At the end of his final chapter Garrett Stewart remarks on Hardy’s refusal in Tess of the D’Urberviles (1891) to claim any redemptive meaning for the violent sacrifice of his heroine. Instead, “what is saving in Hardy is the difficulty of skimming . . . the need, so often, to reread his very sentences to get either their exact picture or their point” (219). Novel Violence is itself dedicated to making skimming difficult, both in the intricate density, sometimes even obscurity, of its own writing and in its insistence on slowing the pace of reading by attending to small-scale aberrations of syntax, figuration, sound effects of alliteration and assonance, or the sort of phonetic ambiguity that marks the moment of the villain’s violent annihilation in Dombey and Son (1846-48), as the doomed Carker “uttered a shriek—looked round—saw the red eyes [of the approaching train] close upon him . . .” (7). (Stewart’s italics, a frequent tactic, here used to underline the double sound and sense of S/Z, adjective/verb.) Dwelling or stumbling on such effects retards any straightforward reading for the plot with the resistance of “prose friction.”
Stewart has been a perceptive and ingenious reader of minute details throughout his distinguished career as a critic of nineteenth-century British fiction, as well as in his studies of film and painting. Now, in addition to scrutinizing four major Victorian novels, he also strenuously theorizes this microstylistic approach as “narratography,” setting it in dialogue with the macropoetics of structuralist narratology and the genre theories of Ian Watt, Michael McKeon, and especially Georg Lukács. Three sections, 20% of the book, as well as lengthy footnotes and theoretical excursions within the other chapters, are devoted to these issues. Stewart proposes his sort of “novel criticism as media study,” working into the fine grain of the linguistic medium, as a methodological alternative to new historicism and cultural studies, but for reasons I’ll return to, his book is unlikely to exert the sort of influence he hopes for (220).
However we take his theoretical claims, Stewart’s readings of his chosen texts are always original and striking. Little Dorrit (1855-57) assumes a different shape when read for its “omitted person plot,” the missing story of Arthur Clennam’s actual mother, whose existence is briefly mentioned only in the novel’s jumbled denouement. Stewart fills this lacuna with an imagined prequel like those recently manufactured for many other Victorian novels, a device that is both witty and illuminating as a gloss on the foreshortened syntax of the final paragraph, which assimilates Arthur to the mothering role played by Amy Dorrit: “not being able to love the mother, or mourn her, the son has become her” (49). A similar shift of perspective displaces the melodrama of Anne Brontë’s The Tenant of Wildfell Hall (1848), where Stewart’s focus on the exchanged journal and letters by which the story is framed discloses another process of transference, one that reflects the novel’s commercial and libidinal transactions with its audience: “Brontë is narrating the story of fictional impact itself” (100).
The chapters on The Mill on the Floss (1860) and Tess of the D’Urbervilles grow more ambitious, largely through Stewart’s greater use of Lukács’s Hegelian idiom to link small-scale features with large generalities about temporality in the nineteenth-century “novel of disillusionment” and the relations of form and force. Registering the forward thrust and lateral disturbances of the linguistic medium, “narratography is a reading of form for its own temporal force” (169). But here the limitations of Stewart’s approach become more troubling. The violent deaths of Maggie Tulliver and Tess Durbeyfield amplify the earlier sufferings of Helen Huntingdon and Arthur Clennam’s nameless mother and might seem to consolidate a unifying theme for Novel Violence, but Stewart is actually not much concerned with violence at the level of character and event. He repeatedly invokes Roman Jakobson on a “violence purely linguistic” and Lukács on the “extreme violence” of irony, until all violence becomes figurative and its ethical implications are absorbed into the aesthetic.
For a critic so dedicated to the study of nineteenth-century fiction, Stewart is remarkably suspicious, even hostile towards realism. This accounts for his apparently anomalous inclusion, in a book otherwise concerned only with the Victorian novel, of a chapter on Poe’s tales, whose self-conscious phonetic play and foregrounding of graphic inscription displace narrative representation and anticipate the anti-realistic effects of high modernism: “all that is left is style itself” (71). Such self-enclosed subjectivity, where “the descriptive motive [is] forfeited entirely to the inscriptive” was the object of Lukács’s indictment of modernism; for Stewart it becomes exemplary (71). His narratographic readings reproduce Poe’s resistance to realistic recuperation from within mainstream Victorian novels, locating effects like those of modernism or even postmodernism produced by the friction of their prose.
The most important precedent for Stewart’s project, as he partly recognizes, is the virtuoso “slow-motion” reading of Roland Barthes’s S/Z (1970). Stewart rightly insists on the difference between Barthes’s disruptive fracturing of Balzac’s Sarrasine (1830) into “lexias” for ideological analysis and his own way of preventing skimming in order to savor small stylistic features. They are alike, however, in their anti-realism, which in Barthes is politically motivated by a refusal of bourgeois ideology but in Stewart is simply an aesthetic preference. They are also alike in their misleading air of seeming to offer methodological models. S/Z, with its elaborate apparatus of codes, appeared to demonstrate a systematic method that could be adopted by others; many failed attempts eventually showed that it was a brilliantly unrepeatable performance—unrepeatable by Barthes as well as his imitators. Stewart’s efforts to proselytize for the widespread adoption of narratography are also unlikely to bear fruit. Narrative Violence is an impressive feat of close reading, and one can hope that it will encourage renewed attention to the language of fiction, but those who try to follow its lead will have to devise their own methods.
Peter Garrett is Professor Emeritus of English and Criticism and Interpretive Theory at the University of Illinois at Urbana-Champaign. He is the author of Scene and Symbol from George Eliot to James Joyce (1969), The Victorian Multiplot Novel (1980), and Gothic Reflections (2003).