An analysis of the state and direction of processes of language change that affect different morphological features in North Western Catalan enables us to obtain quantitative data on the behaviour of various factors with explicative value –social as well as linguistic, sociological, and pragmatic– that may at the present time influence the evolution of these processes. In this article we focus on the influence of a pragmatic factor, speech style, and here we also present a proposal for delimitation internal to a communicative situation

Contents

1. Introduction

2. Speaking style as an explanatory factor in processes of variation and linguistic change

2.1 Criteria for the definition of the speech style explanatory factor
2.1.1 Segmentation of the discourse
2.1.2 Degree of preparation of the discourse

3. Analysis of a case

4. Conclusions

5. Bibliography

1. Introduction

In linguistics, quantification has often been based on descriptive statistics, and this has made it possible to describe and summarise data in terms of frequencies. In this context, variationist sociolinguistics has clearly opted for looking for systems of data processing that involve inferential statistics. The latter refers to analysis of probabilities that will enable us to calculate levels of significance and reliability and thus make available a rigorous explanatory model to the study of linguistic variation. The second probability-based model proposed by Rousseau & Sankoff (1978), based on the VarbRul packages including the GoldVarb 2.0 application for Macintosh used in this study, is one of these statistical models, and one much used to measure the influence of linguistic and extralinguistic factors in the emergence of linguistic variants, since it converts real frequencies into theoretical probabilities by means of mathematical procedures.

Precisely this aspect of data processing, deriving from the sociolinguistics of variation, has been much used in different types of sociolinguistic research, especially on phonetico-phonological variables. This is the most studied linguistic level in the case of Catalan too (see Pradilla, 2002) the majority of which studies for the moment are moving away from the classic Labovian geographical distribution model consisting of analysing neighbourhoods or big cities, since the Catalan studies concentrate mainly on linguistic transition areas and / or areas far from the big conurbations.

Another aspect arising out of sociolinguistics and more particularly sociolinguistic variation studies, and frequently used in this field as a technique for the collection of data is the Labovian sociolinguistic interview, which provides a system to control stylistic variation, intended to elicit different speech styles (see Pradilla, 2001). In broad terms, Labov's technique (1966: chapter 4) sets out to obtain and to characterise five speech styles: casual, careful, formal, + formal, + + formal. The criteria used in our study to delimit a stylistic continuum produced in a pre-existent communicative situation, the plenary sessions of Tremp local government that make up our corpus, differ from the Labovian method we have just described and, in its own modest way, draws on a system based on contributions from the analysis of conversation, as we shall show below.

In this article (1) we propose to analyse the behaviour of a pragmatic variable, the speech style, as a factor for variation and linguistic change in different elements of nominal morphology in an area of interdialectal transition in the North Western Catalan major dialect area, the Conca de Tremp. The analysis is based on the use of frequency and inferential calculations (2) carried out with the GoldVarb 2.0 program.

2. Speaking style as an explanatory factor in processes of variation and linguistic change

As we have already mentioned, the contribution made by this article concerns the analysis of the effect of an explanatory factor of a pragmatic nature on morphological elements. We shall now look at the parameters used (a) for the delimitation of the communicative situation and (b) for determining the categorisation of the speech as either prepared or spontaneous.

2.1 Criteria for the definition of the speech style explanatory factor

2.1.1 Segmentation of the discourse

In sociolinguistics the difficulties that arise when delimiting the stylistic continuum have generated considerable disagreement over the establishing of the parameters involved in its segmentation. In our case, delimitation of the explanatory variable speech style relies on a previous factor, the segmentation of the communicative situation in speech events.

Segmentation was effected based on the criterion of hierearchic structure proposed in Tuson (1995: 49-51) for the study of conversation. (3) According to this criterion, the largest of the units into which conversation is structured is interaction, which can be seen to be delimited principally by the rituals of opening and closure. (4) The application of this notion to our analysis allows us to consider the totality of discourse that occurs in a local government plenary session constitutes a single interaction, since the rituals of opening and closure which are specifically indexed –s’inicia la sessió– (the session is declared open) / s’aixeca la sessió– (the session is declared closed) serve to frame the total communicative activity in question and establish the limits of the legal validity of the decisions taken. Continuing with Tuson’s criterion, the second of the structural aspects that characterise conversation and form an integral part of the interaction is sequence. In general it is accepted that the sequence becomes delimitable basically on thematic criteria. Thus, sequences that form part of every local plenary session would coincide with the different items on the agenda (order of the day). On the third level, Tuson has exchange, understood as a minimum unit of dialogue, comprises two or more turns within a sequence. While on the level immediately below this is the speech turn or utterance made by each participant which, as Adam states (1992: 158), is the maximum unit at monologue level.

