logo ELT Concourse teacher training
Concourse 2

Language and society: sociolinguistics


This is an overview focusing on the nature of the study of sociolinguistics, some of the main ideas and categories of concern.
It also comments on the implications for English language teaching, as you might expect of this site.
For more on sociolinguistics in general, and those aspects of ELT most closely influenced by sociolinguistic ideas, please consult the bibliography and related links at the end.

Sociolinguistics is a very large area of study which overlaps with many others including, but not limited to, anthropology, sociology, ethnology, pragmatics, geography and politics.  The study concerns how language use relates to and is influenced by social factors such as nationality, class, ethnicity, sex, regional origin and education.
It is, or should be, of interest to language teachers because understanding a little of how language use is affected by such factors allows us to select the kinds of language we should be teaching, the settings in which it may be presented and the people we show speaking and writing it.

In what follows, we'll select the key issues in sociolinguistics, summarise what they are about and indicate the implications they may have for language teachers.

Here's a list of the contents to take you to its various sections if you are looking for something specific.
Clicking on -top- at the end of each section will bring you back to this menu.

How influence works Language Dialect and accent Speech communities High- and low-status varieties Variation and social class
Variation and ethnicity Variation and register Variation and interaction Variation and sex Summary diagram Selected bibliography


Influence acts on variety

Sociolinguistics is mostly to do with influences on language use from various sources.  Influence, however, acts in two obvious ways:

  1. It can act to limit variety so, for example, the compulsory wearing of a uniform or a law constraining people's actions both act to limit variety.
    In language, some influences, such as an education system which teaches 'correct' rules for the use of pupils' first languages, act to limit the varieties which would naturally be present and could make people less able to communicate effectively.  It is also noticeable that the international spread of mass media such as film, the internet and television and the ease with which people now travel across national borders and encounter speakers of a range of varieties of their own language may also act to homogenise a language and suppress variety quite effectively.
  2. It can act to change and increase variety so, for example, if musicians are asked what influences have shaped their music, the question really is: How have other people's productions varied your production?
    In language, the further apart people are in terms of culture or simple distance, the more likely it is that the local or regional influences will affect which language people speak, how they use it, their vocabulary, the grammar they employ and the ways in which they speak and write.
    If dialects of a language are separated by geographical, political or cultural barriers, the tendency is clear for them to drift apart while maintaining some commonalities.  Eventually, the process leads to a form of speciation which, like species in the natural world, are no longer able to produce viable offspring.  In other words, the dialects morph into languages proper and mutual comprehensibility is gradually lost.
    We can watch this speciation happening in real time when borders are inserted into areas once considered single nations so, for example, in erstwhile Czechoslovakia both Czech and Slovak were, for the most part, mutually comprehensible dialects of the same language.  With the separation of the country into two republics, many now designate the two varieties as independent languages and the expectation is that they will drift apart and may eventually become less and less mutually understood.
    There are those, for example, who see this process occurring with US and UK varieties of the language but we are not there yet and countervailing influences of mass media, internet use and so on are also acting to homogenise the language.  The jury is still out on which influence will be the stronger in the long run.



Variety: language, dialect and accent

The three most obvious indicators of variety in language are often assumed to be easily identifiable but, when it comes to it, are much more difficult to define than to recognise.


Defining a language

It's a simple matter to state without fear of successful contradiction that people speak difference languages.  So, for example, we can easily recognise that most people born and raised in Japan will speak Japanese and most of those born and raised in Turkey will speak Turkish and so on across the 195 countries normally recognised to have independent existences.  So far, so straightforward.
However, there are over 7000 human languages recognised by, for example, Ethnologue.  That's around 36 languages per country.  It is, in fact, impossible to identify a single nation on earth in which everyone speaks the same language.
Even very small countries exhibit dialect and language varieties.  For example, San Marino, with a population of fewer than 35,000 people has only one official language (Italian) but Romagnol (a dialect or language depending on one's point of view) is widely spoken there as it is across the whole of northern Italy.
Monaco, a state with around the same population (of whom only a minority of 25% are actually nationals) has French as the official language but Monégasque, a dialect of Ligurian, is widely spoken, as is Italian.
Papua New Guinea with a population of around 9 million is home to 851 different languages.  That's roughly one language for every 10,500 people.

The linguistic determiner for the identification of a language is the concept of mutual intelligibility and for most purposes, that's a reasonable criterion.  It is quite simple, again, to be able to state that English and German are not mutually intelligible so, linguistically, are autonomous languages.  Equally, Russian and Urdu speakers cannot communicate with each other using their respective languages because the languages are not in the least mutually intelligible.
However, mutual intelligibility is not a digital, on-off phenomenon and a little thought reveals that there are degrees of mutual intelligibility.  Educated speakers of, for example, Danish, Norwegian and Swedish have very little difficulty understanding each other when they speak their own languages but it is not easy to state that the three languages have no independent existence.
It is also the case that Hindi and Urdu are mutually comprehensible languages and speakers of them have little difficulty understanding each other.  They are, linguistically speaking, dialects of the same language but to assert that they are not independent languages also flies in the face of geographical, political and social facts.  The two varieties are written, for example, using mutually completely unintelligible scripts; Hindi in Devanagari and Urdu in a modified Perso-Arabic script.  Today, Urdu is the official language of Pakistan and Hindi one of the official languages of India.
A notorious example is the language (or languages) spoken in erstwhile Yugoslavia.  Bosnian, Croatian, Montenegrin, and Serbian are, in their spoken forms, mutually almost completely intelligible but political, social and to some extent religious pressures have resulted in their being defined by their speakers as separate languages.  All four dialects have in the past been written in a variety of scripts and today the languages are divided by the use of the Cyrillic or Latin alphabets.  Bosnian, Montenegrin and Serbian are written using both alphabets but Croatian only in the Latin alphabet.
Until recently, the varieties were generally subsumed under the umbrella term Serbo-Croat but today the preferred term is, with due respect to alphabetical ordering, Bosnian-Croatian-Montenegrin-Serbian.  Political pressures are, unsurprisingly, strong in defining the varieties as separate and independent languages.

