Navigation – Plan du site

AccueilNuméros84A variationist analysis of the su...

A variationist analysis of the subjunctive mood in a sociolinguistic corpus of spoken Francoprovençal

Une analyse variationniste du mode subjonctif dans un corpus sociolinguistique du franco-provençal oral
Jonathan R. Kasstan

Résumés

Les études variationnistes sur le subjonctif dans les langues romanes restent centrées sur les langues dominantes des états-nations. A l’inverse, très peu de recherches se concentrent sur les variétés dites « minoritaires » (souvent qualifiées de « langues régionales ») avec lesquelles ces langues dominantes sont en contact. Le cadre labovien invoqué pour élucider la variation du subjonctif n’a pas encore été testé de manière adéquate sur ces langues menacées. En utilisant un nouveau corpus sociolinguistique novateur de variation en francoprovençal parlé, l’analyse variationniste présentée dans cet article offre une investigation préliminaire de la forme synchronique du subjonctif dans une variété, et évalue dans quelle mesure les facteurs sociaux et linguistiques influencent la sélection des variantes. Les résultats présentés suggèrent un nivellement progressif du subjonctif, qui devient de plus en plus dé-sémantisé dans la communauté. Les données disponibles suggèrent une perte progressive du subjonctif qui est dû à de multiples facteurs, d’une part une discordance des temps verbaux, d’autre part la perte de régularité morpho-phonologique, sans qu’aucun facteur social joue un rôle dans la variation.

Haut de page

Texte intégral

Acknowledgements

The author thanks Anne-Gabrielle Bretz-Héritier for her help with the research design and in securing participant interviews as part of this project, and Philip Comeau for his helpful comments on a draft manuscript. The author is indebted to the speakers from whom the data presented here have come. This work was supported by a Leverhulme Trust Early Career Fellowship (ECF-2017-584).

1. Introduction

1Reduction and loss in the subjunctive mood and its productivity in speech has been a recent research focus in variationist-sociolinguistic approaches to grammatical variation in Romance. This work, which prioritises the distribution of and conditioning factors contributing to subjunctive production, has in particular sought to build on long-standing debates regarding the supposed desemanticisation of subjunctive-mood morphology as well as distributional differences in subjunctive vs. indicative constructions across Romance varieties (e.g. Carlier et al. 2012 and Lamiroy & De Mulder 2011). While earlier scholarship has demonstrated little consensus in terms of what factors constrain mood selection, whether a semantic contribution is made by the use of the subjunctive, which verbs trigger the subjunctive, or whether subjunctive use coalesces around more formal speech styles, the privileging of variationist protocols in addressing such empirical questions has borne fruit. Recent studies such as Poplack et al. (2018) for example have marshalled large-scale corpora in order to upend classic claims vis-à-vis canonical subjunctive selecting contexts across Romance, and have provided detailed lines of evidence in arguing for clear desemanticisation effects cross-linguistically. What remains untested however is the extent to which such a framework can also contribute to classic debates concerning structural change in severe language endangerment. In advanced cases of sociolinguistic change such as these, reductions in the frequency of complex sentence types like subjunctive subordination are frequently reported in the literature, often with little empirical weight (cf. Tsitsipis 1984, Hill 1989, Nützel 2009). Moreover, while most quantitative studies on the subjunctive in Romance have focused on dominant nation-state languages, the few existing studies on regional, minoritised Romance varieties in contact with these dominant languages may well offer an alternative picture of variation and change in the subjunctive paradigm. Yet they remain heavily understudied, with little understanding of their current synchronic shape.

2This paper has three aims. First, it introduces a sociolinguistic corpus of spoken Francoprovençal, a severely endangered Romance language in contact with French and Italian in Europe, and English in North America. Second, it provides an account of the synchronic shape of the subjunctive paradigm among 36 speakers of one variety of Francoprovençal as spoken in Savièse (Canton of Valais, Switzerland). Francoprovençal benefits from a rich tradition of research in diachrony, but the present state of the language remains very much under-documented from a quantitative variationist perspective. Third, a preliminary analysis of factors contributing to subjunctive selection in the sample is presented. The evidence, taken from over sixty hours of natural speech samples and structured elicitation tasks, and triangulated with descriptive accounts and existing linguistic-atlas material, suggests that a change is progress is underway in the community towards a progressive weakening of the subjunctive mood, as one might predict given the existing literature. However, a synthesis of findings from the variationist analysis (n=176 tokens of the subjunctive) deployed here offers nuance in understanding this change. The evidence suggests that a decline in subjunctive use in the community is likely of multiple causation, involving (a) a loosening of tense matching in matrix and subordinate clauses, (b) variability with the indicative mood, and (c) morphophonological reduction.

2. On the subjunctive mood in lesser-studied Romance varieties

3The subjunctive mood is perhaps one of the most widely studied morphosyntactic phenomena in Romance. In the case of French, for example, most studies on the subjunctive stress the volume of work that has been dedicated to delimiting the conditions that trigger the subjunctive, as well as the semantic values that subjunctive forms can carry. What is clear from this extensive body of work, however, is that there is ‘no consensus […] on the way that class of subjunctive-selecting verbs is constituted, nor what the semantic contribution of the morphological form is’ (Poplack et al. 2013: 140). That said, there are areas of agreement in this work, within which it is generally accepted that the subjunctive – which can express a spectrum of meanings associated with attitudes, emotions, opinions, evidentiality, possibility, volition, necessity etc. – is obligatorily triggered after a class of subjunctive-selecting governors (including non-verbal governors) in a matrix clause (even if these triggers can be constrained by the degree to which the user wishes to commit themselves to the reality or truth-conditional value of the complement).

4With the advent of the variationist enterprise in North America, a good body of work now exists on the study of different Acadian French varieties, where work on the subjunctive has been particularly fruitful. For instance, while in spoken Metropolitan French it is generally accepted that the imperfect subjunctive had fallen into disuse by the 17th century (Fournier 1998, cited by King 2013: 39), this is not the case some conservative Acadian French varieties. For example, Comeau (2019) provides an analysis of subjunctive use in Baie-Saint-Marie (Nova Scotia). Unlike in other Acadian varieties, the subjunctive mood in Baie-Saint-Marie is demonstrated to be robust. Indeed, in Comeau’s corpus, the imperfect subjunctive is the most frequent tense used by speakers where falloir is produced in the conditional in the matrix clause. However, in general, Comeau does refer to a decline in the imperfect subjunctive in other varieties of Acadian French, even if it is clearly still in use in Baie-Saint-Marie. On the balance of evidence, Comeau argues that, collectively, the progressive loss of the imperfect subjunctive and the strength of the tense concordance effect (especially for having the conditional in both main and embedded clauses) play a role in triggering the subjunctive in becoming a linguistic variable to begin with. Comeau marshals this as evidence for arguing that the concomitant loss of the imperfect subjunctive with a tense concord effect represents ‘a weak point in the grammar which can allow for the subjunctive to then vary with other moods’ (2019: 41). Studies such as these have demonstrated how the subjunctive can be operationalised as a linguistic variable under classic assumptions that go with delimiting the variable context.

