What is the term for a variation of speech and language based on geographic area?

Abstract

Dialectology, the systematic study of language variation, had its effective beginnings in Europe after 1850, but it has its roots in an age-old fascination with accent and dialect. Dialectology first evolved as a classificatory discipline that concentrated mainly on the speech patterns of the most conservative speakers, called NORMs, an acronym for nonmobile, older, rural males. Its consummation came with the Atlas linguistique de France (10 Vols., 1901–13) directed by Jules Gilliéron at the Sorbonne, which provided an influential precedent for 75 years. The rise of linguistics after 1916 brought with it initiatives for integrating dialectology, but these were mostly resisted. As social changes in the second half of the twentieth century diminished the rural, uneducated, isolated populaces that had been the dialectologists‘ object of study, the discipline declined. However, it was revitalized by two concurrent forces, one technological and the other methodological. The first is computerization, which provides dialectologists with the means of managing the superfluity of data that had been both its strength and its burden. The second is sociolinguistics, the study of language variation on social-science principles, which defines its domain more inclusively in terms of populations and social settings. Dialectology, thus reformed, has again become a thriving intellectual enterprise.

Social Space: The Emergence of Sociolinguistics

The social sciences were subject to a quantitative and objective turn during the 1960s. This also affected the development within research on language. Variationist sociolinguistics, or as it was initially dubbed, “urban dialectology,” grew out of dialectology at this time. Variationist sociolinguists do not focus solely on the geographic distribution of linguistic features as traditional dialectologists do. Instead, they study linguistic variation and change from a quantitative perspective and relate it to speakers' memberships in social categories such as gender, age, ethnicity, and social class. In other words, variationist sociolinguists study how social stratification is related to language use.

The founder of variationist sociolinguistics, William Labov, conducted his initial studies of linguistic variation and change in the rural and insular setting of Martha's Vineyard (Massachusetts), but subsequently turned the spotlight on linguistic variation in urban sites; most famous are his studies of New York's Lower East Side. These studies grew out of Labov's observations that speakers pronounced certain linguistic features (“sociolinguistic variables”) in different ways (i.e., as different “sociolinguistic variants”) and observed how this appeared to be related to the speaker's categorical membership as well as to the attention they paid to their own speech in the interaction. In his most famous study, he demonstrated correlations between New York speakers' socioeconomic status, their attention to speech, and whether they did or did not pronounce the sociolinguistic variable/r/in final and preconsonantal position—for instance, in words like “car” and “fourth” (Labov, 2006, 140). The higher the socioeconomic status and the more attention paid to speech, the higher the frequency of/r/pronunciation.

During the decades following Labov's original studies, scholars applied his methods to research on language use in many other metropolitan areas, mainly in the United States and in Europe. The city was the favored site of investigation within this strand of sociolinguistics. This is because scholars considered urban sites more diverse and dynamic than rural ones. This is, however, a view that scholars increasingly contest in contemporary studies.

In these first sociolinguistic studies, as in dialectological studies, researchers paid little attention to characteristics of place. The urban areas under scrutiny were primarily approached as Euclidean space; however, sociolinguists also took aspects of social space into consideration, since their aim was to explore linguistic variation tied to stratifications in social space.

Researchers from the late 1970s onwards began to apply a more practice-based and dynamic view of “the social” to their studies than the rather static models of social stratification allowed for by traditional variationist sociolinguistics. In their studies from Belfast during the 1980s, Lesley and James Milroy argued for the key role of social networks and practices in linguistic variation within a speech community. They demonstrated how network density and complexity influenced the extent to which speakers would comply with local linguistic norms or use standard features. Another key figure in the development of sociolinguistics, Penelope Eckert, in the late 1980s investigated how college students in a suburban American high school formed different communities of practice and, as part of their everyday practices, displayed highly different geographical orientations. “Burnouts,” oriented toward life in the inner city, were more likely to use innovative linguistic variants than “jocks,” oriented toward suburban space and lifestyles.

