331
Views
44
CrossRef citations to date
0
Altmetric
Articles

Gendered Multilingualism in highland Daghestan: story of a loss

, &
Pages 115-132 | Received 10 Oct 2017, Accepted 20 Jun 2018, Published online: 10 Jul 2018

ABSTRACT

The paper traces the level of bilingualism in several highland villages of Daghestan (Northeast Caucasus) through the twentieth century. We show that historically, men were more multilingual than women, but this was not true to the same extent for all languages. Highlanders’ repertoires suggest a correlation between the social function of the second language and the degree to which its command was gendered. We also explore the dynamics of multilingualism from the generation born at the end of the nineteenth century to the generation born in the 1990s. We show that during the twentieth century local L2s were gradually displaced by Russian, and Daghestanian multilingualism lost its gendered character. We argue that these changes were caused by the introduction of Soviet schooling.

Introduction

  • It has often been observed that multilingualism is more typical of the male part of a community, while women tend to speak only their native language (Weinreich Citation1953, 94; Hill Citation1987; Holmes et al. Citation1996, 722; Kulick Citation1997; Herbert Citation2002; Hoffman Citation2006, 150; Takahashi Citation2012).

  • In the presence of a dominant language associated with an innovative urban society, women are sometimes considered guardians of the native language and culture, while men exhibit a stronger connection to the innovative culture and language(s) (Dabene and Moore Citation1995, 22; Piller and Pavlenko Citation2001; Hoffman Citation2006; Deumert Citation2010). In other situations, women were shown to rapidly acquire a second language and be more prone to language shift than men (Gal Citation1978; Aikio Citation1992; Garzon Citation1998; Trechter Citation2014; Çağlayan Citation2014, 133).

  • The behaviour of women with respect to acquiring a language with a higher overt prestige depends on the specific conditions of a particular language community. For example, the role of the main caregiver is associated with women in most societies, but a social interpretation of this role is culture-specific. A woman can be expected to transfer local culture and language to her children (Paugh Citation2001; Hoffman Citation2006), but she can alternatively be praised for raising her children to be well adapted to modern life and education by speaking the language of the urban culture (Cavanaugh Citation2006; Smith-Hefner Citation2009). As Hoffman puts it, «we simply cannot make a universalising statement about women as linguistic conservationists or innovators» (Hoffman Citation2006, 161).

  • Most of the papers cited above deal with situations where a shift occurs from a native language to the language of a dominant culture with higher prestige, usually urban and often more widely spoken. This paper considers another type of shift. In highland Daghestan (one of Russia’s autonomous republics located in the Northeast Caucasus), native languages remain largely intact (Dobrushina Citation2016). The shift concerns second languages. Based on data obtained in a large-scale sociolinguistic survey, we show how the residents of highland villages switched from communicating with their neighbours in one of the local languages to using one single lingua franca: Russian. We approach the situation from the perspective of gender and address the following questions. Was the traditional multilingualism of Daghestanian villagers gendered and, if yes, how were different second languages distributed among genders? How did the emergence of a new second language and its dissemination through education, social advancement and urban life affect their language repertoire?

  • First, we will show that, historically, multilingualism in Daghestan was gendered. The distribution of languages between genders was more complex than it is usually described in other studies. Due to the extreme linguistic diversity of Daghestan, the language repertoire of the residents of highland villages often consisted of three to six and more languages. We concur with the claim that men tend to be more multilingual than women, but, in Daghestan, this was not equally true for all second languages. Highlanders’ repertoires show a correlation between the social function of the second language on the one hand, and the degree to which it was gender-biased on the other.

  • Second, we investigate the dynamics of the multilingualism from the generation born at the end of the nineteenth century to the generation born in the 1990s. We discover several critical moments at which drastic changes took place in the multilingual repertoires of the villagers. With respect to the acquisition of a new dominant language (Russian), the change among men and women is rather similar. We show that both genders shifted from the traditional pattern of multilingualism to the new one almost simultaneously. The reason for this was that the vehicle for this change was school education, which was non-gendered and compulsory for all citizens of the Soviet Union.

  • Finally, we show that multilingual repertoires responded to this extralinguistic impetus almost immediately. In particular locations, languages were acquired and lost by the whole community within several decades, leading to a complete restructuring of its language repertoire.

  • In section Background, data and method, we provide some necessary background information about Daghestan, its languages and inhabitants, and introduce the method used to collect the data. The sections that follow first present a survey of one highland village in more detail, and then the general results for all 38 villages that have been surveyed so far. We summarise the findings in Conclusion.

Background, data and method

Next section provides a brief overview of the sociolinguistic situation and language density in Daghestan. The method which was used to collect the data is described in section Method and data.