5While quantitative studies on the subjunctive in dominant Romance varieties (French, Spanish etc.) are increasingly available, including on some obsolescing dialects (e.g. as found in some Newfoundland Acadian communities), comparatively far fewer studies have been dedicated to minority and endangered Romance varieties, often labelled ‘regional’ languages. Among them, Jones (2000, 2014) has documented language variation and change in the Guernésiais subjunctive paradigm, where evidence from historical materials as well as Jones’ own fieldwork demonstrates that the subjunctive remains robust in subordinate clauses, across present and imperfect tenses. Moreover, many of the contexts in which the subjunctive was attested in medieval times, she argues, still trigger the mood today (2000: 200). Thus, while Guernésiais might be in terminal decline on Guernsey, Jones nonetheless reports that it is doing so with its morphology intact. Indeed, of all contexts requiring the imperfect subjunctive, Jones (2000: 193) reports a prevalence rate in her corpus of 57%. However, she also concedes (a) that ‘triggers for the subjunctive often appear to be lexically rather than semantically motivated’ (p.189); (b) that most of the examples of the imperfect subjunctive now only involve a handful of verbs: including ‘aver (‘to have’), ête (‘to be’), veies (‘to see’) and v’nir (‘to come’)’ (p.191); and (c) that research participants will avoid the subjunctive in spoken Guernésiais in contexts where they are unsure of a particular morphological form or the conjugation of an uncommon verb (p. 189). More recently Jones (2014) synthesises her survey work on the subjunctive mood across all four Norman varieties (Mainland Norman, Jèrriais, Guernésiais, and Sercquiais), where she observes that the imperfect subjunctive is largely only triggered by fautrar (‘to be necessary’) and verbs of influence. It is also noteworthy, as Jones points out, that ‘the imperfect subjunctive of Insular Norman shows no sign of disappearing before its present tense counterpart’ (2014: 130). Such observations, which tend to run contrary to variationist findings in dominant contact-language settings such as Metropolitan French, remain in a minority. One reason among others why such studies are few and far between is that eliciting tokens of the subjunctive (particularly in the imperfect) remains a significant methodological hurdle. Most recently, Auger & Hendrickson (2021) demonstrate this point clearly in their study of the subjunctive in Vimeu Picard (a threatened regional language spoken in northern France), where they report that there are now so few fluent speakers that eliciting any instances of the subjunctive at all becomes extremely difficult. This underscores the urgency with which such endangered varieties must be documented and described, while ‘native speakers and new speakers are still able to provide data and grammaticality judgements, while also providing expertise as speakers’ [author’s translation] (Auger & Hendrickson 2021: 12).

3. The subjunctive mood in Francoprovençal

6As with its sister varieties, in Francoprovençal the subjunctive mood can be triggered in an embedded clause by a predictable set of verbal matrices (e.g. verbal matrices expressing necessity or obligation, such as fali̩ ‘to be necessary’). However, unlike the dominant languages with which Francoprovençal is in contact (French and Italian in Europe; English in North America), the language does not benefit from a long tradition of codification, standardisation, or elaboration of function; nor is there a widely-accepted prestige norm to select from for these purposes. Further, the language is heavily fragmented, which complicates any presentation of broad patterns of subjunctive usage in the Francoprovençal-speaking region as a whole. As the language is under-documented, it is not possible to give any reliable overall impression of the intactness of the subjunctive mood. However, it is clear that, given (a) the status of the language as severely endangered; (b) its restricted domains of usage; and (c) a dwindling base of attriting speakers belonging largely to an inter-war generation, previous reports on grammatical structure make it clear that linguistic change has been underway for some time. For example, in one of the few grammatical sketches that attempt to provide a broad picture of the subjunctive across all sites in which it is spoken, Stich (1998: 116) suggests that four patterns are possible in Francoprovençal, as outlined in (1-4) below:

(1) ‘Subjunctive mood is maintained with a clear separation of present and imperfect paradigms (this pattern is rare in Francoprovençal but nonetheless attested);

(2) Only the present subjunctive is used, in its regular form;

(3) Only the imperfect subjunctive is used, in its regular form, where the morphology conveys present-subjunctive readings;

(4) A mix of the two temporal references with a present subjunctive governor’ [author’s translation]

  • 1 In an augmented study, Stich (2001: 261-262) further clarifies that patterns (3-4) can be typical o (...)

7While no empirical support or extended commentary is provided to accompany these claims, Stich does state that the proposed pattern (1) is thought to be rare. Further, we can infer from (2-4) that a weakening of the subjunctive mood was observed at the time of writing. Although (2) is not surprising: given the oft-cited claim that the imperfect subjunctive is moribund in spoken French too (see e.g. Jones 2014: 129, King et al. 2018: 16), we might expect similar observations in peripheral minority varieties in contact with French, where all Francoprovençal speakers have been sequentially bilingual and/or French-dominant for some time. Conversely, (3-4) taken together indicate the attestation of variable production of both present and imperfect subjunctive paradigms1. Indeed, Favre (2009: 170) has gone as far as to suggest that imperfect subjunctive usage with a present subjunctive reading is ‘widespread’ in the region. Further evidence for (4) can also be gleaned from the available published pedagogy for learners of Francoprovençal. Recent revitalisation efforts geared towards valdôtain (the label accorded to varieties of Francoprovençal spoken in the Aosta Valley) has led to codification and standardisation initiatives, including the publication of an online grammar. With regard to subjunctive selection, the materials state:

  • 2 All instances of the target language are marked using local orthographical conventions.

