Uses and Variations of Word-initial Pretonic e- in Alguaire

Uses and Variations of Word-initial Pretonic ein Alguaire Josefina Carrera Universitat de Lleida Western and eastern Catalan are two widely spoken dialects that are gradually becoming alike. A phonetic feature they have both had for years has been the similar pronunciation of word-initial pretonic e- (eastern [a] and western [a]) Nevertheless, in some zones of the western dialect a new vowel is taking its place in the pretonic position: [e]. We have attempted to describe this ongoing linguistic change in a western speech community: Alguaire, situated about 15 km. from Lleida, the county Seat. It begins with a two-tier analysis of 36 informants: on a productive level (a statistical study of phonetic realizations) and on a perceptive level (evaluation of the acceptance and tolerance of word-initia1 pretonic sounds). We intend to show the gradual substitution of the autochtonous variant [a] for the written and normative vowel [e], and to determine, at the same time, the linguistic, social and psychological factors that favour this substitution. Nowadays, it seems that languages are going through a standardization process which makes them converge. This is helped by the mass media. The phenomenon we are studying located in the north-western dialect, is an exception to this tendency and it marks a new and distinctive separation from other varieties of eastern dialects. Following Weinreich's, Labov's and Herzog's (1968) theory of the stages of linguistic change, our intention is to analyze the different parts of this change taking 36 informants divided into three generational groups (from 65 to 95, from 25 to 45 and from 8 to 10) which we have complemented with two different variables: sex and socio-cultural status. At the same time, the orientation of this study is twofold. First, the article seeks to present the frequency of both word-initial pretonics in 100 Catalan Working Papers in Linguistics (CatWPL) 3.2 (1994): 1-15 Universitat Autbnorna de Barcelona words starting with ewith a statistical study of phonetic realizations. Second, it analyzes the acceptance and tolerance of the two vowels by the informants after having listened to sentences and words which contain one or two of the word-initial pretonic vowels. 1. The speech community Alguaire is a village located in the western part of Catalunya, 15 km. from Lleida, the county seat, and about a two-hour drive from Barcelona, the largest nucleus of population in Catalonia, a centre of cultural production and administration and the urban nucleus most likely to infiuence standard Catalan. Alguaire's economy is based on farming and levestock. However, in the XIX century there was a textile factory that boosted the economy and also the population of the village. Unfortunately, in 1979 this big source of income disappeared. At present, Alguaire has a population of 2,800, some (14.3%) are Spanish-speaking immigrants' who carne to Alguaire to work in the textile factory and now have other jobs. Nevertheless, the great majority of Alguaire's inhabitants are Catalan speakers. Among other differences between the western and eastern dialects of Catalan we point out the unstressed vocalic system: eastern Catalan has a more reduced unstressed system liaul than the western one /iecaou/ which therefore means western Catalan has more variation in the word initial pretonic esound: e - in eastern Catalan-it basically is [a] in western Catalan-it can be [a], [e] We are talking about people who came mostly from Andalucia. They are a 63.8% of the immigrants. 2. Data analysis The [a] emission of initial etendency is a north-western dialectal feature that has been noted by many linguists including Gili i Gaya (1931), Veny (1982), Badia (1984), Palmada (1991), Recasens (1991), Solans (1991). This phenomenon had different explanations: a phonosintactic link, (that is, the resolution [a] of the contraction of the vowel of the article and the ein initial word position la + escala [ias'kafc] 'stairs'); a vocalic assimilation to the nuclear syllable (eldstics 'braces'), the vocalic quality of the vowel [a] in the north-western variety, etc. Apart from that, we must say that some researches (Colomina (1985), Recasens (1991) Solans (1991)) have also pointed out the detection of a different sound in the word-initial pretonic place: [e] instead of [a], which is the one we have explored. We based the study on a few points which we can summarize as follows: 1) We selected a corpus of words adapted to the characteristics of the analysis, where we put aside the evaluation of spontaneous continuous speech because of the problems involved in finding the word-initial pretonic vowel in an initial position and without elisions or vocalic alterations due to preceding sounds. The corpus of analysis has been formed by one hundred words taken from a basic vocabulary which have different consonant and vocalic adjacent contexts. In order to find connections between the words' origin and the present pronunciation of the word-initial pretonic sounds, we have classified these words according to their etimology. 