Daghestan, its people and languages

The Republic of Daghestan is an area of high language density and diversity. Most of the languages spoken in Daghestan belong to the East Caucasian (Nakh-Daghestanian) family. There are also three Turkic (Kumyk, Nogai, Azerbaijani) and two Indo-European languages (Tat and Russian) spoken in the area (for an overview of languages see Wixman Citation1980, Tuite Citation1999, Koryakov Citation2002, van den Berg Citation2005). The estimated number of languages ranges from 30 to 45.

Daghestan mostly consists of mountainous terrain, and the lowlands near the Caspian Sea were uninhabited until relatively recently.

In contrast to its extreme diversity in terms of languages and situations of language contact, Daghestan is rather homogeneous in terms of culture and history (Wixman Citation1980, 107). The vast majority of Daghestanians are Muslims, and a command of classical Arabic was typical for some part of the population. Arabic was sporadically used for writing in local languages, but indigenous languages remained largely unwritten until the twentieth century.

The main occupation in highland villages was shepherding, and, to a certain extent, crop farming. Women traditionally cultivated fields, took care of the cattle, mowed the grass and made cheese. Shepherding was mainly a male occupation. In poor highland settlements people often practiced seasonal jobs outside the village, such as going for construction jobs, herding, tinning, or oil extraction in Azerbaijan. In some areas, it was common practice to leave the village and go tinning to other Daghestanian villages, to Azerbaijan or to Georgia. As the lowlands were more fertile and rich, seasonal occupations were less common among residents of these areas.

It is usually assumed that language boundaries were maintained not only by the landscape, but also by traditional ethnic and often village-level endogamy. In most Daghestanian villages, partners were taken exclusively from the same family (tukhum – Karpov Citation2010). A corollary of the prohibition on marrying out was that linguistically mixed marriages were uncommon (Comrie Citation2008, Karafet et al. Citation2016). Even now, villages with 300-400 households can have as few as one or two mixed families. This provides for the vitality of the languages in the villages, where even now children still communicate in the village’s native language. The population of highland villages ranges from one hundred to several thousand residents.

With so many languages packed in a relatively small area, it is not infrequent that in two neighbouring villages located at walking distance from one another, different languages are spoken. Economic and social ties between neighbouring villages have always been strong. Interethnic communication requires knowledge of a shared language. There was never a lingua franca common to all of Daghestan (Chirikba Citation2008, 30). Until the mid-nineteenth century, the lowlands were dominated by speakers of Turkic languages, which lead to the use of Kumyk and Azerbaijani as lingua francas in the adjacent territories (Wixman Citation1980, 108–119). The local language Avar was also used as a lingua franca in some parts of northern Daghestan. Most Daghestanian villagers also spoke the language of a neighbouring village(s) and additionally a distant language besides their native language.

The distinction between neighbouring and distant languages is crucial for this paper. As we will show later, the command of neighbouring languages was far less gendered than the command of distant languages (Dobrushina Citation2011).

A language is defined as neighbouring if it is spoken in a neighbouring village. A pair of villages are neighbouring if they have a common territorial border or are separated one from the other by the territory of no more than one other village. This usually means a 30 minutes to 2 hours walking distance from one village to the other. Neighbouring languages were acquired in the course of everyday economic activities, such as exchanging goods, cultivating adjacent fields, or shepherding.

Distant languages were spoken beyond the area of neighbouring villages. They were acquired in the course of seasonal jobs outside the village.

Another major language of the area is Russian. Historically, there were no Russian settlements in Daghestan. Russian was acquired through contacts with Russian administration since Daghestan became part of the Russian Empire in the beginning of the nineteenth century. Nonetheless, the command of Russian was very low before the establishment of the Soviet administration in the 1920s. In the mid-1930s, Soviet schools were opened in most Daghestanian villages (for a brief history of language policy in the Soviet Union see Ulasiuk Citation2011). Russian was one of the main subjects in school, and the official language of instruction in secondary school (from 9 to 17 years old). Russian teachers were sent to Daghestanian villages by the Soviet government.

Russian was the first language to be established as a single lingua franca and as a language of written administration common for the whole area (Daniel et al. Citation2010; Magomedov Citation2010; Dobrushina Citation2016). Although the vitality of local languages has not yet been impaired in the villages, neighbour multilingualism is in obvious decline (similar to what was described for Senegal in Lüpke Citation2010). Most people under 20, or, in some villages, under 40, use Russian when they communicate with their neighbours from other villages.