‘We have remarked that the present subjunctive is used less and less, even in cases where tense concordance demands it. So, it is replaced by the imperfect subjunctive. The following two utterances have the same meaning’ [author’s translation] (Région autonome Vallé d’Aoste 2022: 2).2

(1)

Dze

vouì

Que

te

Vegnèye

avouì

1SG

like.COND

that.COMP

2SG

come.SUBJ.PRES

with. COMP

me.1SG

‘I would like you to come with me’

(2)

Dze

vouì

Que

te

Vegnisse

avouì

1SG

like.COND

that.COMP

2SG

come.SUBJ.IMPERF

with.COMP

me.1SG

‘I would like you to come with me’

8This suggests the possibility that a loss of tense matching is indeed taking place in the subjunctive mood, which accords with Stich’s (1998) description in (4) above, although it is not clear to what extent this is evident in (a) naturally occurring speech production; (b) in varieties other than in valdôtain; or (c) whether there is variation in the production of these forms, or indeed variation in perception of the semantic value of the verbal morphology. These are all open empirical questions, and this article will focus on (a) and (b) specifically, leaving research on (c) for future work. Owing to overviews such as Stich (1998), Favre (2009), and the available evidence presented in the valdôtain learner materials, this study’s primary working hypothesis (H1) will be that speakers will demonstrate variation in the realisation of present and imperfect subjunctive forms where the matrix verb demands present-subjunctive selection for tense concordance.

9To test this hypothesis, the present study focuses on the variety of Francoprovençal spoken in the commune of Savièse in the Canton of Valais (Switzerland, see Figure 1), a methodological choice guided by a number of considerations. First, there is at least one available dictionary and partial grammar (Favre & Balet 1960), which was revised and augmented more recently by a local association dedicated to linguistic and cultural preservation (Fondation Anne-Gabrielle & Nicola-V. Bretz-Héritier 2013). These resources provide an attested historical record of the subjunctive in Saviésan Francoprovençal. Second, data collected from two older fluent speakers (1 male, 1 female) between 1994-2001 for the region-wide Atlas linguistique audiovisuel du francoprovençal valaisan (ALAVAL) (Diémoz & Kristol 2018) offers a baseline for comparison in terms of speech-production data, at least under careful elicitation conditions (see below). Third, while the language is today very much restricted to the most intimate domains of usage, it is still in use in the community (for details on vitality, see Zulato et al. 2018). As the subjunctive is an extremely infrequently occurring feature of the spoken language, the conditions offered in Savièse provide fruitful ground on which to (i) document and analyse subjunctive use, and (ii) test H1 as outlined above.

10In what follows, further details are provided on the social and linguistic context of Saviésan Francoprovençal before turning to this study’s research design.

4. Saviésan Francoprovençal: social and linguistic profile

11Saviésan refers to the variety of Francoprovençal spoken in the commune of Savièse (part of the District of Sion, located in the Swiss Canton of Valais; see Figure 1). Savièse itself is comprised of five villages, although these villages do not in of themselves constitute a salient level of sociolinguistic differentiation, as Luyet argued: ‘All of the villages of the commune of Savièse speak the same patois and all the neighbouring villages speak a different one. Savièse therefore constitutes a well-delimited linguistic system’ (1929: 6) [author’s translation].

12In Savièse, intergenerational transmission of Francoprovençal effectively ceased from 1950 after its use was prohibited in local schools, and at which point parents in Francoprovençal-speaking households began prioritising French on the advice of local education authorities. Research participants consistently report that, from the late 1940s onwards, parents would no longer transmit the language, even if it remained a language of communication in the home among adults. Today, while Francoprovençal is still spoken amongst the eldest generation, the language remains in terminal decline, in spite of valiant efforts on the part of local organisations to engage in language revitalisation strategies (see Zulato et al. 2018).

13Francoprovençal marks the subjunctive synthetically, and in Saviésan this takes the form of /ʃ/+V, where the vowel further marks person and number, as in examples (3) and (4) below (taken from the ALAVAL, Diémoz & Kristol 1994-2001). While final vowels are very common in Francoprovençal, they are unstressed and can often undergo apocope in connected speech (see e.g. Nagy & Reynolds 1997, Kasstan 2015).

(3)

[mĩˈʣeʃɔ]

I

ou

Kyé

Mëndzechó

ean

Dina

3SG

necessary.3SG

COMP

eat.SUBJ.PRES.1SG

Before

Lunch

‘I must eat before lunch’

(ALAVAL, 20.1 SavièseF)

[mĩˈʣeʃɔ]

(4)

Fodri

kyé

Mëndzechó

ean

Dina

should.COND.3SG

COMP

eat.SUBJ.PRES.1SG

Before

Lunch

‘I should eat before lunch’

(ALAVAL, 20.2 SavièseM)

  • 3 For reference, a full verbal tense paradigm of the auxiliary verb é̩tre ‘to be’ is provided in Appe (...)

14As examples (3) and (4) demonstrate, Francoprovençal as a partial pro-drop language can omit overt pronouns, but marks person synthetically on the verb. The ALAVAL also clearly attests to variation in the realisation of word-final vowels, which can demonstrate significant variability in vocalic quality (we return to this below). Examples (3) and (4) provide useful baseline indications for two older fluent Saviésan speakers, but it should be stressed that the ALAVAL relies on a combination of traditional dialectological methods for elicitation, combined with audio-video capture. For all intents and purposes, the ALAVAL interviews should therefore be considered as a very formal style of speech, recorded under careful elicitation conditions. No details are provided on any systematic variation in the subjunctive paradigm among the two speakers sampled from Savièse.3

15While the language is under-documented, one advantage of sampling from Savièse is that it not only formed part of the sample universe for the ALAVAL, but also because recent codification efforts in the region have led to the production of a reference grammar (Fondation Anne-Gabrielle & Nicola-V. Bretz-Héritier 2013). In addition, this same foundation publishes a local journal dedicated to language and culture – L’Ami du Patois – which offers example texts across a range of different genres and dialects in Valais. Therefore, historical evidence can also be called upon as a baseline for comparison with the empirical findings of this study, in accordance with King (2013: 27)’s argument on the importance of ‘considering a range of data sources and of examining all of the extant data for a particular linguistic feature’.

5. Research design

16Fieldwork for the study was conducted in 2019 in the commune of Savièse as part of a larger project on grammatical variation and change in Francoprovençal. Data were elicited from among a judgement sample of 36 participants (Table 2), gathered via the snowball method. Although a random sample would have been preferable for sociolinguistic analysis, a judgement sample is a necessary part of endangered-language research as (a) the target language is often stigmatised, as has traditionally been the case in the present context, and (b) speakers are geographically and socially distributed in the community in a non-random way.

Table 2. Final sample

older speakers

younger speakers

female

13

3

Male

16

4

  • 4 Participants in the corpus are pseudonymised and coded by region, sex, age group and speaker identi (...)