The upshot of all this confusion is that we can define a language in three distinct but overlapping ways:

  1. A language constitutes a separate entity and is not comprehensible to speakers of other languages.  French and Swahili are, therefore, separate languages.
  2. A language is defined socially and politically by its speakers.  So, Bosnian, Croatian, Montenegrin, and Serbian are separate languages and Hindi and Urdu share this characteristic.
  3. A language is defined politically by the nation in which it is spoken.  Danish, Swedish and Norwegian are, therefore, separate languages and Swiss German is separate from German.

To identify to which language a dialect (see below) belongs, we need to look at the notions of correctness which its speakers perceive.  For example, in the borderlands between Germany and The Netherlands there are dialects which are mutually comprehensible on both sides of the frontier but speakers of the northern dialect will defer to Dutch as the standard of correctness and south of the line the reference will be to standard German.

Languages which are often defined as such by non-speakers of them also complicate the picture when the speakers themselves will not be able to understand each other when speaking ostensibly the same language.  So, for example, although Arabic is often seen as a single language and Chinese is similarly described, that is merely a convenience because both languages are better described as macro languages and within them the various forms of the language are not 100% mutually comprehensible (and, in some instances, not mutually comprehensible at all).  By most analyses, there are 14 mutually incomprehensible varieties of Chinese and 20 of Arabic.
The tendency with macro languages is to retain comprehensibility of the written form but for the spoken forms to drift apart.

Overall the picture looks a bit like this:


and, like most summaries, that hides a good deal of debate, controversy and detail.
A rule of thumb for the definition of a language is that it is a dialect with an army and a navy.  That's not a definition linguists would recognise but it makes some political, national and social sense.


Dialect and accent

This is another blurry distinction.
We have already seen that efforts to differentiate distinct languages from varieties of the same language are fraught with difficulty.  So it is with dialects and accents.
Before we go any further, we should make it clear that everyone speaks a dialect and everyone speaks with an accent, no matter which language(s) they speak.
The fact that some dialects and ways of pronouncing a language are considered in some cases to represent a standard form of the language is an accident of history.  The standard form is still a dialect and an accent.

Traditionally, a dialect is a variety of a language which differs from other varieties in three ways: grammar, lexis and pronunciation.
An accent, however, refers purely to differences in pronunciation.
So it is that we can talk about someone speaking English (or any other language) with an accent which betrays their origin or, often, their social class and ethnicity in some cases and we can talk of people using specific dialects of the language(s) they speak.
In this section, we shall mostly be using English as a familiar example, but the same considerations apply to almost any language which has multiple thousands of speakers.

When it comes to defining and describing the dialects of any language, we encounter the usual problem of trying to draw lines on a continuum: where are the boundaries?
For British English, for example, this map identifies some main dialect and accent regions:

and most speakers of British English will be able to identify where a speaker comes from very quickly, sometimes from a single word.
It is less likely, however, that a speaker who uses a non-British dialect of English, say a speaker of an American English dialect, will be able to do so and British speakers themselves are unlikely to be able to draw finer distinctions between the range of accents and dialects which this map does not show, if they are not members of the community which uses the variety in question.  Most English people will have difficulty separating Highland from Lowland Scots and many speakers of English from Ireland will be unable to identify the differences between East and West Midland English speakers.
It is also possible, for example, clearly to distinguish between the three main varieties of East Anglian (Norfolk, Suffolk and Essex) but drawing lines between them on a map is impossible because they merge imperceptibly into each other.
Speakers of English as a second language may be able to detect differences in pronunciation but identifying what they are and which region they are typical of will probably be beyond all but the most advanced and sensitive of them.
In sociolinguistics, this familiar problem is called the discreteness-continuity issue.

The same issue applies to almost any other large language one encounters.  Traditionally, for example, German in Germany is divided into three major dialects: Lower, spoken in the north, Central, spoken across the central band of the country and Upper, spoken south of that line.  However, each of these gross distinctions can be further divided into numerous subgroups (East Low German, Low Saxon, West Central, East Central, North Upper German, West Upper German and East Upper German) and each of those has been chopped up into finer and finer distinctions which no map can adequately show because of the merging and blurring of borders.
Italian, to take another example, is traditionally divided into around 14 recognisable dialects (not all of which are spoken within the borders of Italy).

However, there clearly are languages which we can describe and identify as English or German (or any other) and for that we look to the standard dialect which has, usually for historical reasons, become accepted as the norm to which others may aspire and which is the target of teaching and learning for non-native speakers.  A standard form, where such a thing exists, also acts as the base line from which other dialects are said to diverge.

Not all languages have a widely acceptable standard form to which speakers look for 'correct' usage and pronunciation and that can be a source of a good deal of disagreement and controversy (and occasionally conflict).
French has a standard form, usually close to that spoken in metropolitan areas and called variously le français standard, le français normé or le français neutre and German similarly has a standard form referred to as Standardhochdeutsch, Standarddeutsch, Hochdeutsch or, in Switzerland, Schriftdeutsch (betraying in the last instance the idea that the written form (Schrift) is a standard variety but spoken forms will differ).
On what makes a dialect a standard form, see below under status.



A speech community

Intuitively, we may define a speech community as a group of people who share the same language but there is more to it than that.

At its broadest, we may, for example define as a speech community all those who speak English as a first or second language.  This would imply a community of over two billion people.
At its narrowest, we can define a speech community as two people who speak the same language and use and understand terms and vocabulary unique to themselves.
Clearly there are many variations between these two extremes and they will concern people who share:

Clearly, nearly everyone is a member of more than one speech community.  For example, someone who lives and was brought up in the south of England may be a member of at least the following speech communities:

It is extremely rare for any speaker of a language to be a member of only one speech community.
You may like to pause here and draw up a list of the speech communities to which you belong.


Implications for the classroom

Learners of any language have an ambition to join a speech community.  The questions to answer are:

Membership of which speech community or communities is an appropriate target for them?
What form(s) of the language will be the most appropriate targets to acquire in order to qualify for membership?

The answer will depend on a number of considerations and we'll exemplify them here.