Summary

Phonetics has expanded its territory quickly in the past few decades. Dozens of new disciplines have emerged. By interacting with them, articulatory phonetics has changed enormously so that a split of the discipline has taken place. Half of it remains as articulatory phonetics but is leaving from linguistics. It has become more technology oriented, and many efforts have been made in articulatory modeling, which is more like engineering phonetics than linguistic phonetics.

The other half remains in linguistics but is no longer pure articulatory; it resorts to more parameters than the articulatory parameter. This split is so significant for linguistics that it would reshape linguistic articulatory phonetics and affect disciplines such as phonology, typology, dialectology, and evolutionary/historical phonology. This linguistic articulatory phonetics is becoming more theoretically oriented. It is still part, maybe the core part, of the greater phonetic science, yet differs from other phonetic disciplines in that it is of linguistics, by linguistics, and for linguistics. So it is not ‘pure’ phonetics. In linguistic phonetics, physiological and acoustic parameters are sought after, but they should be phonologically conceptualizable or categorizable, or have linguistic implications for typology or phonological evolution.

Phonetics is to phonology what physics is to chemistry. In phonetics and physics, the basic unit is the smallest analytical element, while in phonology or chemistry, the whole study is established on a structural unit: syllable in phonology and molecule in chemistry. Had chemistry not had its own basic, structural unit, it would not have been independent or autonomous from physics. Phonetics overlapping with linguistics has given rise to the linguistic, articulatory phonetics, which, as part, maybe the core part, of phonology, has the syllable as its basic unit.

Early Studies of Language and Social Class

Since the rise of the discipline of sociolinguistics during the latter half of the twentieth century, social class has occupied a central place in the study of language variation as a key factor offering important insights into social structure. Until the 1960s, most studies of variability were concerned primarily with regional variation or dialectology, following a tradition established in the nineteenth century. These studies concentrated their efforts on documenting the rural dialects which were believed would soon disappear. Only during the latter half of the twentieth century would the concern for status-based differences in language become a primary rather than a secondary focus, when sociolinguists turned their attention to the language of cities, where an increasing proportion of the world's population lives in modern times. The rise of urbanization is connected with an increase in social stratification reflected in linguistic variation.

Research focusing on social dialects is sometimes referred to as social dialectology and occupies a central place in quantitative sociolinguistic research on urban speech varieties, beginning with William Labov's (1966) work in New York City. Although linguists were aware before of differences in language use tied to social class, Labov was the first to introduce a systematic methodology for investigating social dialects and produced the first large-scale sociolinguistic survey of an urban community. Earlier, for instance, Alan Ross (1954) suggested that certain vocabulary differences in English could be classified as U (upper class) or non-U (lower class), e.g., serviette (non-U) versus table-napkin (U), one of the best known of all linguistic class-indicators of England at the time. Labov, however, concentrating primarily on pronunciation differences, called phonological variables, showed how idiolects (or the speech of individuals) considered in isolation might seem random, but the speech community as a whole behaved regularly. Previous investigations had concluded that the speech of New Yorkers appeared to vary in a random and unpredictable manner. Sometimes they pronounced the names Ian and Ann alike and sometimes they pronounced postvocalic /r/ (i.e., r following a vowel) in words such as car, while at other times they did not. This fluctuation was termed ‘free variation’ because there did not seem to be any explanation for it. Labov's study and subsequent ones modeled after it, however, showed that when such free variation in the speech of and between individuals was viewed against the background of the community as a whole, it was not free, but rather conditioned by social factors such as social class, age, sex, and style in predictable ways.