17The final sample under study comprises: older fluent speakers (n1=29) and younger proficient speakers (n2=7). The older speakers are an inter-war generation who acquired Francoprovençal from birth. The younger speakers are a post-war generation who acquired Francoprovençal typically in late or post-adolescence. For this latter group, while French was prioritised in the home by immediate care givers, they were nonetheless exposed to Francoprovençal through its presence in the wider sociocultural environment (e.g. grandparents, parents, older siblings, the community).4

18Owing to high illiteracy rates in Francoprovençal, it was necessary to adapt standard variationist protocols (Labov 1984) so as to ensure that ‘literacy [was] not a prerequisite for participation’ (Blainey 2017: 581). Data collection involved two recorded sessions. In the first session, participants were presented with a project brief detailing the study motivations in very general terms that focused on language documentation and revitalisation. No details concerning linguistic features of interest were given so as to not influence or bias interview recordings. It was made clear during the structured elicitation tasks that the researcher was interested in participants’ own variety of Francoprovençal. Then, a sentence-translation task was administered to participants in order to elicit a more monitored speech style. This aspect of the research design follows Klinger & LaFleur (2007: 336), who employed translation exercises as substitutions to classic reading passages and wordlists, with the aim of maintaining the formality and structural consistency as seen in traditional variationist studies. It was therefore necessary to first prompt participants in French. The items included in the elicitation task (Appendix ii) contain instances of known subjunctive triggers in Francoprovençal, along with other variables under analysis, and fillers were also interspersed to conceal the features of interest. Other elicitation tasks used to target additional variables of interest included participants recounting the picture-book story A Boy, a Dog, and a Frog (Mercer Mayer), although no tokens of the subjunctive were elicited for this task. Lastly, A perceived-acceptability task was also administered towards the end of the first session which does not form the focus of the discussion here. Participants were then invited to attend sociolinguistic interviews in groups. Group interviews were conducted in the target language between older and younger speakers so as to elicit a more casual, less monitored speech style. While conversational modules (Labov 1984) were adopted, the aim of the group discussions was to elicit the most vernacular style, and so rather than the author coordinating the group discussions (as a non-native speaker and a non-community member), they were directed by community volunteers. This enabled the author to fade into the background so as to allow the participants to guide the discussion themselves. As far as possible, both styles are analysed below. However, it was not possible to elicit tokens across styles for every participant in the study, owing to the old age, frailty and isolation of many of the speakers (and thus the fact that not all participants could attend group interviews), and so the data are fragmentary. This is, however, to be expected of research undertaken on severely endangered languages, as compared with the bulk of variationist research on dominant languages in urban settings. The author acknowledges here the constraints that the nature of the data places on the discussion and interpretation of findings.

19A further methodological hurdle concerns the Principle of accountability (Labov 1972: 72): it is necessary to ascertain not only where speakers select the subjunctive, but also where they could have even if they did not. This study follows Poplack (1992) in its approach to circumscribing the subjunctive as a variable. In considering the state of the subjunctive in Laurentian French, Poplack (1992) proposes that the variable context for the French subjunctive be based on local community norms rather than on standard norms, which clearly would present problems from the perspective of the present study. Her approach consisted of first extracting all tokens which evidenced unambiguous subjunctive morphology at least once in her Ottwa-Hull corpus, and then establishing a list of the main-clause verbs (falloir, vouloir etc.) from which to carry out variationist analysis. Therefore, all tokens (whether they involve subjunctive morphology or not) which occurred in an embedded clause following a subjunctive selecting main clause verb were extracted, rendering a locally-relevant picture as to how the subjunctive occurs in a given variety. The set of ‘subjunctive-selecting’ governors identified was therefore taken to be the locus in which subjunctive selection was a possibility. While this methodology was devised for French, it is proposed here that the methodology is also applicable to the study of the subjunctive in Francoprovençal. Thus, all instances of subjunctive selecting verbal governors that preceded the complementiser were extracted where mood choice in the embedded clause was unambiguous. Using this method, n=176 tokens were extracted from a corpus comprising 36 participants. While this token count is low by comparison with typical variationist studies on e.g. the phonetics and phonology of dominant languages, it is consistent with variationist studies on the subjunctive mood in natural speech production (e.g. Comeau 2019), contra, say, larger datasets available from existing corpora.

20For the purpose of this preliminary analysis on the synchronic shape of the subjunctive in Francoprovençal, the variable context is limited exclusively to subordinate clauses triggered by fali̩ (‘to be necessary’); oo͡ui (‘to want to’); and anma (‘to like to’) in the matrix clause. Non-verbal complements as well as crystallised forms that can appear in independent clauses are excluded from the present analysis. The choice of governors is motivated by two reasons. First, as with verbs of necessity and desire in Reference French, these verbs in Francoprovençal encompass all of the semantic and structural features that promote the triggering of the subjunctive: fali̩ in particular has an impersonal subject, and all three governors take a subordinate clause as a complement. Second, this paper adopts Meyerhoff (2017)’s framework that calls for a sociolinguistically oriented approach to language documentation and description. Therefore, the present study aims to (a) offer a picture of the Francoprovençal subjunctive in synchrony, and (b) establish which linguistic and social factors (if any) constrain subjunctive selection. The battery of independent variables that were tested on the dependent variable (embedded clause mood) included: matrix clause verb and tense; embedded clause verb and tense; grammatical person in the embedded clause; presence or absence of final vowel on the embedded subjunctive form; speaker sex, age, speech style. The analysis adopts as its primary statistical measures: descriptive frequency statistics, Fisher’s exact (two-tailed) tests of statistical significance, and mixed-effects logistic regression where possible.

6. Results

21Table 2 presents the overall subjunctive rate in the corpus. As the data suggest, rates of subjunctive usage for oo͡ui and anma are robust, although token counts are low. Conversely, there is variation in use of the subjunctive for the governor fali̩, by far the most frequent in the corpus, where 79% of tokens were selected in the subjunctive, and 21% in the indicative.

Table 2. Overall subjunctive rate in the corpus

Governor

Indicative

Subjunctive

TOTAL

oo͡ui

0

61

61

fali̩

22 (21%)

83 (79%)

105

Anma

0

10

10

TOTAL

22

154

176

22Turning to embedded verbs, as Poplack et al. (2013: 167) make clear, every verb is ‘theoretically eligible to be embedded under a subjunctive governor, and thus theoretically eligible to carry subjunctive morphology’. For ease of analysis, Table 3 gives a breakdown of embedded verbs (n=17) by binning them into two categories based on frequency: Group 1 consists of aa̩ (‘to go’), avo͡uêe (‘to have’), é̩tre (‘to be’), fé̩ṛé (‘to do’) and Group 2 consists of all others. From the table it is clear to see that Group 1 embedded verbs are selected in the subjunctive near-categorically, whereas Group 2 verbs demonstrate more variability.