Method and data

Research on traditional multilingualism in Daghestan requires data on local multilingual patterns from the period before the spread of Russian; we need a window into the sociolinguistic past of the region. The method of retrospective family interviews was specifically designed to obtain quantitative data about multilingualism in the past (Dobrushina Citation2013). The method entails that respondents are interviewed about their own language inventories as well as those of their elder – often deceased – relatives. Only those relatives with whom the respondent overlapped in terms of lifespan, and whom they claim to remember clearly, are added to the database.

These interviews allow us to reach back into the nineteenth century, starting from residents born around 1850, with more dense data from the 1880s onwards. This time span captures the situation typical of the village before the drastic social changes of the twentieth century.

The interviews were all held in Russian.

The method is vulnerable to both individuals’ mistakes and systematic biases. The multilingual situation is often stereotyped, generalised and extended to the interviewee’s relatives (‘Our elders spoke this language, so my parents did, too’).

Another issue is the evaluation of the command of a language. A variety of approaches to the evaluation of language proficiency have been developed in the study of language shifts (see Vakhtin Citation2001); such studies are very time consuming and require the researcher’s command of the language in which the proficiency is evaluated. For our cross-Daghestanian study this was not a viable option, due to the number of languages involved, and because we do not just survey language inventories of inhabitants, but also try to reconstruct those of their relatives who are currently deceased. As a result, we had to rely on the respondent’s self-assessment and his or her assessment of the recollected multilingualism of elder relatives.

Finally, for the eldest relatives (e.g. those born in 1880), only multilingualism at a later age could be reported. Although people acquire their basic language repertoire in their youth, their linguistic inventory may gradually change over the course of their lives, especially if the community undergoes considerable social changes (Chambers Citation2002, 358). As in any kind of apparent time study – which this project to a certain extent is, the effect of life-span change distorts the true dynamics of trends of change.

All this results in distortions of the targeted sociolinguistic reality in our analysis. The only bias that we expect to be truly systematic, however, is the present community’s stereotypical image of the past multilingualism. All other vulnerabilities, although they decrease the reliability of individual records, probably go both ways and should not result in systematic deformations. Therefore we believe that the collected data, biased as they are, provide empirical evidence beyond social stereotypes.

The data were collected in 2013–2016 (see footnote 2). At the time when this paper was written, 38 villages had been surveyedFootnote1 (see ). The villages form 12 geographic clusters of two to four villages with two to three different native languages per cluster.

Figure 1. The clusters investigated so far (map courtesy Yu. Koryakov).

Figure 1. The clusters investigated so far (map courtesy Yu. Koryakov).

The collected data on multilingual repertoires are fairly balanced in terms of gender. So far, the database contains information about 1899 women and 1984 men, with birth years starting from the middle of the nineteenth century.

We will combine two approaches in order to report the empirical results of our study. First, we will present a case study of one village. This enables us to illuminate the social conditions under which languages were acquired and lost. Second, we will take a more large-scale approach and describe the dynamics of change in the whole of Daghestan as it is reflected in the data from 38 villages.

Gendered patterns of multilingualism in highland villages: a case study of Chuni

In this section, we will describe the multilingual repertoire of the residents of one highland village – Chuni. The history of multilingualism in Chuni reflects the history of multilingualism in Daghestan in general, and this case study can therefore serve as a transition to the large-scale history of sex-determined multilingualism in 38 highland villages.

The aim of this section is to show which activities have triggered the multilingualism of Chuni’s inhabitants, how the command of different languages was distributed among women and men, and what has induced the changes in their multilingual repertoires during the twentieth century.

The case study starts with a brief description of the village. Then, we will consider three distinct patterns of multilingualism: the command of distant languages, Russian, and neighbouring languages.

Chuni

The inhabitants of Chuni speak a dialectal variety of Avar, one of the major languages of Daghestan (about 5,00,000 speakers). In their own area, however, they are a minority, as the village is situated midway between two Dargwa villages – Upper Ubeki (Verkhnie Ubeki) and Tsukhta. Residents of Upper Ubeki speak the Tsudakhar dialect of Dargwa, while residents of Tsukhta speak the Akusha dialect of Dargwa. The distance from Chuni to Tsukhta is about 30 minutes by foot, the distance from Chuni to Ubeki – about 90 minutes (see ).

Chuni has about 1000 residents (884 in 2009 according to official data, 1100 in 2014 according to residents’ estimations). In 1895, there were 405 residents (Kozubskij Citation1895).

Marriages were and to some extent still are predominantly endogamous. There was no strict prohibition for women to marry out or for men to bring wives from elsewhere. The main source of partners from outside was the neighbouring Akusha village Tsukhta. Nonetheless, cases of intermarriage were infrequent.

