Research Report: Becoming a Nazi: A model for narrative networks

By Amanda Phillips, Untitled Flight Paths Project

1. Abstract.

   This article explores the potentials of network analysis applied to narrative structures. Using self-reported “Becoming Nazi” tales, Bearman and Stovel map narratives as networks and investigate structural features uncovered by these stories. The phenomenon of Nazi identity is often studied in the field of historical social science, but this study attempts to make an intervention by shifting the question from “why” people became Nazis to “how” they did so. Narrative network analysis reveals structural differences between “becoming” and “being” – a phenomenon that they attribute to a loss of subjectivity.

    

2. Description.

   This study of narrative comes from the perspective of historical social science, which is surprising given the historical roots of narratology and sociology. However, Bearman and Stovel make the contention early on that narratives are essential to their field, that as a discipline concerned with “a plausible account of human action in historical context,” the stories that lie at the center of identity formation are key to the questions that they ask as historical sociologists (73). The authors suggest that personal narratives are the key to explaining how individual characteristics are translated into observed actions; they claim that previous attempts to bridge this gap fall back on “abstractions” and the “non-measurable effects of 'culture', 'symbolic constructions', or 'cultural agencies'” (74). (Here, their methodological isolation from humanities disciplines is quite palpable!) Bearman and Stovel draw their story from a database of becoming-Nazi stories submitted for a competition in 1934.

    

   The choice of network analysis to study these stories is firstly an acknowledgment that networks and narratives have similar structures. Bearman and Stovel point out that “narrative, historical and network data are locally dense, often cyclic, knotted, and characterized by a redundancy of ties” (71). They express the hope that applying network methods to narratives will help them answer the questions of “how” and “why” people became Nazis before the party seized power and political affiliation became necessary. The first step in the process is coding the narrative according to the logic of networks: “This is accomplished by coding as a node every discrete element within the narrative and then representing the author's explicit connection between elements (narrative clauses) as arcs linking elements” (76). Next, they look at the structural features of the graph to determine what, if anything, the network map can reveal.

    

   Their most significant finding was in the difference in structure between the “becoming” portion of the stories and the “being” portion. In order to draw conclusions across narratives, Bearman and Stovel abstract specific narrative moments into event types – macrolevel, local, and cognitive. The composition of becoming and being narratives were similar, with each type of event distributed in comparable ways, but the arrangement and density of these units were quite different. Becoming narrative networks were much denser than being stories, to the point that performing a reduction on the elements revealed a cohesive chain in becoming narratives but very little cohesion in the being stories. In fact, the difference was so marked that Bearman and Stovel claim that the being stories lack narrativity altogether.

    

   By assessing network centrality, Bearman and Stovel are able to pinpoint the importance of individual nodes in the stories. They use a measure of centrality known as power centrality, which measures the centrality of a node only with respect to the centrality of the nodes to which it connects. Using this measure of centrality, Bearman and Stovel find what they suspect are themes otherwise concealed by the Nazi's own account of her journey to becoming a Nazi. Some of these themes include the decentralization of “traditional bases of identity” in the potential Nazi's life and an increase in mobility – in other words, a certain destabilization of identity, which is presumably remedied by the rigidity and order of the Nazi party. Meanwhile, anti-Semitism is not very central according to this measurement, even though it appears frequently in many of these texts.

    

3. Commentary.  What Bearman and Stovel offer is an alternative way of thinking about what networks can mean. Seeing a network analysis methodology applied to narrative has important implications for a project that seeks to measure the structural integrity and unspoken hierarchies of a networked novel. The metrics that Bearman and Stovel use and their subsequent analysis of the results suggests important ways to investigate Flight Paths: while centrality measurements can determine the true stars of the Flight Paths universe, network density might uncover hidden narrative progression in a seemingly disjointed text. The purpose of mapping a narrative as a network is to uncover the patterns obscured by the author's projection of her own theories.

    

   The extent to which this methodology accomplishes something new, however, is still uncertain. Even though Bearman and Stovel sought to avoid attributing causality to nebulous social forces at work, their conclusions seemed to do just that. The fact that their network diagram revealed that instability in life caused many potential Nazis to seek out the rigid domination of a fascist political party suggests that network diagramming may not be any more effective in determining concealed meanings than, say, a psychoanalytic treatment of these same texts. On the other hand, one can also say it is just as effective as alternative forms of analysis and provides a solid quantitative base to appease the scientific mind. This may come down to a matter of taste. Thinking optimistically, however, the results of the study strongly suggest that narrative network analysis warrants more scrutiny to determine if truly new insights can result. Even so, despite the claim that their methodology can be applied to all narratives, it is likely that converting even the simple “literary” texts into a node and arc structure would be extremely challenging given the density of language typically employed in texts of this kind.

    

   Applying the methods directly to Flight Paths is not necessarily the way to test the method in its current form, however. There are major discrepancies between Flight Paths and the Nazi stories in terms of the content and structure of each text. The becoming Nazi stories are short, composed by a single author within a single medium, and focus on a relatively narrow topic. Flight Paths, meanwhile, has multiple authors, multiple types of media components, and very little in the way of a unifying theme. Like the being Nazi portions of the Nazi stories, a straightforward network diagram of Flight Paths would likely reveal a lack of narrativity in the piece. This means that in order for the methodology to be useful, variations must be made to the analytic strategies proposed by Bearman and Stovel. For example, networks maps of different types of content will be necessary: perhaps a network for narrative, a network of thematics, and a social network of relations between contributors of the Flight Paths universe. The project is already divided into discrete units that can easily serve as nodes for a very basic map.

    

4. Resources for Further Study.

    

   Bearman, Peter S. and Katherine Stovel. “Becoming Nazi: A model for narrative networks.” Poetics 27 (2000): 69-90. ScienceDirect. Web. 11 Feb 2010.

    

   Franzosi, Roberto. Quantitative Narrative Analysis. Thousand Oaks, CA: Sage, 2010.

    

   Smith, Tammy. “Narrative boundaries and the dynamics of ethnic conflict and conciliation.” Poetics 35.1 (Feb 2007): 22-46.