  1. The larger the speech community of which membership is the target, the more generalised and less specific will be the language needed to join it.
    Learners who need to operate across a range of cultures (English speaking or using English as a lingua franca) will not be helped by being taught language forms which are specific to any particular setting.
    So such a learner needs a non-regional accent and non-regional lexis and structure to be readily understood by and understand a range of speakers of the language.
    If, for example, a learner of English merely wishes to operate reasonably well in the language and communicate fundamental, non-specialised ideas to other users of the language (native speaking or otherwise), an acceptable knowledge of the systems of a standard form of English will be adequate.  The more regionally specific such a standard is, the less useful it will be so the focus will be on an idealised non-regional form.
  2. The more specialised the field in which a speech community operates, the more register-specific will be the language which is used.
    So, for example, if a learner wishes to become a member of a speech community concerned with the study of astrophysics, psycholinguistics or economics, the acquisition of register-specific lexis and structures will be a priority.
  3. Learners who wish or need to integrate to any extent, permanently or temporarily, into an English-speaking culture will need to acquire the language specific to that culture.
    So, for example, a learner who needs to live and operate successfully in Australia will have very different needs from one whose ambition is to be part of the culture of a Caribbean island.
  4. Learners and users of the language as a tool for communication in certain fields (industry- or company-wide for example) will need to be members of an often entirely non-native-speaking group of users of English and for this purpose may not need to control the whole range of structure, have a near-native pronunciation or use very much non-technical language.
    This is English as a Lingua Franca (ELF) or English as an International Language (EIL) and the language used may be severely restricted and almost, some say, pidginised.  The guide to varieties of English, linked below, has more on how this variety works.
  5. Learners who have more than one aim for how they will use the language, for example, as a student of an academic subject in an English-speaking institution, as a member of the wider student body with whom they are studying and as a resident of the country or area in which they are studying have more complex needs and may need to aspire to membership of multiple speech communities.

There are further obvious implications, of course, that concern both the teachers' behaviours and the materials that are used.
If the models the teacher provides, either personally or via the materials used in the classroom, are recognisably those of a limited speech community (by region, class, ethnicity, nationality or field of communication), then learners will be ill prepared to operate successfully outside that community.
Depending on the learner's need for the language, that may not matter but limiting input to language identifiably of one type of a speech community only will be unhelpful if the speech community's norms are mismatched by the settings in which the learner needs to operate.
In other words, teaching American English to someone who needs to operate in a British or South Asian community will be counterproductive.

Most English language teaching materials are produced in what Kachru (1985) defined as an inner circle country.  In this regard, these may be limited to the UK, the USA, Canada, New Zealand, Australia and Ireland and are overwhelmingly produced in the first two of those nations.  For more on Kachru's helpful analysis, see the guide to variety, linked below.
The forms of the language that materials writers choose to present, despite their best efforts to eradicate cultural influences, will normally be identifiably of these speech communities.




Above, we identified the fact that in many countries and language areas (rarely the same thing) one particular dialect of a language is often regarded as the norm from which other dialects differ.
In British English, this has emerged as the speech and language use of educated southern English people in and around the London area.  It was not always so and other dialects of English have variously gone in and out of fashion as providing the standard to which others may, or may be forced, to aspire.
That this dialect became and remains to a large extent a standard variety is an accident of history dependent on economic factors for the most part but also influenced by the fact that London (and the triangle based on it with its other corners marked by Oxford and Cambridge) has long been the intellectual, commercial, governmental and legislative centre of the country.

The accent of the standard form of British English is sometimes referred to as RP (Received Pronunciation) and is the pronunciation which for many years was the required norm in public schools, broadcasting organisations and similar settings.
For example, in southern British a word such as heart will be pronounced as /hɑːt/ with no audible /r/ sound.  However, speakers of many regional accents in Britain and elsewhere retain the /r/ sound in such words and they will pronounce the word as /hɑːrt/.  That will also, incidentally, be the standard pronunciation in most of north America and in many other English-speaking regions.
Nowadays, handbooks of pronunciation, unless they are concerned to describe regional differences in British or other Englishes, will use the RP pronunciation of phonemes as the standard forms.  Thus, for example a word like bath is transcribed on this site and elsewhere as /bɑːθ/ although very large numbers of people in Britain and across north America with pronounce the word as /bæθ/.

Along with RP, a standard dialect of English has also emerged with a set of rules of usage which are familiar to most learners and teachers of the language and often enshrined in school textbooks and other reference texts.  Dialect differences abound, of course, but are usually described, as here, concerning the ways in which they differ from a presumed standard so, for example, the standard form of:
    She was sitting by the window
will be, often disparagingly, compared to the non-standard:
    She was sat by the window
    John needs his hair cut
will be considered equally as an exception to the standard:
    John needs his hair cutting

Standard dialect and standard accent are, however, separable.  It is not necessary to speak standard English with an RP accent (and, indeed, quite rarely done).  Standard English is spoken in many local and regional accents.


Implications for the classroom

Because these two forms of the language have become standards, it is to them that most learners of English will look for models.  Few learners, unless they are very young or very talented, or both, will ever acquire an RP accent in English and most will never get close.  However, when the pronunciation of any word is in question, it seems to make some sense to identify it using something akin to an RP accent rather than a regional variety which may be less frequently encountered and may even be seen as of lower status.
Teachers with particular strong regional accents should, therefore, attempt to move towards RP rather than provide models which are deviant from it.

The same considerations apply to the grammar and lexicon of the language.  Although in many parts of Britain the term nanny refers to a live-in, paid child carer, in many other parts of Britain, the term refers to a grandmother.
An utterance such as:
    Happen he were late to the meeting
which might be a natural form in the East Midlands and areas further north and east exhibits both lexical and grammatical deviance from standard English and may be translated as:
    Perhaps he was late to the meeting.

Unless we are helping learners whose horizons are limited by residence in particular parts of England and who will not need to use their English in a standard written form or outside limited and defined areas, teaching dialect forms such as these will not be very helpful.

