The Body in Flannery O'Connor's Fiction: Computational Technique and Linguistic Voice (review)

Donald E. Hardy's innovative book performs an exacting investigation into representations of the body in Flannery O'Connor's prose. The crux of his study lies in the bodily nexus of the sacramental, incarnational, and grotesque. The body in O'Connor's work is a well-worn critical path to be sure, and it is in a way surprising to discover another study on the subject. Yet, Hardy's book proves delightful and daunting insofar as it opens fresh methodological possibilities for studying not only O'Connor's fiction but also anyone's fiction.

To the familiar territory of the Christian Southern Gothic, Hardy brings his unique blend of linguistic and computational expertise. The premise of the book is that while so much has indeed been made of the body in O'Connor's fiction, the existing studies treat the body in an almost purely thematic—or in Hardy's words "abstract" (5, 145)—manner. In contrast, Hardy sets forth a meticulous stylistic analysis of language involving the body in O'Connor's works, as he "looks to find linguistic form and/or semantics in the text as object of discovery" (146). In the process of mining O'Connor's texts for empirical data, he drills down to the level of single words and groups of words, finding meaning in the arrangement and distribution of the elements of O'Connor's language.

The book is designed to address scholarly audiences with specialties ranging from O'Connor to Southern Studies to linguistics to literary theory. As a result, a good deal of Hardy's writing is foundational: after the introduction, he devotes a chapter each to a primer on grammar and to one on computational statistics. Two analytical sections follow: one on body parts and body actions and one on reflexive pronouns and the "bifurcated self" (107). In the chapter on grammatical voice, Hardy patiently teaches his readers about the grammatical structure and semantic function of the middle voice, which is neither active nor passive but lies somewhere in between. He explains that the middle voice refers to "a verb whose subject is affected by the action of the verb without being an explicit passive goal" (29). In this way, the subject of a sentence retains its grammatical identity—it does not become an object of the sentence per se—while at the same time serving as the goal of the verb's action. Hardy emphasizes this particular construction in O'Connor's language because it raises the question of agency—or rather it questions who possesses agency in the narratives (40, 46). The ambiguous status of the agent in the middle voice illuminates the struggle for control between body and spirit that defines so many of O'Connor's characters (49).

To say that Hardy blends literature and linguistics is only partially adequate. To these he adds a third component: computers—and the capacity that programming holds for deepening and widening the empirical and interpretive study of texts. Through the use of his own concordancing program, Textant—an acronym for Text Analysis Tools—(52), he has compiled extensive data on terms denoting body parts and on verbs describing actions involving body parts. These data are generated by various computational functions including frequency, collocation, and range. Frequency is a count of the number of times a particular body-part term appears in O'Connor's texts (58). Collocation is a measure of the "frequencies of combinations of words" (71), and range determines the distribution of word-occurrences across different segments of the texts (76). While the formulas Hardy covers are nearly impenetrable to an untrained reader, he provides an abundance of data-rich tables that are accessible and revealing. For example, the most frequently occurring words in O'Connor's fiction are "face," "Jesus," "head," "eyes," "mouth," "foot," and "Bible" (81). Even a cursory reading of these statistics supports Hardy...