2) Another important aspect was the choice of generational groups. They consist of children from 8 to 10 years old (III),2 adults between 25 to 45 (11) and also from 65 to 95 years old (I). Despite the fact that informants do not follow correlated areas of population, we intended to make a global analysis of three different stages of the speech community. In order to define the boundaries of all the samples we concentrated on native Alguaire speakers3 who have at least - one parent from Alguaire. We have also taken into account the sex and socio-cultural status, which has been differently analyzed in each subgroup taking into account socio-economic and cultural background (that is, level of schooling, hobbies and personal activities). Children have been classified with a more subjective criterion which is the analysis of cultural activities of each farnily. 3) Each informant was interviewed twice. First, to analyze the production results, they had to answer a questionnaire of closed questions that were structured in a semantic order,4 and with 50 additional words without word-initial pretonic ein order to hide the object of the interview. We have also taken care to avoid answers with possible vowel changes because of vocalic contact, such as in una ermita [uncr'mitc] ('an hermitage'), where the word-initial pretonic is elided. That is, the preceding context should be either consonant or pause. The Roman numerais refer to the different age groups: I11 (8 to 10). I1 (25 to 45) and I (65 to 95). By this we mean people who speak the north-western Cataian variety with the distinctive traits of the Alguaire area 4That means words were classified in different semantic fields to make easier the informants answers After that, a second test was carried out. It was based on the acoustic perception of word-initial pretonic vowels in words with different frequencies of the word-initial pretonics. This second test had two parts: one analyzed the acceptance degree (Pons 1992) of [a] (if the informants considered a sound was usual or unusual, if they used it, etc) after the perception of the wordinitial pretonic [a] in isolated words as well as in sentences recorded from a male Alguaire speaker. The other one had the purpose of evaluating and contrasting the perception of [e] and [a], recorded twice in identical isolated words and sentences for the same speaker. This perception interview had different aims that were separately evaluated by each age group. Briefly stated, are: a) acceptance of the perception of [a] in isolated words and sentences; b) individual relations between the emission of their own word-initial pretonic vowel with the connotation of the perceived one in the same words; c) possible differentiation of the wordinitial pretonics [e], [a] having listened to both of them in the same words and sentences; d) particular selection and evaluation of each vowel.5 3. Results As the study has two parts, the explanation of the results will be given in two sections: a) data of production analysis, b) resul ts of perception test. The data regarding the production interview are divided into different items. The first one presents the factors that have altered the appearance of one of the two word-initial pretonic vowels and have favoured different phonetic possibilities which show the high degree of variation in which the initial word-initial pretonic is submited. They are: a) consonanta1 prothesis as in [den'tra], len'tral entrar ('come in'), b) vocalic and syllabic aphaeresis, ['k+ipsi], Ie'klipsil eclipsi ('eclipse') or [boralBo], [ezporalBo] esborrador ('board wiper'), C) [a] or [e] inflexion to [i], as in [is'tiw], les'tiwl estiu, ('summer'), d) change of eto [o] in The evaluation was taken from the following qnestions: 'which is the pretonic vowel you use?', 'which one do you think is the proper one?', 'Can you tell me what you think about the other one?'