In present-day linguistics the term variety is used to refer to any variant of a language which can be sufficiently delimited from another variant. The grounds for such differentiation may be social, historical, geographical or a combination of these. The necessity for a neutral term such as variety arose from the loaded use of the term dialect: this was not only used in the neutral sense of just a regionally bound form of a language, but also with the implication that the linguistically most interesting forms of a language are those spoken by the older rural non-mobile male population. This view is understandable given the origin of dialectology in the nineteenth century, that is in the heyday of historical linguistics. Nowadays, sociolinguistic attitudes are prevalent and the need for a term which can include the investigation of urban users of a language from a social point of view became evident. The neutrality of the term variety must be stressed. It simply refers to a distinguishable variant of a language. This means that there are a large number of varieties of any given language. The sole criterion to be fulfilled by a particular variety is delimitation vis à vis other varieties. Dialects within a variety framework are frequently referred to as regional varieties and sociolects as social varieties, though the label dialect can be retained if used objectively.
1) Dialect Strictly speaking the term dialect refers to a geographical variant of a language. However, it is used loosely, not only by non-linguists, to talk about any variety of language. For sociolinguistic purposes one must distinguish various sub-types of dialect.
The term dialect is used to denote a geographically distinct variety of a language. There is no reference to the social dimension of language here. It is also important to stress that the standard of a language is nothing more than a dialect which achieved special political and social status at some stage in the past and which has been extensively codified orthographically.
a) koiné This is a term deriving from ancient Greek ‘common’ and refers to the situation where, in a group of dialects, one is predominant and used outside of its natural boundaries as a means of inter-dialectal communication. This was the case with Athenian Greek and the remaining dialects in Classical Greece and - at least for writing - also held for West Saxon vis à vis the other dialects of English in the Old English period.
b) patois This is a French term which refers to a dialect which is unwritten and as such without a literary tradition. The (French) term dialecte conversely refers to a geographical variety which has an associated literature. This use is to be found in other countries of Europe as well, such as Sweden.
2) Standard and non-standard In a countries where languages have a long written tradition and a literature, such as the majority of countries of the West, it makes sense to talk of a codified standard. By implicit or explicit comparison with this standard one can then classify other varieties as non-standard. In each country there is a term for the standard. In England there are various terms such as The Queen’s English, Oxford English, BBC English, Received Pronunciation whereas in Germany the standard is known simply as Hochdeutsch. Of the terms for standard (British) English, only Received Pronunciation these finds favour with linguists. Although the laypeople may use the other terms indiscriminately and although they may not be able to be precise about what they mean by them, they are always able to recognise them and may not infrequently be in a position to imitate them also. Here one sees that the receptive ability of speakers is greater that their productive ability.
There are a number of further labels which are used to refer to language variation along various axes. Students should be aware of at least the following three terms.
| Diatopic | Refers to variation in language on a geographical level. |
| Diastratic | Refers to variation in language between social classes. |
| Diachronic | Refers to variation in language over time. |
3) Vernacular This term refers to the language spoken naturally by the inhabitants of a country (cf. German Volkssprache) as opposed to a possible classical language which may have a position of dominance in cultural or ecclesiastical spheres.
4) Mutual intelligibility of dialects Here one is confronted with the problem of how to distinguish between language and dialect and with the related problem of how to decide what a language is. One way of characterising ‘language’ and ‘dialect’ is to regard languages as a collection of mutually intelligible dialects and a dialect as a recognisable variety within this group.
The criterion of mutual intelligibility is not entirely successful. One common problem with this criterion is that some languages like Norwegian, Swedish and Danish are usually considered, for political reasons, as different languages but speakers of these three languages can readily understand and communicate with each other. It may also be the case that dialects belonging to the same language lack mutual intelligibility. German, for instance, would not be considered a single language because some types of German are not intelligible to speakers of other types. Furthermore, mutual intelligibility may not be equal in both directions. It is often said, for instance, that Danes understand Norwegians better that Norwegians understand Danes. This leads us to another difficulty: the criterion admits of degrees of more or less because many Swedes, for instance, can readily understand many Norwegians but not in the same way as they do Swedes.