Chuni lies about 1300 metres above sea level. There are many water springs, and according to local people, the land is more suitable for agriculture than for cattle breeding. Local products were sold or exchanged in the numerous marketplaces in the neighbouring Dargwa speaking villages. Chuni is situated on the way from lower lands to the Aya-bazaar and Tsudakhar markets. Travelling traders used to stay in Chuni overnight or at least for several hours.

During Soviet times, Chuni was part of a big sovkhoz, ‘Kavkaz’, consolidating Chuni and four Akusha villages (Ebdalaya, Tsukhta, Uluaya, Naskent). It meant close communication with Akusha Dargwa speaking people during common agricultural activities, and some shared institutions, including the legal court.

Distant languages were acquired mostly outside the village. Cattle breeding led to temporary migrations and distant sheepherding, in particular to the lowlands near Manas (Kakashura), where Kumyk is spoken, and near Izberbash. This is how Chuni men acquired a command of Kumyk.

There was a period, mainly in the 1950s, when people from Chuni had a custom of spending several years or even decades working in Stavropol and Kalmykia, where many people speak Russian as an L1. The migration to these places was so common in Chuni, that there is hardly a single person in our database born after the 1920s, whose life was not in some way connected to Stavropol or Kalmykia. According to some local opinions, cattle breeding in these regions was profitable at first, but lost its economic attraction later on.

Living in Stavropol and Kalmykia presumably enhanced the command of Russian. Notably, there are almost no reports of Chuni people speaking Kalmyk, which means that they communicated with speakers of Kalmyk in Russian.

In the 1960s, Chuni villagers started to cultivate and sell cabbage. The location of the village proved to be good for growing it, and cabbage became the main source of residents’ income. As of 2014, cabbage was still the main trade.

The first Soviet school was launched in Chuni in 1928, which is several years earlier than in most other villages. In the beginning, it had no building of its own. Lessons were given in private houses. In 1933, a primary school building was constructed. At that time, primary school entailed four years of education and started at the age of seven. Schooling was first extended to seven years after 1951, and then to eight years in 1967. Сhildren who wanted to study more than seven or eight years continued their education in other villages (Tsukhta, Naskent, Levashi, Gergebil’). Since the 1990s, a full 11-year school functions in Chuni.

From the very start of the school, Chuni children had a lesson called ‘mother tongue’ (Russian ‘rodnoj jazyk’) where standard Avar was taught. Avar was the language of instruction at primary school, and Russian at secondary school. As in all of Daghestan, until the 1960s Russian teachers worked at the school in Chuni alongside local teachers. Several years ago, the language of primary education was changed to Russian in most schools. Chuni people use both Avar and Russian for writing.

In sum, the command of six languages is mentioned in the interviews with Chuni residents: their native language Avar (which is spoken by all residents without exception, and will therefore not be discussed further in this paper); the languages of the neighbouring villages Akusha Dargwa and Tsudakhar Dargwa; distant languages, which were not spoken in the vicinity – Kumyk; Russian; and Arabic, or, more often, only Arabic script (see ).

Table 1. Languages of Chuni people.

Distant language in Chuni

Chuni residents regularly mention the command of one distant language among people from the older generation – Kumyk. There are no Kumyk villages in the neighbourhood. The command of Kumyk was usually acquired when the cattle were driven to the lowlands near the Kumyk villages (which were reachable by foot in several days). The cattle spent several months there, and shepherds stayed with the cattle for the whole period. Herding was exclusively a male occupation; women did not take part in it. In the rare cases when a woman was reported to have a command of Kumyk, this was usually because she had some Kumyk-speaking relatives.

Some other distant languages such as Kalmyk, Lak, Chechen and others are also mentioned, although infrequently.

The Chuni subsample consists of 216 people born in the period from 1850 to 2003. There are 115 men in the subsample and 101 women. shows the dynamics of the command of Kumyk in Chuni among men. No women are reported to speak Kumyk in our database. The command of Kumyk was thus a gendered practice.

Figure 2. Knowledge of Kumyk among men in Chuni for the whole time period (percent of all men).Footnote3

Figure 2. Knowledge of Kumyk among men in Chuni for the whole time period (percent of all men).Footnote3

The command of Kumyk underwent drastic changes. The number of men who spoke Kumyk dropped from 40% among the generation of men born in 1910s to 20% among those born between 1940 and 1949.

By the 1960s years of birth, the pattern of distant multilingualism had become almost extinct. Only two men born after 1950 were reported to have some command of Kumyk (from a total of 75 about whom we have information).

The loss of Kumyk might have been caused by two factors. The first factor concerns changes in farming practice: since the 1960s, Chuni people switched from sheep breeding to growing cabbage, which was more profitable, so their contacts with Kumyk lands weakened. The second factor, which is relevant for all of Daghestan, was the spread of Russian as a lingua franca.