Unlike previous dialectological studies, which generally relied on one person (usually an older male) as representative of a particular area, most sociolinguists choose a sample of persons with different social characteristics. Labov's survey was based on tape-recorded interviews with 103 people who had been chosen by random sample as being representative of the various social classes, ages, ethnic groups, etc. to be found in New York City. This approach solved the problem of how any one person's speech could be thought of as representing a large urban area. Thus, while idiolects (or the speech of individuals) considered in isolation might seem random, the speech community as a whole behaved regularly. Using these methods, one could predict that a person of a particular social class, age, sex, etc. would pronounce postvocalic /r/ a certain percent of the time in certain situations. Through the introduction of these new methods for investigating social dialects by correlating sociolinguistic variables with social factors, sociolinguists have been able to build up a comprehensive picture of social dialect differentiation in the United States and Britain in particular, and other places, where these studies have since been replicated.

3 Linguistic Contributions to Area and International Studies

Many time-honored endeavors of linguists (investigation of unknown languages, research on the relations among languages, preparation of descriptive and pedagogical materials) yield valuable results for area and international studies. Relevant methods and results are discussed under three broad headings below.

4 Social Dialectology

Social dialectologists are concerned with tracing the relationships between linguistic variation and social factors. The pioneering research of William Labov determined the direction of subsequent research in the area for several decades, both in terms of the research questions he considered worth addressing, and in relation to the innovative methodology and quantitative (‘variationist’) analytical techniques he developed. His influential survey of the New York East Side (Labov 1972a) demonstrated that linguistic variation was not random or ‘free’ as previous dialectologists had tended to assume, but rather patterned systematically according to factors such as the social class, gender, age, and ethnicity of the speakers, and the formality of the style of speech they were using (see Language and Social Class; Language and Ethnicity).

The basic patterns which Labov identified in New York were replicated in many other urban speech communities including cities in Canada, England, Australia, and New Zealand. In Europe too, variationist approaches gradually displaced those of traditional regional dialectologists, and were adopted in urban dialectology research with similar results. Researchers found, for instance, that stable variables, which typically included grammatical variables, such as multiple negation, and phonological variables, such as final -ing (e.g., he don't know nothin') in English, tended to divide the community clearly into two major social groups. Linguistic variables involved in change, however, were typically more finely stratified with a gradient pattern reflecting a continuum from the lowest to the highest social groups. In the great majority of surveys, women lead sound change whether towards the standard or towards the vernacular variety. In New Zealand, for example, women are leading the rapid spread of the high rising terminal, as well as the merging of the vowels in words such as EAR and AIR. Recent variationist studies indicate that in specific speech communities variables such as class, ethnicity, social role, and style often interact to modify such general patterns.

Labov formalized the idea that linguistic change could be studied in progress; he analyzed ‘apparent time’ data demonstrating the spread of postvocalic (r) in words like car and card through different generations of the New York speech community. This approach to the study of language change has since been used by sociolinguists in speech communities worldwide.

Labov also used quantitative methods to collect people's subjective evaluations of different social accents he played to them, a method which has been widely used since in language attitude research. He demonstrated that, despite wide variations in their linguistic production, there was a startling uniformity of evaluative norms and attitudes to language among members of the same speech community. The New York accent, for example, was universally disparaged, a pattern replicated in many other urban speech communities. On the basis of this analysis, Labov suggested that the most useful defining characteristic of a speech community was precisely such shared linguistics norms. Others have suggested shared rules of use, density of patterns of interaction, or a shared pattern of variation. More recently, some sociolinguists have begun to explore the concept of the ‘community of practice’ as a useful concept in sociolinguistic research which moves the focus to the ways in which communities are constructed through shared goals, activities, and verbal resources (see Holmes and Meyerhoff 1999).

Labov set the methodological standard in social dialectology, influencing all subsequent research. He used an extensive interview schedule which included word lists, reading passages, questions skilfully designed to elicit casual speech, and innovative attitude elicitation techniques. Later refinements include the use of a network approach by Lesley Milroy and James Milroy in the study of working class Belfast. Instead of using social survey results and consequently interviewing strangers, they used networking (the ‘friend of a friend’ approach) to contact potential contributors. They also operationalized the network concept in terms of density and multiplexity, demonstrating that those who were most closely integrated into the local community used more vernacular forms (Milroy and Margrain 1980).

