Occitan is one of the largest minority language groups in Europe. It has been estimated that over three milion people can manage a conversation in the language and that over two million use it at home. Its territory is vast, covering a third of France. Given the size of this territory the problem of sampling and representation are vast. Faced with this situation the survey was undertaken in the region of Midi-Pyrénées (capital Toulouse, though this city and its immediate surroundings were excluded from the sampling area because of the very low observed levels of Occitan) with the intention of giving an adequate representation of language use by speakers in one specific area. However readers are warned that the results can in no justifiable way be generalised to the whole territory where Occitan is spoken.

The fieldwork was undertaken by a team recruited and supervised by the Université Toulouse-Le Mirail, in early 1996. The questionnaire was discussed with the supervisers, and shortened both in the range of situations in which data were requested, and in interviewees jobs, as this was regarded as a pointless exercise in view of the totally marginalised position of what is still widely known as the patois.

Though they were instructed to recruit active bilinguals, some of the interviewers had only a passive knowledge of Occitan, though in all cases their attitude was positive. The sampling sites were chosen to cover all the departments in the region. 281 subjects were interviewed.

Table 1 gives us an indication of the extent of language group endogamy and also an indication of the extent of ability, and inter-generational process or language production/ non-reproduction:

OCCITAN:

It is evident that even two generations ago the extent of French was widespread, and that by the generation of the parents of the respondents those who were not entirely proficient in French were few in number. On the other hand the data for Occitan suggests that there is a considerable extent of language group exogamy, even for the grandparents of the respondents with about 20% of the partners being involved in marriages where one or other did not speak Occitan. Furthermore, by the respondents generation more siblings spoke no Occitan than had some knowledge of the language. Since almost half of the respondents had not lived outside of the municipality of current residence and that fewer than 10% had lived outside of Occitanie, this is somewhat surprising. The partner also tended to be from within the region, with only 18% having lived outside the region during their life. Furthermore, only 39% of the respondents with partners claimed that their partner spoke Occitan 'very well', or 'quite well' while a further 28% spoke a little. The geographical mobility of the parents was also limited, with less than 15% of the mothers and 18% of the fathers having lived outside of Occitan for any period. That is, there is little relationship between language group endogamy and Occitan geographic endogamy.

OCCITAN:

FRENCH:

Table 2 gives clearer picture of competence. Among the respondents very few have other than a complete competence across all the dimensions in French. On the other hand a substantial proportion - 20%- of the respondents could not speak the language, and a further 28% spoke little Occitan. The figures for reading and writing suggest that the extent of literacy in Occitan is limited, with only 19% claiming and adequate writing ability and 41% a comparable ability with reference to reading. This is not surprising given that only 14% spoke Occitan as their first language and a further 31% spoke both French and Occitan as their first language. For the remainder either French or another language was the first language. Thus the sample consists of an extremely limited number who have a high degree of competence across the range of abilities.

Given what has been said above about the abilities of the different generations by reference to the respondents, the figures in table 3 are hardly surprising:

They indicate that more half of the respondents (55%) reported that their parents spoke to each other, when the respondent was a youngster, in Occitan or Occitan and French. However, fully 41% spoke only French with each other.

Table 4 throws more light on the preceding data. It indicates that the level of exclusivity in the use of Occitan in the family when the respondent was a child was extremely limited, and that the majority of respondents were raised in families where French was the sole language, though the use of Occitan with grandparents is sizeable. This indicates the limited extent to which the language has been subject to reproduction since, as we shall, see the extent of community use is equally limited.

We have already indicated that the ability of the partner of the respondent in Occitan is limited. for the 67% of the partners who had some ability in Occitan we find that 19% spoke only Occitan with their father while a further 18% spoke both Occitan and French with their father. The figures are similar for communication with the mother. On the other hand only 2% spoke Occitan only with their children while a further 13% spoke both languages with their children. The rate of inter-generational non-reproduction is significant. The figures for communication with other relatives lies between these two sets of figures.

Given the information in table 5 it is hardly surprising that children use so little of the language together (table 6). None used only Occitan and only 7% used any Occitan.

It already seems to be case that Occitan is, by and large, excluded from family life. This is reflected in table 7. Occitan on its own is used at meal time only by 5% of the families, and it should be remembered that this question spans generations. Similar percentage use Occitan exclusively with parents and partner but considerably less with the children. The same discrepancy is evident by reference to a mixture of Occitan and French but with less use of this mixture with the partner. Thus Occitan is used by 23% of respondents at meal time, mainly mixed with French. It appears in 27% of the cases with the mother and father, in 23% of the cases with the partner and in 18% of the cases with children. What we do not know is the nature of the relationship between French and Occitan in the cases where both languages are used.