5) Polylectal grammars Obviously, speakers of different dialects are able to understand each other more or less. This can be seen with speakers of both English and German. The reason is that the linguistic systems involved do not differ fundamentally. The understanding of different dialects implies that the speakers know the overall system of the language (group of dialects) and use it just as they convert underlying forms to surface forms by rules in syntax for instance. This view is what is called polylectal, from ‘lect’, meaning form of language. However, we have reason to question the knowledge of common underlying forms. The phenomenon of hypercorrect utterances, e.g. the introduction of an /r/ into the pronunciation of lager /lɑ:ɹgəɹ/ by speakers who pronounce the final -r in English in general, proves that underlying forms are not present for all speakers, i.e. they introduce the /-r/ where it does not exist in Received Pronunciation because they feel the back pronunciation /ɑ:/ implies a following /-r/ which is true of r-ful dialects but not of those without syllable final r.
Speakers understand more dialects than they speak. One explanation for this states that speakers of different dialects are able to communicate because of their receptive competence which is part of their native speaker competence in general. It might then seem legitimate to construct polylectal grammars, grammars that incorporate more than one variety. This leads one to the question of how much polylectal competence speakers possess, what the speaker is able to do with his/her knowledge, and, for instance, where polylectal grammars start and stop on a continuum. It does not necessarily follow, however, that the polylectal grammar is the best way of accounting for mutual intelligibility. It may be the case that the degree of mutual intelligibility is related to the differences, rather than the similarities, between the two grammars in question.
6) Geographical dialect continua In many parts of the world, if we examine the rural areas, we are faced with a situation which is known as a geographical dialect continuum. There are linguistic differences between the dialects of different villages in rural areas, which are sometimes larger and sometimes smaller but the further we get from a definite starting point in an area, the larger the differences become. The striking point is that a chain of mutual intelligibility links all the dialects spoken throughout the area. At any point on this extensive continuum, speakers of one dialect can understand speakers of other dialects who live in adjacent areas to them. In other words, dialects on the outer edges of the geographical area may not be mutually intelligible, but they are linked by a chain of mutual intelligibility. At no point is there a complete break, but the cumulative effect of the linguistic differences will be such that the greater the geographical separation, the greater the difficulty of comprehension. This situation is clearly illustrated by German dialects which form an uninterrupted continuum from the Dutch border in the north west to the Hungarian border in the south-east. The varieties in these extremes are not mutually intelligible but at any two points on the continuum they are.
7) Social dialect continua Dialect continua can also be social rather than geographical. A good example of this is provided by the situation in Jamaica. The initial linguistic situation in Jamaica was such that those at the top of the social scale, the British, spoke English, while those at the bottom of the social scale, the African slaves, spoke Jamaican Creole. English, the international and prestigious language of the upper classes, had a considerable impact over the centuries on Jamaican Creole. Since Jamaican Creole was recognised as being similar to English, although these two languages are not mutually intelligible, it came to be regarded as inferior or debased vis à vis English because of the social situation.
As a result two things happened. First, the ‘deepest’ Creole, what is known in linguistics as the basilect, is now a good deal closer to English than before. Secondly, the gap between the varieties from relatively standard English to ‘deepest’ Creole has been filled and forms the social dialect continuum. The problem with this social dialect continuum is that there is no point on the continuum where English stops and Creole starts. Therefore, any division into two parts would be linguistically arbitrary. The result is that Jamaicans are taken to speak English. In fact, some Jamaicans do speak English, some do not, and some speak varieties where it is not really possible to judge. To deal with this scale between standard English and Creole, linguists use the terms acrolect (nearest to the standard), mesolect (the mid range) and basilect (furthest away from the standard).
8) Autonomy and heteronomy A useful concept in looking at the relationship between the notions of ‘language’ and ‘dialect continuum’ is the concept of heteronomy. Certain varieties on the West Germanic dialect continuum, for example, are dialects of Dutch while others are dialects of German. The Dutch dialects are heteronymous with respect to standard Dutch, and the German dialects to standard German, the respective superimposed autonomous standard varieties.
Heteronomy and autonomy are the result of political and cultural rather than purely linguistic factors and are therefore subject to change. A useful example of this is provided by the history of what is now southern Sweden. As a result of war and conquest, parts of Denmark became Swedish. The Danish dialects spoken on that part of the Scandinavian dialect continuum finally became dialects of Swedish, even though the (former Danish) dialects themselves did not change at all linguistically but they had become heteronymous with respect to standard Swedish rather than Danish.
Just as the direction of heteronomy can change, so formerly heteronymous varieties can achieve autonomy, often as a result of political developments, for example Middle Low Franconian became autonomous with the independence of the province of Holland and evolved into Dutch, the official language of the Netherlands. While ‘new’ languages may develop the linguistic forms need not be new. In other cases, political separation may lead not to autonomy but to semi-autonomy as in the case of Swiss German. It is also possible for autonomy to be lost, and for formerly independent varieties to become heteronymous with respect to other varieties. Scots was formerly an autonomous variety, but it has been regarded as a variety of English for the last two hundred years (movements are currently in progress to achieve at least semi-autonomy).