Current variationist research can be found in the journal Language Variation and Change with material ranging from the finest vowel and consonant distinctions to the broadest features of interactional discourse. For instance, the use of quotative be like (as in so she was like you aren't really going out wearing that thing and he was like why not) provides one example of an interesting, relatively new discourse marker which is being well documented as it spreads from American English into other varieties (e.g., Tagliamonti and Hudson 1999).

25.3 Dialect maps

Language mapping has had a long history, with linguistic variation, i.e., areal differentiation based on linguistic features, its primary focus. These maps can be based on different levels of grammar, for example phonological (sound) maps, lexical (word) maps or syntactic (grammatical/syntax) maps. Early large-scale atlases include Wenker's (1888–1923) initial Deutscher Sprachatlas and Gilléron and Edmont's (1902–1910) 13-volume Atlas linguistique de la France (ALF). The first dialect maps were hand-drawn, and displayed the locations in which individual variants of lexical items were used, based on either survey methods (Wenker) or field worker interviews (Gilléron and Edmont). This research field is called dialectology, and these works are called dialect maps or atlases, which can be either display maps or interpretive maps (Chambers and Trudgill, 1998). Display maps simply ‘transfer tabulated responses for a particular item onto a map, putting the tabulation into geographic perspective’ (Chambers and Trudgill, 1998: 25), while interpretive maps involve more analysis by linguists, for example by adding isoglosses between geographic areas which show differences. A handy modern example of this difference is the Algonquian Linguistic Atlas (Junker et al., 2005–2017), which plots audio recordings uttered by speakers at different geographic points (display maps), and the Algonquian Languages Dialectal Maps (Cenerini et al., 2017), which analyses the data in the Algonquian Linguistic Atlas to interpret dialectal lines between related Algonquian languages (interpretive maps). For an excellent overview and history of language mapping see Lameli et al. (2010).

Linguists have of course long recognized that language varies within social groups. Lameli et al. (2010) show for example that Schmeller (1821) and Wegener (1880) both distinguished linguistic features as spoken between different socio-economic groups. However, traditional large-scale dialect atlases focused on geographical analysis of these dialectal phenomena with the speech of the most ‘local’ speakers, namely NORMs (non-mobile, old, rural males) (Chambers and Trudgill, 1998), chosen to best represent the ‘deepest’ dialect of a region. Although a vast majority of dialect atlases have traditionally focused on NORMs, there are some language mapping projects which take into consideration social variables in addition to geography (see Lameli et al., 2010: 612–622 for an overview and examples). These projects use elements such as shading, colour, trees or frequency charts to help the reader visualize the multidimensionality at work in the language variation under study.

Over the years, technological developments helped make the development of dialect atlases less arduous and more visually appealing. From 1950 on, some speakers were audio-recorded, but most projects relied on field workers to make impressionistic notations based on their ear as the data of study (Labov et al., 2006). Today, we have much evidence that acoustic analyses of recorded data are far superior to impressionistic transcription analyses, and with the increased capability and accessibility of good quality audio-recordings and acoustic analysis software, recordings have become necessary components of variation analysis (Labov et al., 2006), creating an entirely new subfield of variationist study called sociophonetics (see, for example, Di Paolo and Yaeger-Dror, 2010; Kendall, 2013; Thomas, 2011). Increasingly, dialect atlases are published online using GIS and representative audio files, such as the Atlas of North American English (Labov et al., 2006) or Junker et al. (2005–2017), where the capturing of audio can enhance the user's understanding of the variation. These new dialect atlases, using GIS mapping tools and audio files to geo-tag linguistic features onto maps using online data, have made dialect maps an increasingly common way of displaying and popularizing linguistic research.