TABLE 7:LANGUAGE OF THE HOME

The relevance of using Occitan for opening telephone conversations is extremely limited as is revealed in table 8.

TABLE 8: LANGUAGE USED TO OPEN TELEPHONE CONVERSATION

The contrast between what the respondents evaluate as the past use of Occitan and the current use presented in table 9 is quite startling. Whereas almost three quarters claimed they often heard Occitan in the streets in the past, this only applies to about a quarter of the respondents at present. The same difference applies to the other contexts, with the frequent use in shops reducing from 37% to 11%, in church from 18% to 5%, and in societies from 15% to 10%. As many as three quarters claim that they never hear Occitan in the Church, 62% claim they never hear Occitan in voluntary associations, and 35% claim they never hear it in shops. While Occitan is still heard in the range of contexts it is clear that there has been a profound change in use across all contexts within the memory of the respondents.

PAST:

PRESENT:

Current participation in Occitan cultural activities is limited (table 10). The most popular activity involves musical activity, something which does not insist upon a profound knowledge of the language. Fewer than 20% of the respondents claimed even a minimum of involvement in the other three activities. The number involved in such activities on a regular basis is small.

Table 11 gives an indication of the tendency for networks to be structured by reference to language and of the extent of overlap between ability and use among members of social networks, at least as perceived by the respondents. The very fact that over a third of the respondents are unable to assess the proportion of even friends who can speak Occitan gives a clear idea of the limited use of the language in such circles:

a. ABILITY:

b. USE:

It is evident that almost half of those who responded to this question perceive themselves as living in neighbourhoods where the majority are able to speak Occitan. Similarly almost a third of respondents are of the opinion that the majority of their friends speak Occitan, and about the same proportion who attend cultural activities do so within a context where the majority are perceived as being Occitan speakers. On the other hand, fewer than 20% of those who responded would claim that Occitan is used at least as much as French in such contexts. Over half of the respondents who answered the question claimed that they only used French in all of these contexts. By reference to perception of ability, the 'most Occitan' contexts is the neighbourhood, followed by friendship networks. This order suggests that French dominates social networks more than one would expect by reference to the incidence of Occitan ability.

Table 12 gives an indication of the contexts within which Occitan tends to be used. It clearly has no context within which it has exclusivity. Furthermore, it is only in a limited range of contexts - sports, fishing, politics, cafe and restaurant that it has very much of any relevance. The two contexts which do stand out as having a fairly high use of Occitan are the family and by reference to the environment. It is evident from these figures that French dominates the lives of most people within the area.

If the activities in which at least 40 respondents participate are ranked by the use of Occitan (on its own or together with French), it is clear that fishing and hunting (practised only by a minority), environmental issues, visiting friends and cafés and bars are the situations in which the presence of Occitan is significant. The low use of Occitan in the church is striking.

Table 13 gives a broader context to the ability to use Occitan and the willingness to do so. The extent of possibility appears to be extremely limited with only a few exceptions - the café, local authorities, shopping and religion. There also appears to be a greater difference between ability and willingness by reference to high-ranking functionaries and liberal professions - solicitor, doctor, dentist etc.- and some authority figures - library and police.

When the situations are ranked according to the possibility of using Occitan, Municipality, Speaking with café manager, Town hall, Speaking with priest and Shopping are those scoring highest (though the scores are quite low, perhaps reflecting the limited competenmce in Occitan of some respondents). Use is highest, when reported as possible, in Speaking to a local councillor and when Shopping, though in neither case is Occitan use reported by an especially high proportion of respondents. At the other extreme, Occitan appears hardly to be used by any respondents in the Tax Office and the Travel Agent: and in general, in dealings with most liberal professions.

The unavoidable conclusion is that the public use of Occitan is highly restricted.

The society appears to be fairly secular with only 11% of respondents claiming toattend church regularly and a further 25% sometimes:

Within the Church Occitan appears to lack any relevance, having far less of a place than Latin, with French inevitably being the language of religious activity.

Education is a key agency both by reference to supplementing the family role in terms of reproduction and by reference to its production function. It would appear that the availability of education through the medium of Occitan is extremely limited. The vast majority of the children were and are educated exclusively through the medium of French. Only 1% stated that there was a 'calandreta' school within reach of them and a further 3% said there was such a school 'fairly close' to them.