Table 3. Embedded verbs by group

Embedded verbs

% of pool

% subjunctive rate

Group 1

29

94

Group 2

71

85

Table 4. Subjunctive and tense matching

embedded

TOTAL

matrix

Present

Imperfect

Present

157 (94.6%)

9 (5.4%)

166

Imperfect

1

2

3

Conditional

9

0

9

TOTAL

167

11

178

Table 5. Subjunctive and tense matching (fali̩ only)

embedded

TOTAL

matrix

Present

Imperfect

Present

96

4

100

Conditional

5

0

5

TOTAL

101

4

105

23The results also suggest that there is some tense mismatching present in the corpus. As Table 4 shows, most tokens of the imperfect subjunctive are largely observed following a present-tense matrix clause, and vice versa, although they are very few in number. In narrowing down the analysis, no tokens of the present subjunctive emerge in imperfect matrix clauses when just the governor fali̩ is selected, but four tokens of the imperfect are found in embedded clauses selected by a present-tense governor, three of which are offered by just one speaker, as in (5), below.

(5)

[fuəʃ]

a

cha̩te

o̩uṛé

lo

matën

ou

kyé

To

foo͡ue̩ché

A

mijo̩n

PREP

seven

hour.PL

ART

Morning

necessary.3SG

that.COMP

2SG

be.SUBJ.IMPERF

PREP

house

‘At seven o-clock in the morning you must be at the house’

(VaMO17.1.78)

24In relation to Stich’s (1998) description of possible permutations in the subjunctive mood, then, we observe in our sample of 36 Saviésan speakers a pattern akin to (4) in the results, with a mix of both present and imperfect temporal reference across matrix and subordinate clauses, but this is largely only observed for two speakers in the sample who provide the bulk of the imperfect tokens.

25Turning to internal linguistic constraints on subjunctive selection, the research design considered whether grammatical person might trigger subjunctive use in the embedded clause. However, the vast majority of tokens occur in the 2nd person singular (see Table 6), skewing the effect of grammatical person as a factor group on subjunctive selection. This is an artefact of the research design, in that the sentence-translation task largely included 2nd person constructions. This issue aside, singular personal and impersonal forms predict subjunctive selection to a statistically significant degree (Fisher’s exact, two tailed, p = .0001).

Table 6. Subjunctive, grammatical person and number

Person/Number

Indicative

Subjunctive

TOTAL

1st sg.

3

23

26

1st pl.

2

1

3

2nd sg.

9

117

126

2nd pl.

2

3

5

3rd sg.

2

12

14

3rd pl.

3

0

3

TOTAL

21

158

176

26It was noted above that the subjunctive form in Francoprovençal consists of /ʃ/+V, where the final vowel encodes person. However, in connected speech, final vowels tend to undergo apocope in unstressed syllables, as has been noted in other varieties of Francoprovençal (see Nagy & Reynolds 1997 on Faetar). In the present corpus, there is variation in the realisation of final vowels, where a subjunctive form is produced with no final vowel in 79% (n=123) of tokens. In 99% of these cases, the following complementiser in the subordinate clause consists of a preposition or article with an initial front vowel (most often /a/, e.g. avou͡éi [ˈavwi] ‘with’, a [a] ‘the’). Concerning the 21% of tokens that did contain a final vowel, we observe that half of them (n=14) are followed by a part of speech consisting of an initial consonant (we discuss this observation in more detail below).

27On the social factors, no a priori assumptions were made about the extent to which the subjunctive might co-vary with external-linguistic constraints, so classic social variables were included in the analysis. Beginning with speaker sex (Table 7) and age (Table 8), few distributional differences are observed among females and males, and across age groups in the corpus. As the tokens are distributed unevenly across factors and factor levels, tokens were fitted to a mixed-effects logistic regression model using the package lme4 (Bates et al. 2019, R Core Team 2019), with speaker as a random intercept. However, neither social variable was selected as statistically significant in the model (see Table 9).

Table 7. Embedded clause mood choice and speaker sex

Indicative

Subjunctive

TOTAL

Female

15

80

95

Male

6

75

81

TOTAL

21

155

176

Table 8. Embedded clause mood choice and age group

Indicative

Subjunctive

TOTAL

Older

18

114

132

Younger

3

42

45

TOTAL

21

156

176

Table 9. Output of model fit to data from 36 speakers, males and females, older and younger

Fixed effects

Estimate

Std. Error

z value

PR(>|z|)

(Intercept)

-0.9542

0.2777

-3.436

0.0006

Sex

0.6014

2.7420

2.766

0.84901

Age

1.1542

2.6118

0.442

0.65856

model_1<- glmer(EMBEDDED_CLAUSE_MOOD_CHOICE ~ SPEAKER_AGE + SPEAKER_SEX + (1|SPEAKER_ID), data=dat, family=binomial)

28Owing to the fact that the existing literature has demonstrated clear stylistic differences in subjunctive selection in dominant-language contexts (e.g. Gadet 1992:89-90 on low-register spoken French), speech style is also considered here (Table 10).

Table 10. Embedded clause mood choice and speech style

Indicative

Subjunctive

TOTAL

casual speech

7

0

7

translation task

14

155

169

TOTAL

21

155

176

29As Table 10 shows, very few tokens were realised in casual speech, but the available evidence indicates that where a subjunctive governor is produced in this style, only tokens of the indicative emerge in embedded clauses in the corpus, a pattern reversed in careful speech (a pattern we return to below).

7. Discussion

30Recall that the two research questions advanced above pertain to (a) the synchronic shape of the subjunctive paradigm in Saviésan Francoprovençal, and (b) whether patterns in production map onto Stich’s (1998) broad description outlined above. Concerning (a), at least two observations can be made. First, the subjunctive appears to vary with the indicative, and, in the present dataset, the variation emerges with the most frequent governor in the corpus (fali̩). This is illustrated in examples (6-7) where the subjunctive form for étre (‘to be’) is instead drawn from the indicative paradigm (cf. Appendix i). That the subjunctive and indicative moods should vary in Francoprovençal is not surprising given that this is also widely attested in both dominant Romance varieties (e.g. Poplack et al 2013 on French, Digesto 2022 on Italian), and other regional minoritized varieties with which they are in contact. For instance, in the Croissant linguistique – an area comprising central Gallo-Romance varieties transitional between oïl, oc, and Francoprovençal-speaking zones – Escher (2022) identifies that the present subjunctive can fall together formally with the present indicative, and functionally with the imperfect subjunctive. What motivates this variation in Francoprovençal remains an open question that cannot be determined by the low token count presented here for just one governor. Tentatively, in the case of (6-7), it is noteworthy that there is little morphological salience distinguishing subjunctive and indicative forms for étre in Saviésan Francoprovençal, and previous work in other contexts has demonstrated this factor to be an important predictor in subjunctive selection (Poplack et al. 2013: 167).

