This dissertation uses a functional systemic approach to language to examine the construction of the respective first-person narrators of nine of the most popular, commercially successful and therefore influential African American slave narratives published between 1837 and 1862 (Roper, Grandy, Douglass, Brown, Bibb, Northup, Ball, Jacobs, Picquet). This corpus of more than 410,000 words was scanned for various linguistic features such as transitivity of verbs, nominalizations, and several syntactic features. The texts chosen differ as to their methods of production. Some of them were written by the first person narrators themselves, while others were either extensively edited, dictated to an amanuensis, or in some other way controlled. The dialectics of creation and representation through language results in the leading question in this study: how do the first-person slave narrators identify and create a personality for themselves through their texts? This dissertation thus focuses on the linguistic means by which the first-person slave narrator creates what is defined as a “discoursal self”, which helped the narrators to achieve one of their most important goals, namely, to be accepted as reliable. The dissertation consists of six chapters. Chapter 1 introduces a sociohistorical account of slavery, resistance against slavery, abolition, and the development of the slave narrative. The chapter shows that the African American slave narrative was not a completely new and original genre but an amalgamation of a variety of preexisting white and black literary forms. The second chapter is more theoretical and deals with discourse, power, and ideology in the slave narrative. Chapter 3 approaches the language of the slave narrative. The small corpus of scholarly texts that tackle language and style in this genre is reviewed. As an alternative to these approaches, critical discourse analysis (CDA) according to Norman Fairclough is suggested. It eliminates the a priori categorization of specific linguistic features as stylistically significant, because it is based on a functional view of language that perceives linguistic expression as choice on various levels. Every choice is considered meaningful and, according to its presence, absence, or clustering in a given co-text, potentially stylistic. Michael Halliday’s systemic functional grammar is introduced as the basis for the ensuing text analysis. Chapter 4 introduces the first quantitative observations about the density and distribution of the first-person singular pronoun in the narratives. This characteristic is then placed in relation to syntactic condensation in the forms of ellipsis, finiteness and nominalization, all of which are reviewed quantitatively. Finally, this chapter introduces the system of transitivity according to Halliday, Matthiessen, and others. It explains the distinction between the individual process types and provides a quantitative overview of the individual transitivity profiles within each narrative. Chapter 5 represents the main part of this dissertation. Each of the nine narratives is analyzed individually as to the presence of the I-pronoun in the text and the use and distribution of process types. In this way patterns of foregrounded or favored usages against absences of others emerge and contribute to the discoursal selves that the individual narrators present/construct of themselves. These preferred usages in general as well as in their local distributions are examined in detail. The quantitative observations supply the basis for further qualitative analyses derived from a large number of examples from the texts. Thus it is possible to show that each of the narratives is linguistically unique, which results in an individual construction of the respective I-narrator. The use of pronouns, process types and syntactic reconfigurations reveals how control over the activities as well as over the text is constructed, which can be directly related to issues of power. The Summary provides a synopsis of the previous quantitative and qualitative analyses and associates the quantitative results with characteristics of written and oral texts in general. Finally, it thus becomes possible to rank the nine narratives on a cline between predominantly oral and chiefly written characteristics.
|