Bearing in mind the indications of the local team, the questionnaire was adjusted so as to make just a simple reference to the possible presence of Occitan in the education of the interviewees' children, and then analyse if necessary the answers to this open-ended question. (Q21). However, since only 21 interviewees -a very small number as expected- stated that their children had not been, or were not being, educated exclusively through the medium French, no further analysis was undertaken.

The survey thus confirms the marginal role that education plays in Occitan language reproduction.

The interviewing team felt, given the totally marginal presence of the Occitan language in all mass media, that far a sample of the size chosen, ne valuable conclusion would be reached by including items on cultural reproduction.

Fewer than a third of the respondents provided information concerning the language spoken by the director of their company, primarily because most of the others claimed to be self-employed or housewives. Of these the majority claimed that the directors spoke no Occitan, and of all respondents only 6% claimed that half or more of the directors spoke Occitan. On the other hand 42% supplied information pertaining to the language of their co-workers, with most of these claiming very little Occitan-speaking ability among their co-workers. It is therefore hardly surprising that very few spoke Occitan with their directors, co-workers, subordinates or even clients. Only 5% claimed that any Occitan was employed by the administration where they worked.

In conclusion, the role that Occitan seems to play in the world of work is at best marginal, where it isn't totally absent.

What stands out from the data on subjective identity in Table 16 is the overlapping identities that pertain to the majority of the respondents:

Clearly more respondents claim French identity than any other. For many, therefore, any other dimensions of identity are subordinate to that of the state. It is also interesting that more regard themselves as not being Occitan than those who do. That is, the idea of a relationship between the language and the space called Occitanie does not overlap. Cross tabulating the data from this question allows us to ascertain the degree of overlap between identities. As stated previously, the contradiction between the claim for French identity and either Occitan or Midi-Pyrénées is small, being 14% for the first case, that is those who claim Occitan identity but not French, and 17% for the latter. On the other hand, 38% claim both an Occitan and Midi-Pyrénées identity while 34% claim neither. The remaining 28% claim one or other but not both.

These attitude items do serve to give a picture of the perception of the language among the respondents. It does not seem that they see other languages as more important to any great degree, but neither do they feel that the language should enter public administration. On the other hand care should be taken with this second item which refers to 'privileging' the language rather than explicitly linking it to employment. The prestige element is more clearly expressed in item six where opinion seems to be split. There is some support for the idea that Occitan is unsuitable for science and 'the modern world'. The two items which seek to tap the relationship between status and language - items 5 and 11 are polarised. On the one hand the majority do see it as a sign of inferiority, but on the other hand, there is less support for the idea that the language is outdated. This contradiction is probably explained by reference to the outmoded nature of the language is expressed by reference to the perception of others 'La plupart de gens...' whereas item five makes direct reference to the respondent. There is no strong support for the idea that the language is disappearing which is surprising given the preceding discussion. At the same time there is no strong support for the need for the young to learn the language. This may well be linked to the fact that Occitan is not perceived as a defining characteristic of the Midi-Pyrénées region and the overriding French identity to the extent that this region is viewed as a region of France.

Finally table 18 indicates the opinion of the respondents by reference to the interest of different agencies and people in the Occitan language. A score of 5 indicates the highway position. As can be seen, only the respondent's own declared interest (5·98) and that of the respondent's family (5·08) come out on the positive side, and not very clearly at that. Indeed, given the small extent to which ability is translated into use by the respondents, even this must be treated with reservation.

Substantial proportions of "friends" were also positively evaluated, but even here there was considerable negative opinion. The perceived interest of regional, general and local councils varied, though it was only regard as great by a small number of respondents.Central government came out remarkably negatively, with as many as 69% stating that it is viewed as having the minimum of interest in Occitan. This perceived lack of interest was reiterated by reference to commercial activities with in-migrants, private enterprise and, especially, banks, all being viewed in negative terms by reference to interest in the language. The other negative view was the Church, which 72% of all interviewees viewed as displaying no interest whatsoever in the language.

The survey presents a very negative view of the Occitan language group in the region studied, by reference to its potential for continuation. The base from which it might reproduce itself is eroding quickly and the infrastructure necessary in order to produce the language is virtually non-existent. This situation is compounded by the negative identity vis-à-vis the language and the absence of any basis for increasing language prestige as a motivational source.

©Euromosaic