Refined vs. Middling Styles in the Lincoln Reminiscence: Comparing the Rhetoric of Formality and Familiarity

Abstract

This essay discusses the competing rhetorical styles of two volumes that appeared in the 1880s to remember Abraham Lincoln. One volume, edited by Alan Thorndike Rice, remembered Lincoln in a refined-official style. A second volume, by William Herndon and Jesse Weik, captured Lincoln in a middling-vernacular style. Using automatic coding and close reading, the authors show that Herndon-Weik’s middling-vernacular style put a focus on the “personal” Lincoln. Rice’s essayists, instead, featured an “official” Lincoln set apart from the everyday man. The authors argue that these contrasts were a contributing factor to the different critical reception they received.

Notes

1. The authors would like to thank Theresa Enos and RR peer reviewers Barry Brummett and Peter Cramer for their helpful guidance on the essay.

2. We draw on John Bodnar’s conception of “vernacular” memories as smaller-scale remembrances that emphasize personal “experience” and “feelings.” “Official” memories, conversely, are larger in scale and perpetuate the patriotic ideals of the institution and its leaders (13-14, 26, 44). As first-hand remembrances reliant on personal experience, reminiscences exemplify the characteristics of vernacular memory (Peterson 83). Yet, stylistically, reminiscences could assume the rhetorical trappings of official memory, especially when the rememberer boasted a position of institutional power and authority. We contend that the memories published by Rice’s authors appeared more “official” because they wrote from the perspective of White House insiders, using a refined style that was more formal than familiar. “‘Official’” memories, Kendall R. Phillips and G. Mitchell Reyes contend, “are largely tied to the broader narrative of the nation-state” (3).

3. Through his DICTION program, Roderick P. Hart pioneered automatic coding for rhetorical analysis, and his work exemplifies the benefits of automatic coding in the field of rhetoric. He notes that automatic coding is faster, more transparent, and more “dependable” than human coding (Hart, Childers, and Lind 17). Another advantage of automatic coding is that if the coding categories are mutually exclusive, the computer coding achieves one hundred percent coding reliability. This was the case for the automatic coding environment we used in the present study.

4. The “F” stands for the F test that is a ratio of between- and within-group variation across the Herndon-Weik and Rice volumes. The larger the F value, the more Herndon-Weik and Rice differ on a particular language feature. Every F value is associated with a corresponding p- or probability value. For the personal register, an F value of 5.44 is associated with a p value of 0.022. P values of .05 or smaller are conventionally understood to be statistically significant. All the other F and p values mentioned in this section should be read in exactly the same way.

5. Whether words/phrases can count in multiple categories depends on the specific detail of the computer program used. The pattern-matcher used in DocuScope is flexible and can match entries multiple ways but will typically make one “best” match based on length of the string of words.

Additional information

Notes on contributors

David S. Kaufer

David S. Kaufer is Professor of Rhetoric in the Department of English at Carnegie Mellon University. From 1994-2009, he was Head of the Department. His research interests are in the qualitative and quantitative analysis of rhetoric, writing and written information, and technologies for text analysis and text collaboration.

Shawn J. Parry-Giles

Shawn J. Parry-Giles is Professor of Rhetoric and Political Culture in the Department of Communication and the Director of the Center for Political Communication and Civic Leadership at the University of Maryland. Her research centers on the study of rhetoric and politics, with a focus on presidential studies.