University of Illinois Urbana-Champaign

Cycles of agreement: Romance clitics in diachrony

Maddox, Matthew Leroy

Content Files
MADDOX-DISSERTATION-2019.pdf

Permalink

https://hdl.handle.net/2142/104875

Description

Title
Cycles of agreement: Romance clitics in diachrony
Author(s)
Maddox, Matthew Leroy
Issue Date
2019-04-19
Director of Research (if dissertation) or Advisor (if thesis)
MacDonald, Jonathan E
Doctoral Committee Chair(s)
Hualde, José I.
Committee Member(s)
  • Yoon, James
  • Camargo, Martin
  • van Gelderen, Elly
Department of Study
Spanish and Portuguese
Discipline
Spanish
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Date of Ingest
2019-08-23T19:55:47Z
Keyword(s)
  • Clitics
  • diachronic syntax
  • reflexive pronoun
  • null subjects
  • null objects
Abstract
In this dissertation I investigate two linguistic cycles in Romance per van Gelderen’s (2011) framework: the Subject Agreement Cycle (SAC) and the Object Agreement Cycle (OAC). These grammaticalization cycles turn pronouns into agreement morphology on the verb. Both cycles are comprised of three stages. At stage (a) the pronoun is a full DP. At stage (b) the pronoun is reanalyzed as a D-head and at stage (c) it is reanalyzed as a T-head in the SAC or a v-head in the OAC. I extend the SAC to account for the grammaticalization of impersonal pronouns. I show that Modern French on is at stage (c) of the Impersonal Subject Cycle (ISC). Old Spanish (OldS) omne was on this cycle but it disappeared only to be replaced by Modern Spanish (ModS) uno. I propose that the reason for this disappearance was that impersonal subject pronouns will only be reanalyzed if personal subject pronouns are being reanalyzed first via the SAC. I also build upon van Gelderen (2011) by examining the OAC in Spanish in more detail. I show that in OldS, object clitics were full DPs and thus OldS was at stage (a) of the OAC. Based on diagnostics of coordination, modification, and movement, ModS object clitics are more deficient than OldS object clitics. Patterns of clitic doubling are evidence that standard ModS is at stage (b) of the OAC while Rioplatense Spanish is at stage (c). I adopt an analysis of accusative clitic doubling (ACD) based on Harizanov (2014) and Kramer (2014) whereby the object merges and moves to Spec,v (object shift) as a DP. I also show how object movement feeds the OAC. Object movement results in the object pronoun being in Spec,v, where it is associated with the v-head. Since ACD depends upon object shift to Spec,v, we expect languages that have developed ACD to have had object movement at an earlier period. This is the case for Romance. I show that the reflexive clitic se in ModS has been subjected to a type of OAC which I label the “Reflexive Object Cycle” (ROC). The ROC is a grammaticalization cycle that takes a reflexive object pronoun and turns it into a valency-marking morpheme, a Voice or v-head. I present evidence that in Latin and OldS, the reflexive pronoun was a full DP. It was later reanalyzed as a D-head and then a Voice head. This reanalysis is supported by diagnostics of interpolation, modification, coordination, and doubling. I demonstrate that null subjects and null objects relate to the stages of the SAC and the OAC. Null arguments are allowed in a language only if that language has reached stage (c) of the relevant cycle. I extend a D-feature and topic-identification type of analysis based on Holmberg (2005, 2010) and Holmberg et al (2009) to the licensing of null objects; i.e., null objects are licensed by a D-feature in v. I argue that this D-feature is only present on v in some varieties of Spanish because the clitic’s D-feature has been reanalyzed as a feature of v, which is a result of how the OAC works. I show how clitic left-dislocation and ACD are tied to the stages of the OAC, which accounts for their cross-linguistic distribution. I propose a typology of null object languages based on Holmberg’s (2005, 2010) typology of null subject languages. As for null subjects, I illustrate how the SAC and the ROC have interacted in the history of Spanish to give rise to passive se (Pass¬se). In order to develop passive se, two elements are need: null subjects and a reflexive Voice head. These elements are present due to the SAC and the ROC, respectively. This also accounts for the presence or absence of passive reflexive constructions crosslinguistically. Passse develops when a language has subject pro and se as a Voice head.
Graduation Semester
2019-05
Type of Resource
text
Permalink
http://hdl.handle.net/2142/104875
Copyright and License Information
Copyright 2019 Matthew Maddox

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Spanish, Italian and Portuguese