Acquisition of Russian in Chuni

Before the 1930s, Russian was acquired as a distant language: about 60% of men born between 1910 and 1929 could speak Russian, while the command of Russian among women was much less frequent. In the biographies of those men who spoke Russian, military service is usually mentioned, and participation in the Second World War. Their biography may also include having an important position in the village administration (for example, being the head of the kolkhoz or of the local party section), which required communication with Russians. More drastic changes started among the generation born between 1930 and 1939. Both genders demonstrate a dramatic increase of competence in Russian (with up to 80% of women speaking Russian and 90% of men – ). This change correlates with the launch of a local school in 1928. The school was established by the Soviet authorities, and Russian was the main subject there. People born in the 1930s went to school almost without exception (4 grades at the beginning, and 7 grades starting from 1951, which meant studying from 7 to 10 or 13-years-old). Starting from the generation born in the 1930s, male and female patterns of the command of Russian converged almost completely (see ). At present, Russian is spoken by close to a hundred percent of the population.

Figure 3. Knowledge of Russian among men and women in Chuni for the whole time period.

Figure 3. Knowledge of Russian among men and women in Chuni for the whole time period.

The emergence of Russian as a second language influenced the command of other second languages. As we will soon see for the neighbouring languages, Russian displaced the knowledge of other languages and assumed the role of a single lingua franca for Daghestan.

Neighbour multilingualism

As mentioned above, Chuni is situated between two Dargwa villages, in one of which the Akusha dialect is spoken, while in the second the Tsudakhar dialect is spoken. The native language of Chuni is Avar, which is related to the neighbouring Dargwa varieties only very distantly. While their Dargwa neighbours never spoke Avar, the residents of Chuni used to speak both Akusha and Tsudakhar Dargwa. In the generations born between 1910 and 1939, there was almost no difference between genders in respect of their ability to speak neighbouring languages. According to our survey, 90% of women and 100% of men spoke two neighbouring languages.

In the generation born between the 1950 and 1959, the ratio of people who could speak at least one neighbouring language dropped by 65% among men and by 50% among women (). In the next decade (1960–1969), only 20% of women and 40% of men could speak some language of the neighbours. At the present time, people born after 1980 communicate with their neighbours in Russian, irrespective of their gender.

Figure 4. Knowledge of Tsudakhar and (or) Akusha among men and women in Chuni over the whole period.

Figure 4. Knowledge of Tsudakhar and (or) Akusha among men and women in Chuni over the whole period.

Summary

To sum up, in Chuni the command of distant languages like Kumyk and Russian was most gendered for the generations born before 1929: mainly men could speak these languages. Starting from the generation born between 1930 and 1939, Russian spread among all residents of Chuni, irrespective of gender.

Among the people born after the 1970s, gendered patterns of multilingualism no longer exist. All Chuni people speak two languages: Russian and their native language Avar (see ).

Table 2. Gendered patterns of multilingualism in Chuni across generations.

The fact that the command of Kumyk and Russian among earlier generations was sex-determined, can be explained by the gendered patterns of occupation. Women did not take part in herding, hence they did not acquire Kumyk. Women did not travel outside their own village in order to get an education or a job, nor were they called up for military service – hence they did not acquire Russian.

Does the loss of gendered patterns of multilingualism signal that the patterns of occupation became less gendered? The answer to this question is no. It rather signals the emergence of a new, non-gendered practice. Compulsory school education, imposed by the Soviet authorities in the 1930s, was equally applied to girls and boys (unlike some other parts of the world, e.g. Nforbi, Citation2013). This practice was not accepted by Daghestanian people immediately, and for several years there were attempts to leave girls at home or to take them out of school at an earlier age than boys, but equal education for girls and boys was eventually adopted (Dobrushina Citation2008). At school, Russian was acquired by females and males to the same extent.

There is, however, one significant difference concerning the changing language repertoires of men and women. The level of command of neighbouring languages was never gendered, but women have lost them more abruptly than men. Akusha Dargwa dropped from 100% among women born in 1940s to 40% among those born in the 1950s, while the command of Tsudakhar Dargwa dropped from 90% to 10%. Men show a smoother change (from 100% to ∼60% for both languages).

To conclude, the emergence of a non-gendered lingua franca played a key role in the loss of the command of neighbouring languages and gendered patterns of multilingualism in Chuni.

So far, we analysed gender patterns of multilingualism in one highland village. The quantitative data for one village, however, are not very rich. For example, there are only seven males in the 1910s, which is not statistically representative. In the next section, we will consider gendered patterns of multilingualism in 38 villages, and will see that the cross-Daghestanian patterns of multilingualism are very similar to those of Chuni.

