Abstract of my Ph.D. dissertation in linguistics at Korea University, Seoul, Korea

Abstact

 

On the morphemizations of Korean honorific endings

 

 

Name : Kim Hyun Ju

Department : Korean Language and Literature

Thesis Adviser : Chung Kwang et al.

 

 

The aim of this dissertation is to identify how Middle-Korean honorific endings develop so to form new morphemes. Middle Korean honorific grammatical categories consist of three different kinds: Honorifics for the subject include {-ɨshi-}, {-as*jə}, {-ɨsjosjə}, those for the listener include {-ɨi-}, {-as*jə}, {-ɨsjosjə}, and that for the object includes {-sʌp-}. Of these, {-as*jə} and {-ɨsjosjə} are a member of two different categories at the same time ― the reason why each of them can be a member of two different categories is that they function as imperatives where the subject is always the listener.

In this dissertation, usage-based methods are adopted for tracking and evaluating the changes honorific endings pass through. In the usage-based viewpoint, concepts or presuppositions such as connections of words in the mental lexicon, the frequency effects, fusion before coalescence, and morphemes as prototypical categories are natural and harmonious. Addition to these, a heuristic concept, named morphemization is invented to cover several similar phenomena, e.g,, grammaticalizations, lexicalizations, analogies, reanalyses, and so on.

Miscellaneous and detailed investigations are taken in chapter 3 and 4. Chapter 3 is to investigate how honorific endings become lexical morphemes, and Chapter 4 is to investigate how honorific endings become grammatical morphemes. Through these two chapters, the diachronic processes on how new honorific morphemes are constructed are explored. Besides, usage-based methods prove themselves useful through these two chapters.

In chapter 5, some general linguistic inquiries arisen from previous chapters are discussed. First, it seems that morphemes can be nicely seen as prototypical categories. By employing this, allomorphs' reorganizations, allomorphs' family-resemblance effects, and changes of allomorphs' distributions can be naturally explained, cognitively. Second, it can be suggested that every morphemization has restrictions on semantic relations. Adjacent elements of similar senses can easily fused, but those of uncompatible semantically nor hardly construed as one semantic unit are not easily fused. Finally, It is suggested that two different types of patterns can emerge from connections made between words in the lexicon ― One is to attach or detach, and the other is to alternate. In inflectional languages, alternating patterns are predominant, and in agglutinative languages, attaching or detaching patterns are predominant. Of course, attaching or detaching is also a kind of alternating in broad sense.

 

* If you find awkward expressions, let me know.