(6)

[ʃɔ̃]

i

ou

kyé

chon

o͡uéi

3SG

necessary.3SG

that.COMP

be.IND.3PL

PREP

‘They must be there’

(VaFO3.2.182)

(7)

[ʃjɔ̃]

i

ou

kyé

chion

o͡uéi

3SG

necessary.3SG

that.COMP

be.IND.3PL

PREP

‘They must be there’

31Second, the results presented make it difficult to conclude whether or not the subjunctive still forms part of everyday spoken Francoprovençal in Savièse. First, concerning the structured elicitation task, the subjunctive remains robust in the data to a point, but in this preliminary analysis the distribution of tokens is skewed heavily towards fali̩, with only a small number of embedded-clause verbs (n=17). The analysis here is limited by the constraints of the research design, with just three matrix governors considered; a larger study may reveal other frequent subjunctive-selecting governors. In casual speech the subjunctive is clearly waning as a low-frequency variable: of the seven tokens in the corpus (all of which have fali̩ as governor), all contained an indicative in the embedded clause. This pattern is observed for both older and younger speakers, and, in an apparent-time design such as the one this study presents, such evidence would suggest a community-level decline in subjunctive use. However, this can only be confirmed with real-time evidence comprising too additional data of the variable in casual speech. It is necessary to stress though that such an imbalance of subjunctive tokens across styles is already well observed in other Romance studies.

32Whether or not the subjunctive has a productive semantic reading in Saviésan Francoprovençal is one further question that can be asked in light of the data presented above. The results demonstrate that the governor fali̩ accounts for nearly two thirds (60%) of the pool of available matrix verbs triggering the subjunctive in the corpus, much like studies on dominant Romance languages with which Francoprovençal is in contact. As the subjunctive may now be largely confined to embedded clauses with fali̩ as governor, on balance it is tempting to conclude that there is no semantic function and that subjunctive selection in Francoprovençal is now largely only lexically motivated. It is also clear that the findings presented here align with research on non-dominant, regional Romance varieties too, such as Guernésiais (Jones 2000: 189), which make similar claims. However, to confirm an absence of semantic reading more robust data would be required to compare discursive meanings associated with subjunctive forms against the internal factors discussed here.

33Concerning (b), following Stich (1998)’s descriptive observations on the state of the subjunctive in Francoprovençal, this study’s working hypothesis (H1) was that speakers would demonstrate variation in the realisation of present and imperfect subjunctive forms where the matrix verb demands present subjunctive selection. While this study does document some evidence of tense mismatching among speakers in Savièse, the prevalence rate of just 5.4% of imperfect forms in present-tense subjunctive selecting contexts, from just two speakers among a sample of 36, is at best marginal for a phenomenon described elsewhere as ‘widespread’ Favre (2009: 170). However, why imperfect subjunctive forms have emerged in the data at all is also worthy of discussion. Table 4 (above) illustrates that in at least one instance, an imperfect matrix-clause governor resulted in present-tense subjunctive selection in the embedded clause, and a further nine imperfect-subjunctive tokens were found in present-subjunctive selecting contexts. While there is very little available data to augment the analysis, we might posit that a levelling out of the imperfect subjunctive, along with a loosening of tense matching, may be giving rise to a decline in the subjunctive in spoken Francoprovençal in general, given too the style data presented above. A similar argument is proposed by Comeau (2019) in Baie-Sainte-Marie Laurentian French as a possible explanation for why the subjunctive becomes a linguistic variable to begin with. The evidence presented here suggests that Comeau’s observation may be generalisable. However, such a claim would need to be further tested on an augmented dataset.

34The battery of internal factors that were coded as part of the research design did not bare substantial fruit as far as assessing which factors are most likely triggering the subjunctive in an embedded clause are concerned, and part of the issue relates to artefacts of the research design itself. For instance, a majority of tokens were elicited in the 2nd person making it difficult to assess grammatical person as a predictor of subjunctive selection (as has been observed elsewhere, e.g. Lindqvist 1979), even if the distribution in the corpus was statistically significant (Fisher’s exact, two-tailed, p = .0001).

35Variation was also observed in the realisation of final vowels on the subjunctive form. Above it was suggested that this low realisation rate was linked to the phonological environment, in that those forms most often were followed by a complementiser in the subordinate clause consisting of a preposition or article with an initial front vowel, and that where final vowels were realised, in about half of cases the following part of speech consisted of an initial consonant. This observation is supported to a certain extent by evidence from the ALAVAL. Among the few available examples from the two Saviésan speakers in the Diémoz & Kristol’s (2018) corpus, where the subjunctive form precedes a consonant, the vowel is realised, although it is frequently reduced to schwa as one might expect, given the stress pattern in Francoprovençal (see Kasstan 2015). Where the subjunctive is followed by a part of speech consisting of an initial vowel, two strategies can be observed: either speakers produce a reduced form, or speakers compensate with an epenthetic palatal consonant (cf. 6 & 7, below).

(6)

[ˈwiʃə]

[a]

i

Paṛe

Wi

kyé

Me

o͡uéchó

a

chi̩che

o̩uṛé

DET

Father

want-PRS.3.SG

that.REL

1

wake.SUBJ

DET

six

o’clock.PL

‘The father wants me to wake up at six o’clock’

(ALAVAL, 20.1 SavièseF)

(7)

[desøˈneʃə]

[ʃa]

i

Paṛe

Wi

kyé

Me

Désónéchó

a

chi̩che

o̩uṛé

DET

Father

want-PRS.3.SG

that.REL

1

wake.SUBJ

DET

six

o’clock.PL

‘The father wants me to wake up at six o’clock’

(ALAVAL, 20.2 SavièseM)

36While the ALAVAL data come from just two speakers, which were gathered under careful elicitation conditions, these observations concur to a certain extent with the evidence presented above. However, the production data presented in the present study appear to suggest that apocope is now more likely than not. It may therefore be the case that a change in the form of the subjunctive is taking place, in that apocope is being regularised, which in turn could be posited to be linked to a decline in the subjunctive in spoken Francoprovençal in general. Previous work has already demonstrated that competition between subjunctive and indicative forms can be amplified by a reduced morphological system, leading to greater ambiguity (Jones 2000: 196). Further evidence from a larger sample size would be needed to make this claim more robust.