In other words, to state the case rather too forcefully, whatever variety of English teachers habitually use, they should teach and present as models standard English forms only.
If that means losing a strong regional accent and learning to use a standard dialect and pronunciation, well, then, it does.




To start with, we need to get away from the idea that informal speech is somehow less admirable than formal speech.  All varieties of all languages can be used in formal and informal ways.
It is not the case that the use of slang expressions, contractions and other spoken simplifications are somehow indicative of lower-class or uneducated speech.  Even less is this an indicator of lazy or incorrect use.

Nevertheless, there do exist social-class dialects in most languages and these are, for native speakers, usually instantly recognisable.  In English, for example:
    It was her what stole it
instead of
    It was her who stole it
or even
    It was she who stole it
is, for most people, an indicator of social class differences, the latter two being higher status than the first.
What we have identified here is known as a sociolect.
English is not alone, of course, in having forms of the language which signal upper-, middle- or lower-class speech as we shall see.

All societies are, to a greater or lesser extent, stratified.  That is the basis of social class analysis.
Even discussing social class raises some people's hackles and there are as many ways to classify social class as, probably, the number of people investigating the area.  Usually, classes are identified in socio-economic terms and that involves measuring, for example, income levels, educational backgrounds, cultural participation, occupations and so on.
Very loosely, most analyses of western societies divide classes grossly into lower, middle and upper strata but those divisions are also further subdivided into smaller units.
In other societies, class may be defined by caste, by birth or by occupation very precisely.
As a working definition we may divide classes in most societies into the traditional three groups and see what effects, if any, class membership has on the types and forms of language that people use.  Like this:

Class Description Language accent Language dialect
Upper A small group of wealthy and usually highly educated people who control wealth and access to wealth and power.  In some analyses, such people will constitute only around 2 or 3% of the population. A clear tendency to avoid regional accents and speak a standard form of the language. Uninfluenced by region with a strong tendency to use standard (prescribed) grammatical and lexical forms.
Middle A large group of people (often 50% of a population) which is frequently subdivided in to Upper, Middle and Lower bands.
People in this group have income and wealth above subsistence level, may be property owners and work in skilled occupations, sometimes highly paid.
Educational levels will be quite high with often a university degree or other professional qualification.
People in this group will often speak with a mild but recognisable regional accent.
They may deliberately emphasise or suppress a local accent depending on perceived formality and the setting.
There will be local deviations from a standard dialect but people in this group will often be adept at switching between dialects depending on setting and levels of formality.
Lower A large group often subdivided into Upper, Middle and Lower bands.
People in this group will often have an education up to secondary school but rarely beyond although those in the upper band may have work-based qualifications.
At the lower end, people may have a very basic education and work in low-skilled, poorly paid jobs or be unemployed.
People in this group often use a strong regional accent and are not able to switch to a standard pronunciation. Deviations from prescribed lexis and grammar are frequent, often to the extent of using terms and grammatical forms unrecognised in other native-speaking regions.

There is, naturally, no assertion here that all members of all classes use language in the ways mapped out in this table.  After all, the ways one speaks are not written into law and there is no reason why people cannot decide to take on a different dialect and a different accent to suit themselves and their aspirations.
The point is, however, that this is the way to bet.

Frequent large studies of the use of language analysed by social class have detected significant differences between people's use of language related to their social class.  For example, in English:

This list could be greatly extended and will vary, except for upper-class accents, depending on regional differences.  The general rule is, however, plain: the lower down the social ladder people are, the more likely they are to use regionally affected accents and non-standard grammatical and lexical forms.

It has been noted in many studies, incidentally, that women are much less likely to use regional dialect forms than men.
For a little more on why this may be, see the guide to Sex, below.

We have chosen examples here of the use of English because that is this site's main concern but other language and speech communities will exhibit similar phenomena (although the details will vary, of course).
In languages such as Japanese, which have complex honorific structures, differences in address patterns will be instantly recognisable and based on perceived status and power differentials.  In that language, differences in how a person is referred to or addressed will vary from adults of equal status, people of higher status (including customers), people of junior status, small children, senior colleagues or authority figures such as teachers, doctors and lawyers and so on.
In other languages, such as Tamil and other Indian languages which developed in a society divided into various castes, there may be differences in the pronoun system and in the forms of the language used between and among people of different castes.  Tamil, for example, has very different variants and forms will vary depending on whether the language is being used formally, informally or officially and much of that depends on real or perceived class differences.
In European languages, similar class-based differences are discernible with a tendency, for example, in French for upper-class speakers to prefer the vous-form over tu-forms more frequently than those in lower classes, even among close friends and relations.  In German, there are noticeable class differences in the use of certain case forms for some prepositions and the list could go on.
The fact that class-based variation exists in English will not, therefore, come a surprise to most learners of English.


Implications for the classroom

It may not be surprising to most learners that class-based differences in language use exist but recognising them will be very challenging.  Native speakers of all languages can make quite accurate decisions concerning people's class and background from the way they speak and can often assign people to certain classes and occupational types based on very little linguistic information.  From that information, they can adjust their own speech and the topics they choose to discuss to match the perceived status, backgrounds and interests of others around them.  They may even feign membership of classes other than their own by a deliberate (or semi-conscious) variation in how they speak to others.
Many native speakers can also switch codes easily and use, for example, a local accent and dialect in some social settings (in order to emphasise that they belong to a speech community) and avoid such uses in other, usually more formal, settings in which the use of something approximating to standard dialect forms and pronunciation is valued.
Non-native speakers, even advanced users of a second language, have no such skills for the most part and will depend less on how something is said because they are focusing on what is said.  They will usually be wholly unable to switch codes, having learnt only one dialect and one accent.
In other words, subtle and often unconscious decisions about others based on how they use language cannot be transferred to a second language because the signals are completely different.
If we wish, therefore, to equip learners with some even very basic sociolinguistic skills, the area has to be taught.

