5.3 Change in real time

5.3.1 Reviewing the past

One efficient way of overcoming the logistical difficulties as well as the methodological issues, in studies of change in apparent time, is to search for studies previously carried out on the speech community, to use the results obtained for comparison. In other words, to use the past to explain the present.

This research approach, which has been developed to good effect by certain scholars working on linguistic change, has the disadvantage that the data are often too fragmentary and not always very good quality, but also has the advantage that the evidence is objective, in the sense that the data are not from a given type of study or a specific research project. A further methodological problem here involves the typical tension between qualification and quantification, since the observations that a variant occurred "sometimes" or "frequently" in the past cannot be compared with the quantified frequencies of more recent studies. What is more, the conservatism of the traditional dialectologists is notorious; it is well known that in general they limited their recording of forms to variants that had already been dated to previous periods of the history of the language.

5.3.2 Repeating the past and returning to the scene

In seeking to overcome the methodological problems discussed above, a new methodology was eventually adopted which involved repeating or replicating observations made in the past, by returning to the scene where the language variation and suspected language change was originally observed.

This new focus came into being thanks to the development and implementation of two types of study:

a) Replication or trend studies, where the study is carried out with the same population, and the same data collection methods are used, plus the same techniques of analysis, but x years later (usually 10 ~ 15 years later).

b) Studies based on the same original sample (panel studies). These involve the seeking out of the same individuals, with same monitoring of the changes in linguistic behaviour using identical instruments (questionnaires, etc) as before in the study on apparent time. I propose the term "sample study" to refer to this type of study of linguistic change in real time.

5.3.2.1 Replica studies

Trend or replication studies are the simplest way of returning to the past, but they raise a substantial number of methodological problems. In the first place, if a large community is involved, it may be difficult or impossible to include any of the individuals who had taken part in the original sample. Furthermore, the community would necessarily have to be one that remained demographically stable, otherwise the changes could, and probably would, be externally motivated. This does not mean that externally motivated variation and change are unimportant but in this case one would be hard put to say whether they were produced as a result of internal linguistic factors. (Bailey and Maynor 1987).

5.3.2.2 "Sampling studies"

This type of study seeks to go back to use the original sample of informants or consultants, and thus entails the locating of the same individuals who had taken part in the study of change in apparent time. These individuals are then given the same questionnaire, in the same sociolinguistic interview with the same formal tests as in the original research design. Instances of this type of study include a) the sample study carried out in Montreal in 1984 by Thibault and Vincent (1990) based on an original study in apparent time by Sankoff and Cedergren, in Sankoff and Sankoff (1973); b) the analysis of a single individual over a period of time carried out by Brink and Lund (1975) within the framework of research on the dialect of Copenhagen, and lastly, the study by (1987) on the Besaran dialect of Yiddish, which contrasts with standard Yiddish, based on the recordings of a single speaker, the folksinger Sara Gorby, over several decades of her life.

Any research context which sets out to look at the relationship between linguistic change in real and apparent time would need to be based on two sets of principles which emerged in the studies of linguistic variation and change carried out over the course of the three decades during which the Change in Paradigm was developing. One of these sets of principles came out of the research that looked at differences and similarities between variation in the individual, and variation in the group, and which sought to confirm the internal and external factors that could explain the uniform distribution of variation in a given community (Guy 1980). The other set of principles had to do with notions of change from above and change from below, already discussed elsewhere (Turell 1995a).

There are a number of published studies which consider the relationship between linguistic change in apparent time and linguistic change in real time. It will not be possible to mention all of them, but I should like to mention one or two of the more significant of those that deal with the Spanish and English languages.

Possibly the most significant study on linguistic change in Spanish is the replication or trend study by Cedergren (1969-1982) in real time in Panama dealing with the substitution of affricative /c/ of Spanish by the fricative /š/ as in words like muchacha (girl) muchos (many) etc. The objective of this study was to show, using newly available real-time evidence, that there actually was a change in progress.

There have been a number of studies of this kind involving English, but perhaps the most significant from the point of view of the relationship between apparent time and real time is the study by Payne (1976, 1980) on the acquisition of the Philadelphia dialect. More specifically, Payne studied the phonological changes (for example, the splitting of the English vowel a into two variants: one tense, the other lax) which were occurring in the dialect which was being acquired by the children of families that were originally from outside the Philadelphia area.

Another interesting study from the methodological point of view is the replication or trend study by Trudgill (1988) of his original study of linguistic variation and change carried out in the English city of Norwich in 1968 (Trudgill 1974). This took on a real-time perspective by virtue of the fact that 17 speakers aged between 10 and 25 were added who could then be contrasted with the group of adolescents in the original 1968 sample. It is true that this only added the comparison of the same age group duplicated by two observations at different times, but it was nonetheless a very powerful way of demonstrating the efficacy of a given methodology.

In the field of research into Catalan, all studies of variation and change carried out to date take a methodological approach working with apparent time. See Alturo and Turell (1990) and Alturo (1995) for the Catalan nord-occidental dialect of the Ribagorça, Pradilla (1995) for Valencian nord-occidental dialect, Plaza (1995) for the Catalan of the Conca de Barberà (central Catalonia) as well as the study of linguistic stratification of Petrer (Valencia) carried out by Gimeno and Montoya (1989). And there are many others which have been carried out but which have not yet been published in scientific journals.

As I see it, these communities constitute linguistic laboratories, and are very relevant and suited to the implementation of "sample" type of study in real time. Such studies can supply data that will enable us to form a clearer picture of the current state of Catalan in terms of its internal structure, the internal and external factors that constrain patterns of language change and variation, what the starting point is for such changes, who are leading the changes, and the route taken. Information of this sort can also contribute to developing a theory of language variation and change, to which we variationists are committed — and to a general theory of language, which we do not yet have, but which we hope to be able to work towards on both theoretical and methodological levels.

8. Bibliography

ALTURO, N. "La variació d'haver auxiliar al Català nord-occidental". In M.T. Turell (ed.), La sociolingüística de la variació (1995), Barcelona: PPU, p 221-255.

ALTURO, N and M.T. TURELL, "Linguistic change in El Pont de Suert: the study of variation of //. Language Variation and Change. [Cambridge] (1990), no. 2, p. 19-30.

BRINK, L.and LUND, J. Dansk Rigsmal I-II. Lydudviklingen siden 1840 med soerligt henblink pa sociolekterne i Kobenhavn (1975). Copenhagen: Gyldental.

BLOOMFIELD, L. Language, London: George Allen & Unwin Ltd, 1933.

CEDERGREN, H. "Panama revisited: sound change in real time". Paper presented at NWAVE 10 [Philadelphia] (1984).

GARDE, P. "Réflexions sur les différences phonétiques entre les langues eslaves". Word (1961), no. 17, p. 34-62.

GIMENO, F. and MONTOYA, B. Sociolingüística. Valencia: Universitat de València, 1989.

GUY, G. "Variation in the group and the individual". In W. Labov (ed.), Locating language in time and space.(1980), New York: Academic Press, p. 1-36.

HOCKETT, C. F. "Which approach in linguistics is scientific?" Studies in Linguistics. (1950), no. 8, p.53-57.

HYMES, D. "Introduction". Language in Society. (1972).