Chair: Helen Elizabeth Ross
Insular Celtic dialects provide relatively lean pickings for the old inherited Indo-European words
(i) *suyo- / *sūnu- ‘son, male descendant’ and
(ii) *dhughHther ‘daughter, female descendant’ (forms which are alive and well in English son and daughter).
This paper will examine the alternative terms used for 'son' and 'daughter' in the living dialects of Goedelic and Brittonic. In addition to the lexical items in dictionaries and literary texts, our coverage for the lexemes ‘son’ and ‘daughter’ in Insular Celtic is, of course, greatly enhanced thanks to the minute geographical coverage available to us in the various linguistic atlases conducted in Wales, Brittany, Ireland, Scotland and the Isle of Man.
The wealth of material for Insular Celtic will be contrasted to the more meagre remnants available to us for Insular Celtic. It will, nevertheless, be possible to come up with an interesting set of data for the ‘son’ and ‘daughter’ in Continental Celtic sources.
Irish has a multitude of idiomatic constructions whose meanings cannot be derived with certainty from their components, but depend on their combination. I present a qualitative analysis of the construction téigh i X lit. ‘go in X’ where X is a predicate noun.
In general, the verb téigh ‘go’ is used with prepositions of direction: go ‘to’ and chuig/chun ‘to’. However, in certain constructions it can also be used with the preposition i ‘in’ which does not indicate direction, but rather location. The meaning of ‘he went into a room’ would be rendered as chuaigh sé isteach i seomra lit. ‘went he insidewards in room’, so that the motion meaning is expressed by the adverb isteach ‘insidewards’. The combination of a motion verb téigh ‘go’ with a locational preposition i ‘in’ alone must signal non-compositional, metaphoric meaning:
DEF footstep.PL PROG go.VN in neglect
‘[One hears] the footsteps fading’
I aim to distinguish a number of semantic types of nouns typically occurring in this construction. Then, it will be discussed whether these semantic types are coherent with the transitive/intransitive status of the construction and, in the case of transitive constructions, the way they govern an object. Finally, any other verbs which could possibly combine with the prepositional phrase will be detected in order to ascertain whether different types of this construction demonstrate an equal degree of boundedness or not.
Thus, I will show the types of the téigh i ‘go in’ constructions and how they differ semantically, syntactically and in the degree of cohesion.
Agglutinating languages are generally analysed into morphemes. For example, a verbal complex in Japanese tabesaseraremasendesita ‘(I) was not made to eat’ is analysed like tabe-sase-rare-mas-en=des-ita (eat-CAUSATIVE-PASSIVE-POLITE-NEGATIVE=COPULA-PAST). In contrast, inflectional languages, including Irish, are not generally analysed into morphemes and grammatical categories are indicated with a dot after the word concerned. For example, in Irish, a nominal phrase an bháid ‘of the boat’ is glossed like DEFINITE.MASCULINE.GENITIVE boat.MASCULINE.SINGULAR.GENITIVE and a conjugated verb bhrisfinnse ‘I myself would break’ is glossed like break.CONDITIONAL.1SG.CONTRASTIVE. However, I do believe that inflectional languages can be analysed into morphemes to some degree.
The present paper attempts to analyse Irish words as detailed as possible and proposes analyses for several grammatical instances. This analysis is based on the orthography but special symbols indicating some sorts of phonological phenomena, such as L for the lenition, i for the slendarisation (i.e., palatalisation) of a stem, etc. When this analysis is adopted, the examples above can be analysed like anL bád-i (DEFINITE.MASCULINE.GENITIVE boat.MASCULINE-SINGULAR.GENITIVE) and L bris-f-inn=se (PASTMARKER break-FUTURESTEM-CONDITIONAL.1SG=CONSTRASTIVE). In this case, inflections are indicated by a hyphen and other elements are indicated by a double hyphen. The morpheme -f-, which appears in the future and in the conditional, can be independently analysed, but the ending -inn cannot be analysed any more.