A more contentious implication concerns the materials which teachers use in classrooms and the signals they send.  Coursebook and materials writers are usually concerned to present language which is:

  1. a standard variety which has wide acceptance in English-speaking societies worldwide.  This usually means either standard American or standard British English.
    In some circumstances, e.g., in the design of materials for teaching immigrant populations, other regional variations may be the targets but the focus will still be on the most widely accepted 'correct' usage.
  2. presented in a recognisable and authentic social setting.  This may mean considering a wide range of roles and occupations for characters in narratives and dialogues making sure that overt class, sex, ethnic or religious bias is minimised.  Most modern materials are carefully scrutinised for implicit biases of this sort.

Unfortunately, a. and b. may not be compatible aims.
If a materials writer presents language spoken by people in the target-language culture from a range of occupational, educational, and social backgrounds, then those depicted as being from lower classes will not authentically be using a standard form or dialect.  Survey after survey and investigation after investigation have shown that people from lower social classes will almost always be using non-standard dialect forms and non-standard pronunciation.  Any hoped-for authenticity in teaching materials is fundamentally undermined, therefore, if such people are depicted or heard using high-status, standard language forms.
Published materials will, consequently, either be:

  1. inauthentically depicting lower-class workers and families or upper- and middle-class people using the 'wrong' dialect and accent
  2. presenting authentic language which is unacceptable as a model because it is too divergent from a standard form of the language and provides a non-portable example of the language most learners are concerned to learn.

Some attempts in some materials have been made to square the circle and provide examples of people using a range of regional or national varieties but the accents used in these cases will not usually be particularly strong (and are sometimes fake) with few deviations from the standard which most learners will be indifferent to even if they can discern them.
Language which deviates from accepted structural norms will very rarely appear and dialect words which have little range socially or geographically will also be avoided.

This contradiction may not matter too much and is certainly irrelevant for those learners who are unlikely ever to encounter people speaking non-standard forms of the language in a native-speaker environment but for other learners who may wish to travel widely in a native-speaking country and may work with people who speak with a genuinely regional or class accent it is important.  When they encounter, for example:
    It were him what said it
    We was having dinner with me mam's friends
it will be somewhat easier for them to follow a conversation if they have been alerted to possible variations in dialect and accent.
If published course and supplementary materials do not, for commercial if no other reasons, provide this exposure, it devolves on teachers to fill the gap.



Ethnicity and nationality

Language, nationality and ethnicity are independent concepts.  No person's ethnicity or nationality determines the language they speak and no person's language assigns them to an ethnic group or a nationality.
It is, for example, not a safe bet to assume that someone born in Argentina speaks Spanish or someone who speaks Spanish as a first language was born in Spain.
It is, accordingly, nonsense to refer to Germans as if this is the same as German speakers or to Dutch speakers as if they all come from The Netherlands.
All that said, there are, of course, connections between nationality, ethnicity and language but they are connections, not determining factors.

We noted above that there are no identifiable wholly monoglot nations on earth and even very small nations often have separate speech communities within their borders.  The same is, of course, true of ethnic groupings.  Some ethnic groupings are very large, straddling national borders and occupying large areas of the earth's surface; others are small and constitute only a part of other nations.
Ethnic groupings are also blurred at the edges and it is almost impossible to define, for example, what exactly constitutes vague groupings such as Indian British or New York Jewish.  Almost any individual one tries to place in a grouping like that is likely to exhibit characteristics which do not fit the stereotype.
However, for the purposes of sociolinguistic analysis, it sometimes makes sense to operate with vague terms like these to see if there are identifiable language-use characteristics which can be assigned to the influence of ethnicity or nationality.



In the USA, around 430 languages are spoken (many indigenous) and in the UK, around 300 languages are spoken.  Other countries around the world exhibit similar (if not such extreme) characteristics.  Germany, for example, is home to at least twelve indigenous languages and another 20 or so are widely spoken by immigrant populations.  France is also home to over 20 indigenous languages and has significant populations of immigrants speaking a further dozen or so.
On the other hand, English is spoken as an official language or widely understood and used in over 50 countries around the world, Spanish is an official language in 20 countries and a significant language in another handful and French is spoken in both American continents, throughout western, central and northern Africa as well as in Indo-China and elsewhere in Europe (including, of course, significant dialects in Belgium and Switzerland).  Arabic is an official language in 26 countries and a recognised one in many more (although the dialects of the spoken language are not always mutually comprehensible).  Other major languages show similar spread although Japanese is nearly confined to the nation of Japan, within which there are, incidentally, a wide range of regional dialects, some so different from standard Japanese that they are barely comprehensible to speakers of other Japonic languages and dialects.
Russia has around two dozen indigenous languages and Russian speakers also form large minorities in at least a dozen other countries.
Greek speakers often refer to anyone whose first language is Greek as Greek by nationality although the reality may be very different.  Greek itself may be divided into nine major dialects and Greece is also home to speakers of around 10 other languages.
Even Australia, a notoriously monoglot country, contains significant minorities of speakers of ten other languages and at least 15 indigenous languages are still spoken (down from 400).
Chinese is a catchall term for a group of at least a dozen mutually incomprehensible languages (in their spoken form) and other large languages such as Malay are also better defined as macro-languages containing many mutually incomprehensible dialects.
This list could be greatly extended.
It is not, therefore, sensible ever to equate someone's nationality with the language they speak.

Nevertheless, in most of these countries, there is a clear tendency for the major languages to be spoken by almost everyone with other languages often referred to as additional or minority languages.  In the UK, for example, over 98% of people speak English but that figure is only 75% for the USA.



In sociolinguistic terms, ethnicity is usually considered in the ways it affects the dialects or accents of a major language so, for example, in Britain, studies have been made of what is very loosely termed black English.  In fact, of the roughly 3% of British people who self-define as black, origins are variable and concern for example, immigrants or their descendants from the Caribbean, Africa, and southern Asia.
At one time, it was felt that black British could be precisely identified from its pronunciation, lexicon and grammar.  Many aspects of the dialect are clearly influenced by Jamaican English and other forms of English spoken around the world but some are unique to the dialect.  As is the way of things, the dialect, now called Multi-Ethnic English, is spoken by people from a variety of ethnic backgrounds most of whom reside in London.
The dialect is recognisable phonemically with many changes to standard English (RP) pronunciations of, for example, vowels and /h/ dropping as well as differences in the pronunciation of some consonants.
It is also recognisable from having taken on some aspects of London English, including but not limited to using, e.g.:
    We was coming
    He were here earlier

and so on.
Lexically, there are many differences with over 25 identifiable adjectives not found in standard English including greezy (bad), hench (strong, fit) and deep (unfortunate or serious).  There are also many nouns which are unique to the dialect including blem (cigarette), ends (neighbourhood) and yard (house or dwelling).  Verbs, too, have been identified and they include pree (stare), gas (lie) and jack (steal).
Even pronouns differ so we find man (first or second  person), dem man (they) and us man (we).