Gendered patterns of multilingualism in rural Daghestan

In this section, we analyse gendered patterns of multilingualism for 38 villages from various parts of Daghestan (see the map in ). The set of languages, as well as the particular circumstances of language contact are different in each cluster. We will not be able to go into details and explain the particular factors which underlie multilingualism in each village, as we did for Chuni.

The aim of this section is to show the dynamics of change in multilingual patterns that are common for the whole area. Traditional social structures were roughly similar across rural Daghestan, and the most important historical events which triggered the changes (such as participation in the Second World War and the start of Soviet schooling) had an equal impact in all of Daghestan.

We will start with a description of the dataset. Then, we will turn to distant multilingualism, followed by the dynamics of bilingualism in Russian and the dynamics of neighbouring multilingualism.

Dataset description and the overall tendency

The dataset includes field data from 38 Daghestanian villages. It consists of 20,512 observations and 13 parameters (https://github.com/LingConLab/2018_suplementary_for_Gendered_multilingualism_in_highland_Daghestan_story_of_a_loss). Each line of the dataset corresponds to the knowledge of one particular non-native language by one speaker. In order to visualise the difference between the multilingual repertoires of males and females, we created two plots based on all observations from our dataset (). Each dot represents a speaker, the left plot is for female speakers, the right one is for male speakers. The horizontal axis shows the year of birth, the vertical axis gives the number of languages known to the individual (from zero to five, the native language is not included). As we can see from these plots, females and males differed in their level of multilingualism. Women knew less languages, and there were more women who didn’t know any language but the native language.

Figure 5. The number of languages spoken by men and women in highland Daghestan.

Figure 5. The number of languages spoken by men and women in highland Daghestan.

Distant multilingualism

As explained above, in certain villages people had some command of one or several languages which were not spoken in their own neighbourhood. Distant languages were most often acquired when people travelled outside their area to earn some money or to purchase something (the road could take from ∼5 up to ∼30 hours by foot or by horse).

In terms of the number of languages, distant multilingualism was expectedly richer than neighbour multilingualism. There are no villages with more than two neighbouring languages in our database, but distant languages can be more numerous. In general, about 12% of the population of Daghestan born before 1910 spoke two or more distant languages.

As our data clearly demonstrated, female and male behaviour differs in respect of distant multilingualism.

represents female and male language competence in distant languages. On the vertical axis, the number of spoken languages is shown. The horizontal axis shows the year of birth. One can see that in 1890-1909, about 25% of men spoke one distant language and about 15% spoke two or three distant languages. Within the same period, only 10% of women spoke one distant language and less than 5% spoke two distant languages. After 1920, the number of men who were able to speak distant languages started steadily decreasing. By the 1950s, the difference between females and males had almost disappeared.

Figure 6. Knowledge of distant languages in Daghestan for the whole time period.

Figure 6. Knowledge of distant languages in Daghestan for the whole time period.

Command of distant languages is very rare among villagers born in the 1980s or later. The most probable reason for the loss of distant languages is the emergence of Russian as a lingua franca.

Acquisition of Russian

The role of Russian has changed significantly during the twentieth century. At the beginning of the century, competence in Russian was similar to that of distant languages. There were no sources of Russian inside the villages – there were no speakers of Russian within the region, no Russian schools and no media. Some villages had madrasas (religious schools) where people were taught to read the Quran. Only several villages had Russian schools that existed since the mid-nineteenth century. In general, Russian was acquired only through communication with Russian speaking people, but since there were no Russian settlements in the Daghestanian highlands, the language was acquired outside of the neighbourhood.

After the establishment of the Soviet Union, the role of Russian changed. Due to the national programme of compulsory education launched by the Soviet authorities in the 1930s, a school was opened in almost every highland village. Russian was one of the main subjects, the main language of instruction, and the native language of some teachers who were especially sent to these regions by the government. Since that time, Russian has gained the prestige of being the language of education, power and cities, and it has quickly spread across Daghestan.

This change of status is clearly reflected in the distribution of Russian bilingualism among men and women.

In the earliest period (i.e. among the generation born between 1890 and 1909), about 50% of the male residents spoke Russian, while only 10% of females were able to speak it (). This distribution strongly resembles that of distant languages at the time (see section Distant multilingualism). Until the birthyear 1919, the ratio of women who could speak Russian remained approximately the same, while the ratio of men has been constantly rising. The quick changes in the level of command of Russian among males within this period might be due to the fact that men born between 1900 and 1919 took part in the Second World War and acquired Russian during military service. At least, this is the most frequent rationale which was given by villagers in discussing the command of Russian by their elder male relatives.

