Wydawnictwo KUL / The Publishing House of the Catholic University of Lublin

Permanent URI for this community

https://hdl.handle.net/20.500.12153/271
Wydawnictwo KUL powstało w marcu 2002 roku w miejsce Redakcji Wydawnictw KUL, która z Uczelnią związana była od 1957 roku. Kontynuując działalność wydawniczą RW KUL, publikuje dzieła naukowe: monografie, rozprawy doktorskie i habilitacyjne z zakresu teologii, filozofii, humanistyki, nauk społecznych i prawa, a także nauk matematyczno-przyrodniczych.

News

Kontakt

Wydawnictwo KUL
ul. Konstantynów 1H
20-708 Lublin
tel. 81 454 56 78
fax 81 454 56 93

Strona www: Wydawnictwo KUL

Browse

Browsing Wydawnictwo KUL / The Publishing House of the Catholic University of Lublin by Author "Chudak, Magdalena"

Now showing 1 - 1 of 1
  • Item
    Initial segments in Irish - variation in mutation patterns
    (Wydawnictwo KUL, 2022) Chudak, Magdalena
    The book investigates the alteration of word-initial segments in Irish. The phenomenon consists in a dialectal variation whereby the phonetic realisation of certain words differs, only or mainly, in the quality of the word-initial segment, e.g. cabáiste/gabáiste ‘cabbage’, beach/ meach ‘bee’, aiste/faiste ‘peculiarity’. The goal of this analysis is to contribute to the description of the process, identify its causes, and propose a formal account of the alteration in the word-based morphology framework. Lexical Relatedness Morphology (Bochner 1993) and Network Morphology (Bybee 1985) are two implementations of the word-based morphology used in this work. The key tenet of both theories is that all the words a speaker encounters are stored in the lexicon in their full form. The words are organised into sets, called “word-schemas”, according to their morphological (derivational or inflectional) relatedness, e.g. act/actor, describe/description/descriptive. Furthermore, following Jackendoff’s (1975) proposal, the principle of “full-entry” is promoted, i.e. each lexical entry is fully specified with its phonetic, morphosyntactic, and semantic properties. Morphological patterns, similar to Jackendoff’s (1975) redundancy rules, emerge from sets of morphological relations on a different level of abstraction, and are represented as abstract word-schemas. The lexical connections as well as lexical entries are characterized by different degrees of lexical strength, which result from the frequency of usage of a given word or morphological pattern. As Bybee (1985) postulates, the more frequent a word or a pattern, the higher its lexical strength. The lexical strength, in turn, determines the ease of lexical retrieval, i.e. the more frequent forms are retrieved faster than the rare ones. The most productive morphological patterns, on the other hand, are those with the strong lexical connections. The central claim made in this work is that the word-initial reanalysis results from the confusion caused by mutation patterns. Irish initial mutations are sets of consonant replacements triggered by morphosyntactic contexts, e.g. after a feminine definite article or the 1st person plural possessive adjective. There are two types of mutations in Irish, i.e. eclipsis and lenition, and their effects are defined by grammar. For instance, [p] in a leniting context is replaced with [f], and in the eclipsing one with [b]. The problem inherent in the mutation system is that it allows for certain consonants to occur both as effects of mutations and as radical (non-mutated) consonants. For instance, the above-mentioned [b] is the initial radical segment in the word bean ‘woman’, but it is an effect of eclipsis in ar bpeann ‘our pen’. In the model of the word-based lexicon, mutations are, on the one hand, embedded directly in the relations between three interrelated forms, i.e. radical, eclipsed and lenited. On the other hand, mutations are patterns which emerge from these sets of forms, referred to as mutation-schemas, and represented as sets of related initial segments specified for the mutation grade (radical, lenited, eclipsed). The representation of mutation as connections between consonants ensures that speakers use them productively with borrowings or neologisms. Also, the knowledge of mutation patterns as relations, and not input-output rules, enables them to retrieve a radical form on the basis of the mutated one. The explanation of the alteration of word-initial segments hinges on two modes of the lexical representation of mutations pinpointed above, i.e. word-level sets of related word-forms, and as abstract sets of connected related. On the word-level, as the word-forms vary in their token frequency, a very frequent mutated form may be retrieved more readily than a rarely used radical, and reanalysed as a radical word. For