. [o'ruyc], [e'ruycl eruga ('caterpillar'), or eto [c] in [ckstrate'restre], Iekstrate'restrel extraterrestre ('extraterrestrial'). Apart from that, the apparition of word-initial pretonic [a] is favoured by the phonetic characteristics of adjacent sounds as [m], [n], and, to a lesser degree, [@I, [sk], [sp], [st]. [a] also depends on the quality of the stressed vowel of each word ([a] as a stressed syllable favours the word-initial pretonic vowel opening as in escala 'stairs'). Another important factor is the frequency of usage of words, which is directy correlated with the use of [a], as in the word esmorzar ('breakfast') which is very frequently used. After that, we present a pattern of variable and invariable word-initial pretonic vowel which points to the fact that the process of substitution of the western variant [a] for the standard [e] is not uniform but gradual and arbitrary. That is, it is not the sarne for each age group and neither for each word: Invariable words. We cannot find many. The ones that always maintain [e] word-initial pretonic are: eclipsi ('eclipse'), edat ('age'), editor ('editor'), etarra ('member of ETA'). Others that keep [a] are: efeminat ('effeminate'), eixuta ('dried'). Variable words. They can change in many directions depending on the age groups as I, I1 and 111: 3.1. Substitution of the word-initial pretonic [a ] in 116 Edifici ('building') I-[a] 11 [e] Educat ('well-mamered') I-[a] 11 [e] Electricista ('electrician') I-[0] I1 [e] We consider that the vowel change is outstanding when more than half of the informants pronounce [e] pretonic. Epia2mia ('epidemic') I-[0] I1 [e] Evangelis ('Gospel') I-[a] II[e] Exagerant ('exaggerated') I-[a] I1 [e] Exigent ('demanding') I- [a] I1 [e] Explosió ('explosion') I-[a] I1 [e] 3.2. Gradual reduction of the [a] as a word-initial pretonic: I [a], II [e], [a]7, III [e ] Elastics ('braces') I-[a] II-[e], [a] III-[e] Equivocar-se ('to mistake') I-[a] II- [e], [a] III-[e] Eri~ó ('hedgehog') I-[a] II-[e], [a] III-[e] Esclops ('clog') I-[a] II-[e], [a] III-[e] Escola ('school') I-[a] II-[e], [a] III-[e] Esquirol ('squirrel') I-[a] II-[e], [a] III-[e] Estrelles ('stars') I-[a] II-[e], [a] III-[e] Examen ('exam') I-[a] II-[e], [a] III-[e] Excursió ('trip') I-[a] II-[e], [a] III-[e] Furthermore, age is crucial for the substitution of the word-initial pretonic [a], and we can say that speakers with ages ranging from 8 to 10 show a lower rate of use of [a] than the second generation, and this use of [a] is greatest in the third group. Socio-cultural background is a decisive factor for the replacement of [a] to [e], because, in general, the most well educated people tend to pronounce [e] instead of [a]. The sex variable reveals that women use [e] more frequently with an overall difference of 15% (for groups I11 and 11). Women's tendency toward linguistic innovation is commonplace in sociolinguistic studies, and is replicated here.8 Despite this, we have to be aware that the group of older women shows the most conservative Here the emission of both pretonics takes a similar percentage: about 50% of the speakers each one. * Labov (1990). Alturo and Turell (1990). Montoya (1992) pronunciations. Sociohistorical reasons can account for this difference: they have grown up in a restrictive social world: the home. At the same time, men had more varied lifestyles because of their jobs, the Civil War, the military service, etc. (See the graphic.) Graph. Age and Sex (ela) Women I11 Men I11 Women I1 Men I1 Women I Men I - others The perception part is based on the study by Lídia Ponsg where she points out that in a substitution process, the existence of two controls1° in the linguistic mechanisms of the speakers shows an imbalance in the evolution and this leads people to believe that the phonetic changes in a language do not entail a conscious presence in the passive linguistic knowledge of the speaker. We can summarize the most relevant observations of this section in the following points: 1) The presence of word-initial pretonic [a] as strange or unusual in isolated words and in sentences appears to be inversely correlated with the degree of [a] productions in children and older people. Informants between 25 and 45, nevertheless, do not follow the pattern and are 9 Pons. L (1992) lodització i apitxament al Vallis. Interpretació sociolingüística i psicolingüística dels canvis fonitics, IEC, Barcelona. l0 he two controls are the linguistic production (active) and the linguistic perception (passive). more aware of the presence of this vowel. (The following graphic shows the percentage of [a] perceived as an unusual sound in each age group). Degree of consciousness of [a] 80 60 % of [a] 40 2 o o l?s molt educat. [aBulkat] 'He is very well-mannered.' Tens uns esclops? [as'kt~ps] 'Do you have a pair of clogs?' (a) gs un exagerat, va dir que havien guanyat i van empatar. [ag3a3e1rat] 'He exagerates, he said that they had won and they had drawn.' (b) 6s un exagerat, va dir que havien guanyat i van empatar. [ampa'ta] 'He exagerates, he said that they had won and they had drawn.' Ensenyen. [an'sej~en] They teach.' excursi6 [askursi'o] 'outing.' esponja [as'poj~~] 'sponge.' equi pe [a' kip] 'team.'