Other examples of ethnically influenced dialects of standard forms of languages exist around the world so the French spoken in some parts of southern France in areas of high rates of immigration from Africa is strongly influenced by African languages and in the USA there are identifiable dialects within dialects spoken by people from a range of ethnic backgrounds including Jews, Italians, Latin Americans, Germans and Scandinavians.


Implications for the classroom

The implications for the classroom are as for the considerations of social-class accents and dialects (see above) and include the need to be aware of the range of differences as well as to avoid assuming that everyone in a particular country speaks the same kind of language or even the same language as everyone else.

Furthermore, we need to get away from assuming that learners' nationalities will reliably tell us about the language they speak.  That may be the way to bet, of course, but not all bets win.

Whether learners of English need to be exposed to or made aware of the variations in dialect and accent which are influenced by ethnicity is somewhat controversial.
Undoubtedly, for certain groups, for example learners who are immigrants in the areas of a country where there are strong ethnically-based dialects and accents, an understanding of the varieties is quite important.  For others, whose contact with such varieties is likely to be very minimal or wholly absent, no such considerations apply.
As usual, it is the learners' needs which constitute the defining variable.




Language does not vary on purely personal grounds concerning where people have been brought up, how they have been educated, where they live, what their ethnic background is and to which social class they belong.  It also varies in terms of the contexts in which it is used.
Here, we shall not be using the tern register as a sloppy synonym for style.  Style concerns levels of formality and, while it is clear that certain social settings often require certain levels of formality to be maintained and others allow very informal language use, these are not the defining characteristics.  What defines the language use in terms of social setting is register.
Here are some examples of the registers which may affect, in particular, lexical choices and grammatical structures, notably the former:

Legal register
will often use terms which are not easily understood by people outside the speech community such as mens rea, tort or habeas corpus.  Such terms are often derived from Latin but other terms such as occupier, assault, causation and so on are defined very precisely in ways that non-members of the speech community would not immediately recognise.
Legal register will also affect grammatical choices with a general avoidance of the use of the first person and quite complex nominalisation in, for example, contractual agreements.  The use of cohesive markers such as hereinafter, heretofore, notwithstanding and so on are also common.
Medical register
is, in many ways similar, with thousands of terms barely or not understood by non-professionals.  Indeed, medical practitioners sometimes receive targeted training in translating medical register into language comprehensible to their patients.
The use of certain formulaic grammatical structures is also noticeable such as The patient presented with ... and so on.
Academic fields
all exhibit specific uses of lexemes which are often unknown or very differently defined outside the field.  A non-specialist would be very challenged, for example, if asked to participate in a conference of specialists in most academic environments.
Indeed, the distinction between popular-science publications and texts written for fellow specialists often lies in the need to define and even avoid terms rather than in the assumption that the reader will already share register-specific knowledge.
(On the Revolutions of the Celestial Spheres, by Nicolaus Copernicus (1473 – 1543) has the forbidding inscription "For mathematicians only".)
English Language Teaching register
is, as a cursory glance at the guides on this site reveals, replete with terms used technically rather in their everyday meanings such as use, usage, function, structure, reference, tense, aspect and so on.  There are also many terms which occur not at all or very rarely outside the register such as deixis, noncompositionality, idiomaticity, interlanguage etc.
In common with other humanity subjects, too, speaking and writing in the field often requires slightly unusual (for some) uses of modal expressions, hedging and reporting verbs.
Games and sports
are notorious for having words and structures unknown outside the field of interest so, for example, the distinctions between court, field, pitch and ground are subtle and professionals will often use terms which even enthusiasts in the areas will not immediately recognise or use.
The use of the present simple in sports commentaries is also indicative of a register-specific use of tense and aspect.
Institutional settings
often exhibit certain language forms, particular in lexis, which are not in widespread, or any, use outside that setting.  For example, in certain schools, and even year groups or classes within a school, a set of often short-hand terms has developed which are only used in that setting by the people affiliated with the institution or a specific part of it.  The same considerations apply to colleges, universities, departments within them and other educational settings.
Companies and other commercial or non-commercial organisations also exhibit a parallel phenomenon so the use of language within even quite small organisations is often, at least to some extent, unique to that setting.  Anyone who has joined a large organisation and been forced very rapidly to come to terms with internal jargon, often acronyms or initialisms, has experienced the need to acquire a setting-specific register.

This list could go on for some time because there is almost no limit to the number of registers which we can identify as having particular lexicons and structural uses.


Implications for the classroom

Implications are obvious because of the range of language that its use in specific registers may influence.
In general terms, register affects:



There are, as Hymes (in Pride and Holmes, 1971:278) pointed out, rules of use without which the rules of grammar would be useless.
Unfortunately, the rules, such as they are, are culturally variable and operate differently across languages and dialects of languages.
This is a huge research field which is frequently revisited so we can only scratch the surface here.  Some examples will have to suffice concerning the ways that cultures and other social factors affect the rules of use.  Some of these are trivial issues, taught in seconds, but others are subtle and very complex.
For more, see the guides listed below.