Despite the advances in technology leading to these audio-enhanced GIS-driven maps, however, much of the rich linguistic variation remains lost in the mapping of the data. As we have seen, linguistic variation and change is driven and characterized by a complex intertwining of social variables, and for the most part, these variables and interactions are not reflected in GIS-driven maps, since by definition, they reflect static geographical differences. In a field sometimes called linguistic variation and change, dialect maps can effectively display only one locus of variation, and cannot display change effectively at all. Heat maps (cf. Katz, 2016) can be used to show the intensity of certain variants as compared to others, but they do not (easily) show changes in progress or gender differences. However, there are (at least) three compelling reasons to want to create data visualization methods for linguistic change as well as for loci of linguistic variation other than geography. The first reason is field-internal; that is, to aid researchers in sociolinguistics: data visualization helps find patterns in data which may not be obvious in other common forms of data storage, such as spreadsheets and databases. The second reason relates to communication and translation of research findings. Given intense public interest in language-related research coupled with increased importance of translating research for an interested but naive public, we seek methods which will allow the complexity of language to be better translated and understood. Third, it is not uncommon in variationist work to add layers of newly collected data to those already collected and analysed; language is ever-changing, and sociolinguists are constantly trying to keep up. Since traditional GIS maps are more or less fixed, iterative study with added data becomes difficult, if possible at all. A new set of maps must be designed with new data, starting from scratch. A better tool would be able to incorporate new data as it is added, which is what we seek in the cybermapping process.

10 The relation of language ecology to other branches of scientific inquiry

Haugen's proposal raises, moreover, the intricate problem of how to range language ecology among other research concerns or research areas and how mark it off from these. In other words, how are we to demarcate language ecology from other research agendas or fit it into a classification of sciences and scientific fields (cf. Eliasson, 1987). As indicated in Table 4, Haugen enumerates a number of scientific areas that are implicated in his ten research questions. Here it emerges that the aims of language ecology overlap considerably with those of miscellaneous linguistic sub-disciplines such as philology, language history, dialectology, descriptive linguistics, psycholinguistics, sociolinguistics, sociology of language, ethnolinguistics, linguistic anthropology, and so forth (cf. 1972, 327). The text does not, however, make it clear exactly how these apparent overlaps are to be understood. Most easily, we may associate them with a view of language ecology as a research perspective alone — a point of view that may be pursued within the bounds of various sub-disciplines. But when Haugen says that language ecology is a research field, an uncertainty arises as to where a given scientific problem in reality belongs, to language ecology or to some other linguistic sub-discipline. To mention just one instance of unclear responsibilities, is it up to language ecology to classify languages genetically (part of Question 1, Table 4) or should this remain the task of historical linguistics? Or alternatively, is language ecology perchance to be seen, at a more general level, as a kind of meta-discipline (cf. Lechevrel, 2010a, 163)? In this case, genetic language classification would trivially be a task of language ecology, because it is a task of what would be a ‘sub-field’, viz., historical linguistics. Still another complication ensues when — as we saw in Section 5 — Haugen on the one hand claims that “sociolinguistics is an ‘ecology of language’ and therefore part of the ecology of society” (Haugen, 1978, 7) and on the other hand that “‘language ecology’ is … one aspect [italics, SE] of the interdisciplinary field of sociolinguistics or the sociology of language” (1979b, 245). But then again, he remarks: “I have been asked why one should say ecology rather than sociolinguistics. I thought of the former as somehow much broader than the latter” (Haugen, 1987b, 81). The picture becomes even more complicated when he underlines the point that “the preservation of language [a topic of language ecology, SE] is a part of human ecology, which in turn is a branch of the larger disciplines of sociology and political science” (1985, 4f.). Possibly, he may at times have had something like the relations in Fig. 3 in mind.

Human ecology would then be located in the intersection of political science and sociology, and sociolinguistics would be a sub-field of human ecology. But whether language ecology is merely to be seen as an aspect of sociolinguistics or whether sociolinguistics is in effect a part of language ecology as in Fig. 3 is not entirely settled by Haugen's statements cited above. Nor does he address how linguistics in its entirety is to be built into such a disciplinary structure.