The aim of this research is to contribute empirical and descriptive data to the study of different aspects of contemporary oral usage and the operation of certain processes of linguistic change and variation. In particular, we are concerned with helping to complete and update the available data on the linguistic characteristics of the Conca de Tremp, an area of intradialectal transition between the Ribagorçanopallarès and Lleidatà areas. We will also analyse the situation and direction of the processes of linguistic change that affect the definite article, certain full forms of the weak pronouns, and demonstrative and possessive adjectives. These analyses were carried out in order to obtain data on certain factors contributing to the evolution of processes of linguistic change and variation(1).

1. Introduction

2. Some theoretical and methodological aspects

3. The speech of the Conca de Tremp: an update

4. Processes of linguistic change and variation: an analysis of some representative cases
4.1. Linguistic variables and groups of explanatory factors
4.2. Results

5. Conclusions

6. Bibliography

1. Introduction

The Catalan language has an important dialectological tradition that has concentrated its efforts on certain areas in the north-western domain, whether because of its interest as an area of interlinguistic transition (Franja de Ponent), interdialectal transition (Tortosí) or, more recently, because of the demographical importance of the area of Lleidatà. However, there are still holes in the north-western sphere that require filling: from a dialectological standpoint, monographs and updates are needed, and as regards the field of sociolinguistics, studies of attitudes in the different speech communities are required (see Sistac, 1998/2000: 227). The speech of the Conca de Tremp, which has scarcely been the target of differentiated study and treatment, presents the characteristics of an area of intradialectal transition which, as we shall try to prove, shares features with its neighbouring subdialects, the Ribagorçanopallarès and Lleidatà areas.

Initial hypotheses suggest that the particular characteristics of the area would permit the study of internal modifications within the language, caused by linguistic factors and other, external factors. These characteristics include: geography (pre-Pyrenean, Mediterranean-style valley); history (late Romanization; appears to be no earlier than the fifth century, and frontier zone); economy (agricultural and cattle-farming area preserving the traditional character of the commerce and services area in its immediate sphere of reference) and society (strongly affected by emigration during the twentieth century; currently has a stable demography, but with one of Catalonia’s oldest populations).

2. Some theoretical and methodological aspects

The theoretical framework of our objectives is based on contributions from the field of varieties in contact and, in particular, from the analysis of processes of linguistic convergence and of the characteristics of the standard variety of Catalan and its relationship with the geographical varieties (Pueyo, 1980; Lamuela, 1994; Segarra & Farreny, 1996; Bibiloni, 1997), especially with north-western Catalan, and postulates used in quantitative sociolinguistics (Labov, 1966; Sankoff, 1978), allowing us to study linguistic change in its development phase and to describe variability in terms of the variable rule by assigning probabilistic values that enable us to predict the probability of variants (using specific software tools such as the VarbRul series of computer programmes).

As regards methodology, a number of preliminary studies (Romero, 1996, 1998a and 1998b) had offered indications for the identification of different explanatory factors (individual, social, sociolinguistic and situational) affecting the process of linguistic change in the features analysed in this speech community. These prospective studies also enabled us to justify the need for complementing frequential processing of data (which helps describe contemporary oral use), with probabilistic processing (enabling us to analyse the regression of certain traditional forms and the ensuing progression of standard forms).

Linguistics has tended to perform quantifications using statistical analysis, based on two main ends obtained by applying different quantitative treatments to the data used: these data are described and summarised using frequencies from descriptive statistics; then, meaning and reliability are calculated and the conclusions are applied to a larger set of data. These operations are made possible by the probabilities obtained with inferential statistics. Quantitative sociolinguistics, therefore, searched for formulas in applied statistics to allow it to incorporate a rigorous explanatory model into the study of variation. Rousseau & Sankoff’s second probabilistic model (1978), the basis of the VarbRul computer programmes (which include GoldVarb 2.0 for Macintosh), is one of the most commonly used statistical models for measuring the influence of linguistic and extralinguistic factors in the emergence of linguistic variants and to convert real frequencies into theoretical probabilities using mathematical procedures.