Figure 7. Knowledge of Russian among men and women in Daghestan for the whole time period.

Figure 7. Knowledge of Russian among men and women in Daghestan for the whole time period.

The speed with which women acquired Russian is astonishing: starting from the birthyear 1920, the number of women who could speak Russian has risen up to almost 40%, and by the 1940s it had jumped up to 70%. This change clearly correlates with the launch of Soviet schools: most people born after 1920 obtained some school education, which led to a certain level of competence in Russian. In the 1940s, in most parts of Daghestan school education changed from 4 years to 7 years.

The acquisition of Russian correlates with a sharp decline in the command of distant languages. The rise of Russian among men started in the generation born in the 1910s – and the generation born in 1920s started to lose command of distant languages (cf. ). Since Russian was acquired by all ethnic groups in Daghestan, it was used for interethnic communication within the republic. This way, Russian has quickly displaced both distant and neighbouring second languages, and achieved an exclusive role of lingua franca for the whole territory of Daghestan.

Figure 8. Knowledge of Russian and distant languages among men in Daghestan for the whole time period.

Figure 8. Knowledge of Russian and distant languages among men in Daghestan for the whole time period.

Since the beginning of the 1950s, there has been almost no difference between men and women in respect of their command of Russian: near to a 100% of the population spoke Russian, regardless of their sex.

Neighbour multilingualism

Most villages which were surveyed in this study lie on the border of two or three language areas (). In most villages, people spoke the languages of their neighbours. The level of bilingualism, however, varied significantly, because bilingualism was often asymmetrical (unilateral). Although the average level of neighbour bilingualism according to our data is less than 50% (in terms of the number of bilingual people), the actual level varied among villages from 98% to 0%.

Neighbour multilingualism differs from distant multilingualism in that females were also engaged in it. shows the level of bilingualism in neighbouring languages among men and women. Vertical axis represents the number of spoken languages, the horizontal axis shows the year of birth. Men who could speak the language of their neighbours in the generations born between 1890 and 1909 were more numerous (about 70% spoke one or two neighbouring languages), but women also had a decent level of command of neighbouring languages (more than 50% could speak one or two neighbouring languages – ).

Figure 9. Knowledge of neighbouring languages in Daghestan among men and women for the whole time period.

Figure 9. Knowledge of neighbouring languages in Daghestan among men and women for the whole time period.

Multilingualism in neighbouring languages was more stable than the command of distant languages. The command of neighbouring languages was preserved almost intact until the 1950s. Since that generation, gendered patterns of multilingualism in neighbouring languages started to become lost.

Among later generations, the ratio of neighbour multilingualism slowly continues to decline. Almost none of the respondents born between 1970 and 1999 speak two neighbouring languages, about 25% speak one neighbouring language.

The dynamics of multilingualism in at least one neighbouring language is presented in . By comparing the generations born in the 1940s and 1950s, we can see that the tendency of women to be ahead of men in terms of losing command of the neighbouring languages (observed in Chuni) is also visible in the data from 38 villages, though it is less prominent. Since the 1970s, however, this has changed. As shown in , during the last decades women spoke neighbouring languages slightly more often than men. This could be due to the increased mobility of younger men. After the break-up of the USSR, unemployment has risen, and men often go to other parts of Russia for several months to earn some money, while women stay in the villages.Footnote2 Neighbouring languages thus became more useful for women than for men.

Figure 10. Knowledge of at least one neighbouring language in Daghestan among men and women for the whole time period.

Figure 10. Knowledge of at least one neighbouring language in Daghestan among men and women for the whole time period.

Summary

In this section, we considered the patterns of multilingualism in Daghestan with regards to the gender of speakers. We showed that within the earlier period (among residents born before the 1920s), gendered patterns of multilingualism were very prominent for the command of distant languages and Russian, while the command of neighbouring languages was spread more evenly. In all three cases, men were more multilingual than women.

The number of men who spoke distant languages was three times higher than the number of women (∼40-45% vs. ∼15%). Gendered patterns in distant multilingualism were lost together with the very phenomenon of distant multilingualism. Since the 1950s the command of distant languages has almost completely vanished. This strongly correlates with the rise of Russian.

Among those born before the 1900s, the command of Russian was four times higher among men than among women (∼40% vs. ∼10%). Since the 1900s the number of men who spoke Russian started to increase. Most probably, the rise of Russian among men was due to their participation in the Second World War. The acquisition of Russian among women started later on (in the 1920s), and was presumably evoked by the opening of Soviet schools in 1930s. Most people born between 1950 and 1959 acquired Russian and at present, bilingualism in Daghestan has almost lost its gendered nature.