37With regard to the social factors coded for in the corpus, while there were small proportional differences by speaker sex and age, no statistically significant distributions were observed for subjunctive selection in the regression analysis (Table 9). The only trend to emerge from the data was that subjunctive selection now only appears to be confined to more monitored styles, as no subjunctive tokens emerged in casual speech (albeit with a total token count in this style of n=7). At the very least, then, the data present some evidence for style shifting at the morphosyntactic level. However, given that the study is under-powered, on balance it appears that the subjunctive is less a sociolinguistic variable than it is a linguistic variable in spoken Francoprovençal, a pattern observed in other variationist studies on the subjunctive in Romance (cf. King et al. 2018, Cerruti & Ballarè 2022).

8. Conclusion

38In introducing a sociolinguistic corpus of spoken Francoprovençal, this study has sought to (a) present quantitative evidence for the synchronic shape of the subjunctive, and (b) assess which social and linguistic factors might be constraining variant selection. Unlike similar studies on other minoritised Romance varieties (e.g. Jones 2000, 2014), the available evidence presented here suggest a potential decline in subjunctive usage in the community. While there is clearly variation in the corpus, social factors appear to play only a marginal role, and no tokens emerge in casual speech at all. In appealing to the internal-linguistic factors, it at least appears that a weakening of the subjunctive mood may well be linked to: (a) a loosening of tense matching, which has been implicated in increased variability in other contexts; (b) variation with the indicative mood; and (c) a breakdown in morphophonological regularity. An augmented study reproducing similar patterns of variation in tense, aspect, and mood selection, and coupled with the variable production of final vowels may therefore indicate an increasingly vulnerable subjunctive mood. However, a number of shortcomings have been highlighted in the research design that may be contributing to this picture, and further real-time evidence would be needed to confirm the observations made.

Appendix I. Example verbal tense paradigm

Table 1. Verbal tense paradigm for é̩tre ‘to be’ in Saviésan Francoprovençal*

Person

Pres. Indic.

Pres. Perfect

Imperfect

Cond.

Simple Fut.

Pres. Subjunctive

Imp. Subjunctive

1sg. (yó)**

chéi

chéi ita

īró

forôo

Chari

chiechó, chècho

foo͡uechó, foechéo

2sg. Tu

t’éi

t’éi ita

īré

Fori

charéi

chieché, chèeché

foo͡ueché, foeché

3sg. (i)

l’é

l’é ita

īré

Fori

Charé

chi, chèeché

foo͡ueché, foeché

1pl. Nó

Chin

chin ita

īron

Foran

Charin

chiechon, chèechon

foo͡uechon, foechon

2pl. vou

éi

éité ita

īré

Fori

charéi

chiéché, chèeché

foo͡ueché, foeché

3pl. (i)

Chon

chon ita

īron

Foran

Charan

chion, chèechon

foo͡uechon, foechon

* Language is marked using local orthographical conventions (Fondation Anne-Gabrielle & Nicola-V. Bretz-Héritier 2013).

** Francoprovençal is a partial pro-drop language, and brackets indicate optional pronouns

Appendix II. Translation task (subjunctive forms only)

Il faut que tu viennes avec moi

Il veut que je vienne avec lui

Je veux que tu viennes avec moi

A 7h le matin il faut que tu sois à la maison

Il faut qu'il débarrasse la table

Il faut que tu débarrasses la table

Il faut que tu chantes à haute voix

J'aimerais que tu viennes avec moi

A 7h le matin je veux que tu sois à la maison

A 7h le matin je veux que vous soyez à la maison

Pendant ce temps j'étais déçu et je voulais qu'il partît

Il faudrait que tu ailles avec lui

Il faut que je traverse le champ

Il faut que tu sonnes la cloche

Il faut que tu sois à la maison

Il faut que tu mettes le flon au four

II faut que je chante

Haut de page

Bibliographie

Auger, J., Hendrickson, R. (2021). Le morphème du subjonctif en picard du Vimeu : phonologie, variation et changement, in Dialectes de Wallonie en l’honneur de Fernand Carton. Liège : Société de langue et de littérature wallonnes.

Bates, D., Maechler, M., Bolker, B., & W., Steve. (2019). Package ‘lme4’. R package version 1.1–21. https://cran.r-project.org/web/packages/lme4/lme4.pdf.

Blainey, D. (2017). Sociolinguistic research with endangered varieties: the case of Louisiana French, in Journal of French Language Studies, 62(4), pp. 576-595, Cambridge: Cambridge University Press.

Carlier, A., De Mulder, W. & B. Lamiroy. (2012). Introduction. Folia Linguistica, 46(2): 287-302.

Cerruti, M. & S. Ballarè. (2022). Sociolinguistic variation, or lack thereof, in the use of the Italian subjunctive. Paper presented to the panel on Sociolinguistic variation in contemporary Italian, International Conference on Language Variation in Europe (ICLaVE) 11, Universität Wien, 14th April.

Comeau, P. (2011). A window on the past, a move toward the future: Sociolinguistic and formal perspectives on variation in Acadian French. PhD Dissertation. York University.

Comeau, P. (2019). When a linguistic variable doesn’t vary (much): The subjunctive mood in a conservative variety of Acadian French and its relevance to the actuation problem, in Journal of French Language Studies, 30(1), pp. 21-46, Cambridge: Cambridge University Press.

Diémoz, F., Kristol, A. (2018). Atlas linguistique audiovisuel du francoprovençal Valaisan ALAVAL. Neuchâtel: Université de Neuchâtel. Online : https://www.unine.ch/isla/home/recherche/projets-finalises/atlas-linguistique-audiovisuel.html; accessed 01/02/2022.

Digesto, S. (2021). Lexicalisation and social meaning of the Italian subjunctive. Cadernos de Linguística 2(3), e609.

Escher, L. (2022). Overlapping subjunctive forms in Gallo- and Ibero-Romance verb paradigms. Revue Romane 57(1): 86-115.

Favre, S. (2009). Qu’en est-il du subjonctif present? In: C. Fréchet (ed.), Langues et cultures de France et d’ailleurs. Lyon: Presses universitaires de Lyon, pp. 163-176.

