The two-fold goal of this dissertation is to conduct a thorough survey of diachronic changes to person marking reference within paradigms (here called Person Marking Referent Shifts or PMRS) across a wide set of languages and then to analyze patterns in the trajectory of those shifts through a formal separation between the syntactic and semantic representation of the ɸ-features underlying pronominal exponence and the actual contexts of use that are available to those forms. As such the body of this disseration is divided into four parts. The opening section introduces the scope and bounds of the project and reviews the literature on person marking and PMRS that led to the current project. It addresses the range of discovered PMRS types and claims of directionality found in Heine & Song (2011), Song & Heine (2016), and Helmbrecht (2015), specifically that person (3 > 2 > 1) and number (PL > SG) hierarchies may both trigger and constrain these shifts. In order to investigate those claims more comprehensively, Chapter 2 broadens the surveys of PMRS seen in those works to a much larger variety of languages from genetically and geographically language families, exposing a number of novel findings. First, it is demonstrated that PMRS is relatively common in the sample compared to estimates based only on more commonly studies families. Further, the survey greatly expands on the types of PMRS as defined by the difference between source and goal referent sets. Finally, the results are found to be inconsistent with previous proposals of strong influences towards uni-directionality in these shifts based on referent accessibility since the data include widespread bidirectionality not present in previous surveys.
However, the case is made that not all shifts represent the same phenomenon. They can in fact be divided into two groups: Independent Shifts and Dependent Shifts. While Dependent Shifts appear to involve restriction of a markers reference to a subset of its original contexts due to the introduction of a competing marker, Independent Shifts involve an extension of the marker to new contexts of use without the concomitant addition of any competing markers to the person marking paradigm. Based on historically recorded trajectories for some shifts and synchronic facts about the person markers’ distribution in others, many of the shifts in the survey are able to be immediately sorted into one of these two categories. The next chapter goes into depth on the semantic and syntactic structure of the phi-features underlying reference to PERSON, NUMBER, and GENDER. Appealing to the usage of person markers, I bolster the claim made in other works that these features often utilize privative, rather than polar, divisions. In these cases, one member of the contrasting dichotomy has a semantically contentful feature defining its contexts of use, while the other (bare) member is confined from use in some of the contexts with which it is semantically compatible due to pragmatic pressure from the existence in the paradigm of the other more semantically specific marker that it competes with. This is expanded on for features such as NUMBER (contentful PL vs. bare SG) and clusivity, which often involves a contrast between one contentful clusive marker, either exclusive or inclusive, and a more general marker.
Privative contrasts capture the details of the Independent/Dependent contrast seen in PMRS very well. That is, Dependent Shifts are ones in which a more feature-rich marker is innovated and enters a paradigm, pragmatically forcing a feature-poor more-general, with which it competes for the same contexts, into a subset of the contexts that are semantically available to it. On the other hand, Independent Shifts involve actual change, or reanalysis, of the featural content of a pronoun. Due to a majority of these cases following a pathway from more to less features (extending their range of compatible contexts of use), a proposal for directionality in Independent Shifts is entertained positing that the shifts always involve a loss of featural content similar to bleaching. Independent Shifts appear to be triggered by the socio-linguistically strategic use of the ‘strong’ feature-rich pronoun in contexts where the hearer is socially expected to take their word for its acceptability. As seen in other cases of reanalysis (see Eckardt 2012 and Deo 2015), some hearers question this accomodation and posit instead that the speaker is using the form to mean something more immediately accessible to both speech act participants in the context. The directionality of these shifts falls out from the assymetric entailment relationship between the feature-rich ‘strong’ form and the feature-poor ‘weak’ forms. The pair form a Hornian < Strong,Weak > dyad, in which S → W but W does not entail S and the use of W implicates that S could not have been used in that context. For example, in NUMBER, the 1PL entails 1SG but not vice versa and the use of the 1SG marker has an implication that the 1PL could not have been used in that context. Thus if a speaker uses S in a context the hearer associates with the weak meaning, there is impetus to reanalyze but if one uses a W form in a context associated with the strong meaning, the utterance is still technically true and would not produce the triggers necessary for reanalysis. This proposal would simplistically separate true reanalysis involving actual feature loss that lead to context extension from cases of pragmatic restriction that lead to context reduction.
However, counterexamples in the data suggest that this proposal is incomplete. To investigate this more fully, the final chapter of the dissertation looks at some of the more peripheral examples of PMRS both within the survey sample and outside of those language families. Specifically, the paper looks into shifts in GENDER, shifts from 3rd person to SAP markers, and those involving feature ‘hardening’ of clusive contrasts and some number contrasts, which show feature ‘gain’ instead of loss. Together these make a strong case that reanalysis directionality in PMRS should not be defined in terms of what is likely, but rather what directions are unlikely to occur given the triggers that are necessary. In other words instead of favoring S > W directionality, the processes involved simply do not favor W > S. When no such <S,W> relationship is present between a potential shift’s source and goal feature sets, directionality does not appear.
|Commitee:||Pye, Clifton, Gluckman, John, Duncan, Phil, Vyatkina, Nina|
|School:||University of Kansas|
|School Location:||United States -- Kansas|
|Source:||DAI-A 82/9(E), Dissertation Abstracts International|
|Subjects:||Linguistics, Language arts|
|Keywords:||Historical linguistics, Person Marking Reference Shifts, Pronoun, Typology, Pronominal exponence, Dependent shifts|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be