The command of neighbouring languages was always more typical of men than of women (70% for men and 55% for women in the generation born between 1900 and 1909), but was never as gendered as the command of distant languages or Russian. Neighbouring languages were more viable than distant ones. The visible decrease started only from the 1950s (in terms of birth years). There was a slight difference between men and women in the rate of loss of neighbouring languages. Women lost the command of neighbouring languages more quickly than men, a tendency which we observed even more clearly in Chuni. This tendency has changed starting from the 1970s years of birth, when women became bilingual in neighbouring languages more often than men. We explained this by the increased mobility of men.

Conclusion

The structure of multilingualism in a language community reflects the structure of its external contacts. Second languages are acquired in the course of speakers’ economic, educational, or professional activities that involve contact with other language communities. A simple expectation is that bilingualism will be gendered if the activities which demand the language competence are gendered. In societies where women’s and men’s roles are significantly different, multilingualism is observed to be gendered (Aikio Citation1992: 58-59).

This is the pattern we find in Daghestan before the 1920s. Only men were involved in seasonal jobs which required travelling outside the village. As a consequence, distant languages (spoken by communities not present in the neighbourhood) were spoken only by men. On the other hand, the whole community was involved in communication with their neighbours. Proficiency in neighbouring languages was therefore gendered to a much lesser extent. Our conclusion is that gender is a factor of multilingualism not on its own but in combination with the language’s functional scope.

The changes in the language repertoire of the Daghestanian people over the twentieth century lead to the loss of gendered multilingualism. At first, the changes were visible only among the men. In the generation born in the 1920s, the level of Russian among men started to rise, but remained the same among women. This had a functional reason; men served in the Russian army during the Second World War. Further on, however, the women caught up with the men. The command of Russian quickly spread among the whole population of Daghestan, irrespective of sex, and eventually displaced other second languages including neighbouring languages. The traditional gendered multilingualism was replaced by non-gendered bilingualism with Russian as L2.

This change came as a response to the implementation of a new, non-gendered practice – compulsory Soviet education with Russian as one of its main subjects. The rapid spread of Russian reorganised the whole system of traditional multilingualism. Not only has Russian brought down the barriers between the many different ethnic groups of Daghestan by becoming the common lingua franca, but it has also levelled the previously gendered bilingualism patterns. In modern Daghestan, there is no difference between women and men in terms of their language repertoire. Gender equality came at a price, as traditional multilingual patterns were lost completely, and the use of local languages for local communication has waned.

The shift from the usage of neighbouring languages to a global lingua franca is observed in other parts of the world as well. As was shown in Hicks Citation2017, reciprocal multingualism on the Solomon Islands has quickly changed to bilingualism in Pijin. Learning other languages became less valuable and necessary (Hicks Citation2017: 869).

Among other things, our data demonstrate how direct the link is between the command of second languages and social practices. The change of the multilingual landscape under the influence of education was strong and abrupt. Within several decades, the whole system of traditional multilingualism was gone and restructured under the influence of the Soviet school system. The patterns of multilingualism in Daghestan, as we have shown in this paper, were closely linked to economic and social conditions, so that the change in language proficiency brought about by changes in these conditions was dramatic.

So far, the observed gender biases in linguistic repertoires could be explained on purely functional grounds. However, there seems to be a minor difference between Daghestanian men and women in the rate of loss of neighbouring languages for which we have no functional explanation. According to our counts, women were losing command of neighbouring languages more quickly than men. This may be a manifestation of a known tendency that women tend to be quicker on their way to innovations. The tendency is not very strong, however, and was not confirmed qualitatively in our interviews with the highlanders. There is also little chance that it may be confirmed in the future, because the generation in question is out of the interviewers’ reach. Therefore we leave this issue open.

Acknowledgements

This paper was prepared within the framework of the Basic Research Program at the National Research University Higher School of Economics (HSE). The first author also benefited from a EURIAS fellowship at the Collegium of Lyon, France, co-funded by Marie Sklodowska-Curie Actions under the European Union’s Seventh Framework for research. The authors are especially grateful to Mikhail Daniel, Samira Verhees, and three anonymous reviewers for their valuable comments on the earlier drafts of this article. We also express our deep gratitude to all our respondents in Daghestan.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported within the framework of a subsidy by the Russian Academic Excellence Project ‘5-100’.

Notes

1 The data were collected as part of the project «Atlas of multilingualism of Daghestan» (http://multidagestan.com/).

2 I express my gratitude to the anonymous reviewer who attracted my attention to this fact.

3 Plots and diagrams were plotted using ggplot2 package (Wickham Citation2009) for R 3.4.4 (R Core Team Citation2018).

References

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.