Favre, C., Balet, Z. (1960). Lexique du parler de Savièse. Berne : A. Francke.

Fondation Anne-Gabrielle and Nicola-V. Bretz-Héritier. (2013). Lexique du parler de Savièse. Édition revue, augmentée et illustrée. Savièse: Éditions de la Chervignie.

Fournier, N. (1998). Grammaire du français classique. Paris : Belin.

Gadet, F. (1992). Le français populaire. Paris : PUF.

Hill, J. H. (1989). The social functions of relativization in obsolescent and non-obsolescent languages. In N. C. Dorian (ed.), Investigating Obsolescence: Studies in Language Contraction and Death. Cambridge: Cambridge University Press, pp. 149-166.

Jones, M. C. (2000). The subjunctive in Guernsey Norman French, in Journal of French Language Studies, 10(1), pp. 177-203, Cambridge: Cambridge University Press.

Jones, M. C. (2014). Variation and Change in Mainland and Insular Norman. Leiden: Brill.

Kasstan, J. R. (2015). Illustrations of the IPA: Lyonnais (Francoprovençal). Journal of the International Phonetic Association 45(3): 340-355.

Kasstan, J. R. & Nagy, N. (2018). Introduction. International Journal of the Sociology of Language, 249, 1–9.

Kasstan, J. R. (2019). Emergent sociolinguistic variation in severe language endangerment. Language in Society 48(5): 685-720.

Kasstan, J. R. (2020). Modelling stylistic variation in threatened and under-documented languages. Language Ecology 4(1): 73-94.

King, R. (2013). Acadian French in Time and Space: A Study in Morphosyntax and Comparative Sociolinguistics. Publication of the American Dialect Society 97. Durham, North Carolina; Duke University Press.

King, R., LeBlanc, C. L. and R. D. Grimm. (2018). Dialect contact and the Acadian French subjunctive: A cross-varietal study. Journal of Linguistic Geography 6(1): 4-19.

Klingler, T. A., LaFleur, A. (2007). L’enquête PFC en Louisiane, in S. Detey & D. Nouveau (eds.), Bulletin PFC N. 7, PFC: Enjeux descriptifs, théoriques et didactiques 2007, 331-340. Toulouse: Université de Toulouse II.

Labov W. (1972). Some principles of linguistic methodology. Language in Society 1(1): 97-120.

Labov, W. (1984). Field methods of the project on linguistic change and variation, in J. Baugh & J. Sherzer (eds.), Language in Use: Readings in Sociolinguistics, 28-53. New Jersey: Prentice-Hall.

Lamiroy, B. & W. De Mulder. (2011). Degrees of grammaticalization across languages. In B. Heine & H. Narrog (eds.), The Oxford Handbook of Grammaticalization. Oxford: Oxford University Press, pp. 302-337.

Lindqvist, C. (1979). L’emploi temporel dans la complétive au subjonctif par un temps du passé en français contemporain (Studia Romanica Upsaliensia, 28). Uppsala : Acta Universitatis Upsaliensis.

Luyet, B. (1929). Système phonétique du patois de Savièse. Publications de la « Komona Valéjana de Zénéva » 3, pp. 6-8.

Meyerhoff, M. (2017). Writing a linguistic symphony: Analyzing variation while doing language documentation, Canadian Journal of Linguistics, 62(4), pp. 525-549.

Nagy, N., Reynolds, B. (1997). Optimality Theory and variable word-final deletion in Faetar. Language Variation and Change 9(1): 37-55.

Nützel, D. (2009). The East Franconian Dialect of Haysville, Indiana: A Study in Language Death. Regensburg: Edition Vulpes.

Poplack, S. (1992). The inherent variability of the French subjunctive, in C. Laeufer, T. A. Morgan (eds.), Theoretical Analyses in Romance Linguistics: Selected Papers from the Nineteenth Linguistic Symposium on Romance Languages (LSRL XIX), The Ohio State University, 21–23 April 1989. Philadelphia: John Benjamins, pp. 235–263.

Poplack, S., Lealess, A. Dion, N. (2013). The evolving grammar of the French subjunctive, Probus, 25(1), pp. 139-195.

Poplack, S., Cacullos, R. T., Dion, N., de Andrade Berlinck, R., Digesto, S., Lacasse, D & Steuck, J. (2018). Variation and grammaticalization in Romance: A cross-linguistic study of the subjunctive, in W. Ayres-Bennett & J. Carruthers (eds.), Manuals in Linguistics: Romance Sociolinguistics. Berlin: de Gruyter, pp.217-252.

R Core Team. (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/.

Région autonome Vallé d’Aoste. 2022. Les modes et les temps des verbes. Available online: https://www.patoisvda.org/site/allegati/8-les-verbes-modes-et-temps-fra_634.pdf [accessed 01/02/2022].

Stich, D. (1998). Parlons francoprovençal : Une langue méconnue. Paris: L’Harmattan.

Stich, D. (2001). Francoprovençal : Proposition d’une orthographe supra-dialectale standardisée. Unpublished PhD dissertation. Université Paris V –Descartes.

Tsitsipis, L. D. (1984). Functional restriction and grammatical reduction in Albanian language in Greece. Zeitschrift für Balkanologie, 20: 288-308.

Zulato, A., Kasstan, J. R., Nagy, N. (2018). An overview of Francoprovençal vitality in Europe and North America, International Journal of the Sociology of Language, 249: pp. 11-29.

Haut de page

Notes

1 In an augmented study, Stich (2001: 261-262) further clarifies that patterns (3-4) can be typical of Valaisan varieties.

2 All instances of the target language are marked using local orthographical conventions.

3 For reference, a full verbal tense paradigm of the auxiliary verb é̩tre ‘to be’ is provided in Appendix i.

4 Participants in the corpus are pseudonymised and coded by region, sex, age group and speaker identifier (e.g. VaMO1 = Valais, Male, Older, speaker #1).

Haut de page

Table des illustrations

URL http://journals.openedition.org/linx/docannexe/image/8722/img-1.jpg
Fichier image/jpeg, 143k
Haut de page

Pour citer cet article

Référence électronique

Jonathan R. Kasstan, « A variationist analysis of the subjunctive mood in a sociolinguistic corpus of spoken Francoprovençal »Linx [En ligne], 84 | 2022, mis en ligne le 31 août 2022, consulté le 13 avril 2026. URL : http://journals.openedition.org/linx/8722 ; DOI : https://doi.org/10.4000/linx.8722

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search