instance, the loan-word fréamh→préamh ‘root’ is a feminine noun, thus it is always lenited after the definite article giving an phrae [ǝn fre:]. If this form is prevalent, in some dialects the initial [f] is reinterpreted as a radical segment and the “new” radical word frae is reassigned to a different mutation schema. On the other hand, a very rare radical word may be misinterpreted as an effect of mutation. For instance, the opposite alteration to the previously described one is possible where fréamh ‘root’ is reanalysed as a lenited variant of a non-existent préamh. Both cases of reanalysis are possible because of the overlaps in the abstract mutation patterns. In the first example, speakers’ reference to the abstract mutation schemas lead to the confusion of the initial mutated [f] (related to the radical [p]) with a radical [f]. In the second case, the mutation schemas serve to ‘restore’ the radical préamh from fréamh ‘root’. Chapter 1 is a brief review of the Irish sound system, with a focus on the phonotactic constraints concerning the beginning of the word in Irish. Next, a synchronic and diachronic account of consonant mutations is provided, in terms of mutation effects and contexts of occurrence, in order to set the scene for the analysis of other types of word-initial segment replacements. The core of the chapter is the introduction of the theoretical model in which the analysis is couched, i.e. word-based morphology. The fundamental aspects of the lexical representation of mutations are presented, as well as their relevance in the context of word-initial alteration. Chapter 2 is divided into two parts. The first one clarifies the terminology involved in the description of the reanalysis of initial segments. The term “alteration” is contrasted with the term traditionally used in this context, i.e. “interchange”. It is shown that the alteration is a replacement process whereby one radical segment is substituted with another initial segment, e.g. miodóg→biodóg ‘dagger’. In consequence, the word is reassigned to a different mutation pattern and mutated accordingly. The term “interchange”, on the other hand, does not capture the derivational aspect of the word-initial change. Rather, it is an observation about a synchronic variation in the realisation of a given word across dialects, e.g. méaróg/béaróg ‘neat little girl’. The second part of the chapter is dedicated to the methodology of data collection, organisation and analysis. The sources of the dialectal variation in initial segments are enumerated and briefly discussed. Chapter 3 seeks to present a unified account of these alteration patterns which are based on consonant mutations. The analysis is couched in the word-based morphology framework. A typology of the word-initial changes is offered, the basis of classification being the type of the reanalysis process involved in postulating the secondary radical. Three types of mutation-based alterations are distinguished, i.e. radicalisation (when a mutated segment is assumed to be radical, e.g. [p]→[b]), deradicalisation (when the radical consonant is interpreted as a mutated reflex of another radical, e.g. [d]→[t]), and misradicalisation (when the mutated reflex is associated with a radical different from its original one, e.g. [m]→[b]). For each type of alteration a general mechanism is provided, as well as a discussion of the relevant dialectal instances. Additionally, two in-depth case studies are included, i.e. the alteration in the noun beach→meach ‘bee’ and in the dependent form of the verb téigh ‘to go’, i.e. deachaigh/teachaigh ‘going’ Chapter 4 concerns the reanalysis phenomena which are not rooted in mutation patterns, but have the same effect, i.e. alteration of the word-initial segment. Three such processes are discussed. The first one is the confusion of initial palatalised coronals [t′], [d′], with palatalised velars, i.e. [k′] and [g′] respectively. It is argued that the alteration results from the phonetic affinity of these two classes of sounds in some Irish dialects. Another reanalysis process is the semantically motivated prefixation of the consonants s- and p-. They are shown to have a phonaesthemic function of increasing the expressiveness of derogatory terms, e.g. leib/pleib ‘fool’, truán/struán ‘wretch’. The last group of alterations are those resulting from the wrong division of the definite article, i.e. an and an t-, before vowel-initial nouns. This development, called metanalysis, is found in the following by-forms: aos/taos ‘people, folk’ or uchtóg/nuchtóg ‘small heap’. As one of the main objectives of the work was to compile the list of the attested by-forms, the last part of the book presents the corpus of these in the form an appended table.