The last of the units into which conversation can be segmented, according to Tucson, is the act, which comprises the illocutionary and interactive functions at any given moment.

In our case, to evaluate correct assignment of different linguistic manifestations in the corpus to one or other of the speech styles in question, we have worked on the data at the level of the utterances of each speaker. To illustrate the procedure, see examples (1), (2) and (3):

(1) The following fragment contains two speech turns generated in the context of one of the exchanges included in the only sequence in the interaction. The example provided is an urgent extraordinary plenary with one item on the agenda. (5) The first utterance is solemn and serves as an introduction to the second, which is a read institutional declaration condemning terrorism. Apart from this providing exemplification of the segmentation, we can ascribe to the categories of spontaneous cases those instances in which one of the linguistic varieties analysed, the masculine definite article. The prepared speech style is underlined at the right and is justified in point 2.1.2.

"A: (el) plenary session has one item on the agenda
an institutional declaration                prepared speech style (a)
which all persons and political groups forming a part of Tremp town council have previously agreed upon, and which will now be read by
(el) clerk to the council                 prepared speech style (a)
S: whereas on the (el) day of the
tenth of April of nineteen ninety-five prepared speech style (a)
the (el) infantry brigadier M                 prepared speech style (a)
someone closely connected to the town and municipality of T
has been brutally murdered in SS
Closely connected since in addition to having been a student at the basic general academy for junior officers, he was married to a local person
and spent his holidays with his family at T where he was much loved and appreciated by all.
and whereas the town council of T is bound to express publicly its (el seu) rejection and condemnation of this act
in the name of the municipal corporation of the people of T
and in general in the name of all people of good faith who live in peace and coexistence
it is proposed that the council in full accord should firstly
express publicly its (el seu) rejection and repudiation of the brutal murder
of infantry brigadier M                          prepared speech style (a)
and condemn in unmistakably clear terms the perpetrators of this act (TPLENS3: 46-72)

(2) The second extract contains seven short overlapping turns produced in the context of one of the exchanges included in the interaction, specifically the reports time toward the end of the session. It is an exchange of short or very short turns in the course of a tense discussion. The turns spoken by the mayor-moderator (A:) attempt to contain the transgressions of councillor (J:), as can be seen in the successive refusal to let the other take a turn. Each one of the occurrences of the masculine definite article which appears in this fragment can been taken to be cases spontaneous speech style, and this too we justify at 2.1.2.

"A: [Mr. J. there is no answer]
J: [and you've ended up saying that the error was administrative
well, you've got a flaming nerve!]     spontaneous speech style (b)
A: [no. I didn't say the error was administrative]
J: [and in second place]              spontaneous speech style (b)
A: And I don't blame anybody
J: and in the (el) second place
A: I assume full responsibility     spontaneous speech style (b) and if you want to take legal action, then
you know the [el] way to do it [angry].
J: I regret your tone (el to)              spontaneous speech style (b)
(TPLENS5: 2040-2051)

(3) The third extract contains two turns produced in the context of one of the exchanges in a sequence of the interaction, dealing with the decision to organise a tribute to a mayor of the time of the República (the regime overthrown by Franco). The first of the speech turns is read, while the second is the reply. We feel that it provides a very good example of the transition from the prepared speech style to the spontaneous speech style –which we comment on at 2.1.2. In this case we can observe the occurrences of the linguistic variable masculine definite article, underlined in the transcript, and the linguistic variable first person plural weak pronoun, in italics.

"C: [the turn is longer]
(:::) you are the mayor of a town council with a budget that doesn't balance and is getting progressively worse
and given these circumstances
you will not find an heir
do not fear a motion of censure,
it will be more and more difficult for us (ens) and more
unpopular                                            prepared speech style (a)
to balance the books.
Mr. A
Mr. Ar
all (tots els) mayors                           prepared speech style (a)
simply because they are mayors
if they are good mayors they
(::) already
already have their own tribute
(:) let us not go into more detail than that
let us not recall the past and let us make sure those days do not return
nothing more
and and let us concern ourselves (preocupem-mos [preoku'pemos]) with
what is happening now                   prepared speech style (a)
I will see to it for the good of all
that this vote (ague(s)t vot) is secret
there will be many different opinions on the subject
and there's no need for - need for grudges
Ar: in terms of,
my party's
the least (lo mínim) I can do     spontaneous speech style (b)
we have been going over and over this issue
(ague(s)t tema) since April
and in our case it was just one of the things that had to be done
just one more
not the only one                       spontaneous speech style (b)
the only (lo únic) thing we have done is put forward
the proposal and get on with it"

  (TPLENS2: 1178-1210)