a narrative argument would have, to decide on a place from which to start in order to arrive at an ending which had already, in part, been decided through my methodological approach. I could go anywhere.

But what about form? A great deal of the criticism of current context-focused narrative approach to literary studies concerns its reductive treatment of the literary text and its negligence of aesthetic questions with regards to meaning. Do networked or mapped arguments really allow for a more balanced appreciation of both form and context or is the node just another way to reduce form and put it in its place, subordinated to the sociological, historical and political contexts which produced and produce the text? It is my opinion that it does, but it does so within the given limits of the environment as a place to perform initial research and explore secondary sources. Part of this capacity has to do with the structure of the model of the text which NewRadial produces in breaking the body of the text up into nodes, which in the Gilman piece represented the individual paragraphs of the story. The short story’s reliance on the divisions of its paragraphs to communicate various levels of meaning as the protagonist descends into madness become more clear once they are isolated into nodes on map, revealing the way in which the form of the text, its aesthetic characteristics are part of its function within the network. The assertion appears counter-intuitive given that the form of the text has been broken up into a model which in many ways bears little resemblance to the original, but this model gives the user two methods of exploration which a traditional book does not and these allow for the development of a new perspective. The first, which is not unique to NewRadial, is the ability to search the text for key or recurring words or phrases, revealing patterns of language which expose the underlying meaning of the text. Although this kind of search is available in almost any digitized version of a text, it has a different impact here because NewRadial does not work with the text, it works with a model. The result of this is that once patterns have been identified through the search, they can be isolated visually by moving the paragraph nodes in which these patterns occur. Manipulating formal elements of the piece rather than approaching its aesthetics as a whole enabled me to create specific inter and extra-textual relationships with other nodes, establishing a relationship between specific formal elements of the piece and the contexts of Victorian patriarchy which Gilman wrote it to address.

What mapping provides that makes it important, then, is a unique environment wherein context and form can exist in dialogue without being subordinated one to the other. By experimenting with a model of my text, relationships emerged not only between my primary text and secondary sources but also between the meaning and form of the piece. The character of my work changed from researching to support my argument to experimenting with the relationships I had created. My exploration of the aesthetics of the story became one and the same with my exploration of its contexts, as mapping the relational edges revealed the two were interrelated. The more connections I made, the more it became clear to me that there were more connections to be made. My experience began to ironically mirror Gilman’s narrator’s interpretation of the wallpaper: like the narrator, I began my work by imaging patterns of relationship between my nodes but the more that I looked, the more that I imagined these connections, the more patterns seemed to emerge as if on their own, underlying the patterns I was drawing like a palimpsest. The more I broke apart my texts and contexts to peer at them from different perspectives, the more I wandered the paths I created, the more I became lost within the emerging patterns. The map, like the moonlight of “The Yellow Wallpaper” illuminated complex relational patterns which had been previously invisible to me. But unlike the narrator, I was not a prisoner of my patterns but an explorer. Using the NewRadial mapping environment, I created new paths through these patterns which liberated new meanings, alive meanings that grew and changed the more that I worked with them. I had left the place of the book behind only to find that my model was a map of that place that led me down paths of Gilman’s story that I had never travelled before.

Whether a network approach can produce a non-reductive literary critical method which can lend equal weight to questions of form and context remains to be seen. What is certain is that the development of such a critical method which allows for an equal appreciation of form and affective impact as well as context and historical effect must begin with a shift in how we view texts and their relationships and this shift is already taking place within the realm of digital humanities.


Felski, Rita. “Context Stinks!” New Literary History. 42.4 (2011): 573-591. Print.


Moretti, Franco. Distant Reading. London: Verso, 2013. Print.


Gitelman, Lisa ed. “Raw Data” is an Oxymoron. Cambridge: MIT Press, 2013. Print.


Leave a Reply