In many Anglophone countries it is conventional not to interrupt a turn but to wait for a signal that the speaker is giving up a turn or for a suitable point in the exchange, often called a transition-relevance point which allows one participant to take over or start a new turn.
In other cultures, however, interruption is seen as a normal part of any interaction and will not be viewed as discourteous or unconventional.
There is evidence that within Anglophone cultures, silence on behalf of a participant in a conversation signals the need for someone else to take a turn and is not tolerated for long.  In other cultures, such as those containing speakers of North American Athabaskan or Dene speakers (which includes, e.g., Navajo) silence betokens no such thing and speakers of these languages remain silent without this causing any feelings of discomfort.
Encounters between English speakers and Athabaskan speakers may result, as Trudgill (2000:132) has suggested in
    Athabaskans ... thinking that English speakers are rude, dominating, superior, smug and self-centred
and English speakers finding
    Athabaskans rude, superior, surly, taciturn and withdrawn
Applying the rules of one's own culture when speaking a different language, especially to a native speaker, or to people who speak a different dialect of the same language is, therefore, fraught with danger.
In some dialects of English, notably, for example, New York Jewish, continuous interruption is considered a sign of interest being maintained and not at all discourteous.
In many cultures, it is considered unusual if listeners do not, from time to time, signal the fact that they are listening and understanding by providing audible or visible signals to that effect.
In others, however, back-channelling noises and interjections are interpretable as interruptions and frowned upon.  In these cultures, notable in east and south-east Asia, listening in silence is considered polite and deferential.
Silence in those cultures which rely on back-channelling will soon make a speaker feel uncomfortable.
expressing opinion
In many cultures, one is free to express an opinion in many circumstances and nobody will consider doing so a sign of arrogance or lack of respect.
In others, personal opinion statements are handled with much greater care and often couched in non-personal terms so, instead of:
    I think this is a poor idea
speakers will prefer something like
    There are those who may not like this idea.
answering the telephone
This is a trivial issue but indicative of cultural differences and the dangers of ignoring them.
In many cultures, it is possible to have an exchange such as:
    Hi, can I speak to Mary?

but in others, such as Francophone and German-speaking cultures, that would be wildly unconventional and the conversation would normally proceed as:
    Is that 66655789?
    This is Annie.  I'm sorry to disturb you.  Is Mary there?

which is far more deferential.
In Anglophone cultures, in other words, there is no obligation to identify oneself unless the need arises so the first conversation may continue:
    No, I'm afraid she's out.
    Oh, could you tell her than Annie called, please?
In almost all cultures, incidentally, children are deliberately taught conventional ways of starting telephone conversations and the training embeds the behaviour very firmly.
Anglophone cultures are notorious for the ways in which speakers apologise when there is no need to, so for example, we may encounter:
    I'm sorry.  Can you tell me which bus goes to Sloane Square?
    Sorry, can I get through here, please?
Such conversations would sound very odd in most other languages and most other dialects.
This is, of course, a teachable area but routines like this are learned at a very early age and overcoming a tendency not to apologise in another language or to apologise when speaking English takes a real effort of will.
Gricean maxims
The four maxims identified by Grice (1989) are
  1. Quality
    Do not say something for which you have no evidence
    Do not say something which you know to be false
  2. Quantity
    Be informative enough but do not over inform
  3. Relation
    Maintain relevance
  4. Manner
    Avoid obscurity and ambiguity
    Be orderly and brief
and breaking any of the maxims is seen as a sign of some deliberate marking or emphasis so, for example, saying
    I'm getting rather tired
is, on the face of things, over informing, but signals that the illocutionary force of what is being said does not refer just to tiredness but signals a request to stop.
Equally, breaking into a conversation about organising a party with something like:
    Oh, have you seen Fred recently?
may seem to be breaking the maxim of relation but signals some important relevance for the speaker that the listener needs to note.
However, such maxims are based on English speakers and it would be perilous to assume either that they are applicable to all languages, cultures, nationalities and ethnic groups or that the breaking of a maxim will always be a signal of some form of communicative marking.
restricted and elaborated codes
The concept of restricted and elaborated codes we owe to Bernstein (1971).  He identified two varieties of language available to its speakers:
  1. Elaborated code is used most frequently in formal situations and contains:
    1. a high proportion of subordinate clauses, using, for example, conjunctions such as although, so, because, in order to and conjuncts such as accordingly, subsequently, therefore etc.
      The use of such devices is generally terms hypotaxis (from the Greek meaning to arrange below).
    2. passive verb clauses using, for example:
          She was arrested
      instead of
          The police took her away
    3. longer and more unusual adjective phrases such as the ancient, British people, a difficult but important issue etc.
    4. less common adverbs, adverb phrases and adverbials such as regrettably, to my astonishment, not wholly surprisingly, without much difficulty etc.
    5. the use of abstract nominalisation such as creation, emergency, issues, calculations and so on.
  2. Restricted code is used in less formal settings and is characterised by:
    1. The use of simpler, and often repeated coordinating rather than subordinating conjunctions such as and, then, but etc and conjuncts such as first, next, later etc.
      This is often referred to as the use of parataxis (from the Greek meaning to arrange side by side).
    2. The avoidance of passive structures and a consequent increase in the use of subject pronouns such as he, she, they etc.
    3. The use of tag questions to elicit agreement such as wouldn't they?, didn't we? etc.
    4. Simple adjectives and the avoidance of long adjective phrases so we find more occurrences of, e.g., nice, lovely, pretty, good, bad etc. rather than pleasing, beautiful, stunning, acceptable, disrespectful etc.
    5. the use of a greater number of concrete nominalisations such as road, car, house, room etc.
More controversially, Bernstein and others have proposed that the ability to use either code depending on the appropriacy of the setting and the needs of the moment is restricted to middle-class children whereas working-class children are confined in general to the use of restricted code and are therefore educationally and career-wise disadvantaged because elaborated code structures and lexis are required to encode complex relationships.  In that view, working-class children suffer from cognitive deficiency because they are unable to think about more abstract and difficult ideas.
This is a view most strongly held by those who believe that thought is dependent on language but on that hypothesis, the jury is still considering its verdict.
There is a guide to language and thought, linked below.
There are two competing views of the way in which the use or non-use of restricted and elaborated codes affects educational outcomes:
  1. Elaborated code is an essential requirement of the educational process which is concerned with the ability to express universal meanings and abstract concepts.  So, non-access to elaborated code means:
    1. cognitive deficiency
    2. inability to handle abstract thought and universalistic concepts
    3. these students fail more often
  2. The use of elaborated code is simply social convention which is conventionally required in educational settings which means that
    1. teachers consider elaborated code appropriate
    2. students who don’t / can’t use elaborated code are discriminated against by teachers
    3. these students fail more often
    This is the view held by, e.g., Labov, who presented evidence that even in restricted code the encoding of complex and subtle ideas is perfectly possible, indeed, routinely achieved and no cognitive deficiency can be identified.  In this view, elaborated code use has nothing to do with restrictions in cognitive abilities.
It will not have escaped your notice the the characteristics of restricted code are often noticeable in the speech of learners of English.