What is an isogloss?
Isoglosses Boundaries between two regions which differ with respect to some linguistic feature are called isoglosses. The term isogloss literally means ‘same language’ (iso + gloss). The term is used in two slightly different ways and is also represented graphically in two different ways. One way of displaying an isogloss is to draw a single line between two regions which are found to differ with respect to some linguistic feature. The single line separating the regions is the isogloss.
The alternative representation links by means of lines the locations of speakers who share the realisation of feature a with those who share feature b. The two lines form a heterogloss referring to those speakers who are at the interface between the two isoglosses. While the heterogloss is more precise at the interface, it is neutral with regard to any claim about those features (a and b). A single isogloss is in turn less precise about the interface, cutting through it arbitrarily. However, there are lots of cases in which both, isoglosses and heteroglosses make the same claim (compare gradual and abrupt transition).
Patterns of isoglosses Certain patterns of isoglosses have recurred time and again in various surveys that have been carried out. Their recurrence is an interesting fact about dialect, but what is also striking is the pattern itself. One of those patterns shows up as a welter of isoglosses that criss-cross one another almost chaotically. A classic example for such a pattern displaying a wild variety of combinations of dialectal elements is the set of isoglosses which separate Low German from High German and which runs east and west across Germany and Holland on a line just slightly north of Berlin. For part of their length they run more or less parallel to one another in a loose sense. Suddenly, at the point where they meet at the Rhine river, the elements go their separate ways so that it is impossible to make useful generalisations about High and Low German. The point in the northern Rhine area is called the ‘Rhenish fan’ which has become an instructive example for isoglosses going their separate ways. This pattern of criss-crossing isoglosses separating even contiguous villages is considered to be typical of a region that has had a long settlement history and is parallel to the observation that linguistic variation increase the closer one gets to an area of original settlement.
Bundles of isoglosses It is important to note that each isogloss plots a single linguistic feature. The significance of a dialect area increases as more and more isoglosses are found which separate an area from adjacent ones. The coincidence of a set of isoglosses is called a bundle, as for instance in the case of the isoglosses running throughout Germany. Perhaps the most striking example of a bundle of isoglosses was obtained by the French survey by Gilliéron and Edmont. The investigated bundles have a particular prominence in the number of isoglosses which come together to form it, and in their closeness throughout the entire area which they cover.
Grading of isoglosses It seems clear that some isoglosses are of greater importance than others - often depending on what particular feature they mark. With bundles it is almost the same. However, in the history of dialectology, no one has succeeded in developing a set of principles for grading isoglosses or bundles of them. Several interesting attempts have been made. One of the most prominent is referred to as dialectometry which describes a formula for indexing the dialectic ‘distance’ for any two speakers in a survey.
Structural categories of isoglosses In attempting to determine the linguistic significance of isoglosses, categorising them according to the type of linguistic feature they describe may be the first step followed by grading them according to their linguistic structure or empirical observations. The categories can be characterised as follows:
Lexical isoglosses describe contrasts in the words used by different speakers to characterise the same object or action, like the use of the words dutch cheese and cottage cheese. Pronunciation isoglosses include most of the examples discussed so far, referring to contrasting pronunciation. It seems appropriate to rank lexical differences as more superficial than pronunciation differences because the former are more likely to be subject to self-conscious control or change by speakers than the latter. This may well have to do with the status of lexis and phonology as open and closed classes respectively.
In phonology, there are also two types of isoglosses. The first one is phonetic, involving contrasts in the phonetic output of two regions as the result of a more general or an additional phonological rule. Differences in phonemic inventories, on the other hand, involve phonemic isoglosses. There are thus two kinds of phonological isoglosses, and it might be tempting to rank them by attributing greater significance to the phonemic type, since it has greater structural significance. The remaining types can be subsumed under the heading ‘grammatical isoglosses’. One subtype is morphological, i.e. it involves inflectional and derivational differences between contrasting regions. The second subtype, the syntactic isogloss, refers to aspects of sentence formation. Both types of grammatical isogloss are quite rare, and thus it is difficult to grade them relative to one another. Finally, another type of isogloss can be referred to as a ‘semantic isogloss’. Semantic isoglosses involve contrasts in meaning from one region to another.