This list can be extended and more information is provided in the guides linked below.


Implications for the classroom

The list above is enough, hopefully, to alert you to the fact that the conventions of social interactions and settings vary across languages, dialects, cultures and ethnic groups.  This means:

and more.




In some languages, but not English, men and women use different dialects.  In extreme cases, this may affect more than just pronunciation, which is often quite variable in all languages.
In Thai, for example, the first person pronoun, I, is different for men and women so a man refers to himself as phom and a woman as dichan.
Other differences have been noted by comparative linguists but they usually concern small and somewhat unusual languages.
However, we noted above, when discussing regional and class accents, women are substantially less likely to use dialect forms which depend on region or social class, or a combination of those factors.  Trudgill sums up the situation like this:

In all cases so far examined, it has been shown that, allowing for other factors such as social class, ethnic group and age, women consistently use forms which more closely approach the standard variety or the prestige accent than those used by men.

He backs this up with figures from studies carried out in the USA showing that, for example, the use of a double negative such as
    I don't have none
is used by 90.1% of lower-working-class men but by only 58.9%v of women from the same group.
Nobody really knows why this is the case but there are some interesting and persuasive hypotheses:

  1. Women are more status-conscious than men and are thus make an effort, especially when speaking to strangers, to use a higher-prestige dialect and accent.
  2. Women are more sensitive to dialect and accent than men and more able to switch easily between codes, as they might when taking part in a survey of dialects and accents.
  3. Working-class speech has connotations of masculinity and toughness to which men aspire more than women and so they are more likely to use the forms.
  4. Men are more inclined to associate themselves with others of their class and may show a greater tendency to use dialect forms and a non-standard accent to show their class solidarity.

The assumption is that in societies in which sex roles are very clearly demarcated, the tendency for men and women to use different dialects and accents will be stronger than in societies where roles are more fluid and less rigidly assigned.


Implications for the classroom

Because most Anglophone cultures are relatively less strict in assigning social role to sex, the case in English is just one of general tendencies and certainly not something from which one can extract a rule.  However, one implication is that if we wish to demonstrate a strong regional accent or dialect, we are probably better off choosing to do so in the speech of working class men than women.  Conversely, if we wish authentically to present examples of high-class prestige varieties, we are probably better advised to use the speech of women than men.
There is a guide on this site, linked below, to how English and other languages handle gender-specific terms.


Here's a summary but it does not show apart from some deliberate overlapping how the influences may act together in complex ways.
influences summary

Related guides
variety this guide focuses particularly on varieties of English including artificial ones such as English as Lingua Franca
gender for the guide to how English and a range of other languages mark gender
turn taking for the guides to areas which all draw on sociolinguistic theory
adjacency pairs
spoken discourse for the guide which considers how interaction is achieved in spoken language
pragmatics for consideration of an area of language study which draws heavily on sociolinguistic themes
the roots of English as a language develops, varieties alter with it
types of languages for considerations of how other factors such as word ordering and transitivity vary between languages
language and thought for a guide to how and whether language is the stuff of thought
discourse index for the in-service index to the area
skills index for the in-service index to the area

Selected bibliography
Some of the following have been used in the writing of this guide.  Others you may find helpful:
Allan, K & Jaszczolt, KM, (Eds.), 2012, The Cambridge Handbook of Pragmatics, Cambridge: Cambridge University Press
Bayley, R, Cameron, R, and Lucas, C, (Eds.), 2013, The Oxford Handbook of Sociolinguistics, Oxford: Oxford University Press.
Bell, A, 2013, The guidebook to sociolinguistics, Malden, MA: Wiley-Blackwell
Bernstein, B, 1971, Class, Codes and Control, Volume I, London & New York: Routledge, Taylor & Francis Group
Cameron, R and Fuller, JM, 2021, Sociolinguistics, Oxford: Oxford University Press
Chambers, JK, 2009, Sociolinguistic Theory: Linguistic Variation and Its Social Significance, Malden, MA: Wiley Blackwell
Grice, HP, 1989, Studies in the Way of Words, Harvard: Harvard University Press
Griffiths, P, 2006, An Introduction to English Semantics and Pragmatics, Edinburgh: Edinburgh University Press
https://www.ethnologue.com/guides/how-many-languages for the estimate of the number of living human languages
Hymes, D, 1971, On communicative competence, in Pride, J. & Holmes, J, (Eds.), 1971, p 278, Sociolinguistics, London: Penguin
Kachru, B, 1985, Standards, codification and sociolinguistic realism: English language in the outer circle, in Quirk, R and Widdowson, H, (Eds.), English in the world: Teaching and learning the language and literatures, Cambridge: Cambridge University Press
Labov, W, 2010, Principles of Linguistic Change, Malden, MA: Wiley Blackwell
Meyerhoff, M, 2019, Introducing sociolinguistics, 3rd edition, New York: Routledge
Sealey, A & Carter, B, 2004, Applied Linguistics as Social Science, London and New York: Continuum
Trudgill, P, 2000, Sociolinguistics: An Introduction to Language and Society, London: Penguin Books
van Herk, G, 2017, What is sociolinguistics? 2nd edition, Hoboken, NJ: Wiley-Blackwell
Walker, JA, 2010, Variation in linguistic systems, London and New York: Routledge.
Wardhaugh, R, and Fuller, JM, 2021, An introduction to sociolinguistics, 8th edition, Hoboken, NJ: Wiley-Blackwell