Language & Society 2019

• In many approaches to language structure, grammars are treated as categorical, invariant and discrete
• Sociolinguists, however, treat grammars as variable and fluid and continuous
• Many formal approaches, as we saw, treat ‘performance’ as of little importance; sociolinguists have demonstrated three things about variation in performance that make it worthy of linguistic investigation.

• Two or more ways of saying the same thing
• Meaning does not change (despite different realizations)
• Linguistic variables can be phonetic/phonological, lexical, morphological, grammatical…

(t) – ‘butter’

[bʌtə] versus [bʌʔə] versus [bʌɾə]

( )  phonetic/phonological variables (idea)

[ ]   variants of variables (actually being said)

1. Variation is functional

2. Variation is a necessary precondition of language change.

3. Variation is highly structured, both in the speech of individuals and in the speech of communities.

investigating the ways in which people choose different linguistic structures for communicative reasons

• obvious on pragmatic level, e.g. politeness 
  Can you..? vs. Would you be so kind..?
   
• There is social meaning attached to the variants 
  formality, poshness, coolness, etc 
  ideological/ indexical field (difference in social meaning, identity marking)

• Linguistic change is universal
• Easy to spot when comparing language to its previous forms
• More or less consistent for a certain time
• Variation is the mechanism through which changes take place
• This mechanism is a gradual replacement, often over decades or centuries, of one feature by another: the old and the new coexist
• Assumption: Most changes begin as sporadic unconscious innovations 
• S-Curve Adoption Model

• Language variation isn’t ‘free’, ‘haphazard’ but structured & orderly. 
• Highly structured, both in the speech of individuals and the speech of communities

It is structured by:

1. Speakers’ social characteristics

2. The audience’s social characteristics

3. Setting and topic

4. Psycholinguistic phenomena (attention, processing, etc.)

5. Language and conversation structure

• identifying social factors that seem to correlate with variable linguistic usage, for example that women use certain variants more than men, and so on.
• sociolinguists have demonstrated how linguistic variation is shaped by the linguistic context in which the variable is found. These linguistic contexts that shape variation are called linguistic constraints.

• Constraint means constraining, influencing, shaping, encouraging, preventing
• A constraint on variation is a factor which influences the proportions of variants found for a particular linguistic variable

• Linguistic differences are analysed in terms of the importance of the legitimised language in the socio-economic life of the speaker.
• Some people have a greater stake in speaking the 'legitimised language' than others. This is not necessarily correlated with social class.
• Sankoff / Thibault: avoir as auxhiliary for all verbs 

Gal: German-Hungarian. Constraints: social class (peasant or not), age (young or old)

once upon a time: ->  only used by peasants , now used by everyone (younger speakers)

Bell: radio announcers under different social contexts

Coupland: Study of travel agency employes (different language among different costumers): /h/ dropping, /-t and d/ deletion, /VtV/

Rickford / McNair-Knox: 'Foxy', more AAVE features with same (as her) skinned interviewer than with different-skinned one

Baker: different social situations/activities influence /t/ glottalling

Seriality

If the first adjective in a turn shows marking, to what extent do subsequent adjectives in that turn also show marking? And if the first adjective does not show marking?

Dative alternation: 

V + DO + to IO

e.g. He gave some food to the hungry cats.

V + IO + DO

e.g. He gave the hungry cats some food.

Probably have the strongest constraining effects on language variation, more important than social and other factors…

Weinreich, Labov and Herzog (1968):

“Linguists are naturally suspicious of any account of change which fails to show the influence of the structural environment upon the feature in question… detailed studies of intimate co-variation among linguistic variables in process of change provide the most persuasive empirical evidence of such systematic effects” (1968: 172)

• observed the deletion of –t/-d in consonant clusters 
• Analysis of different ethnicities, but all working class

Data Set 1:

best friend

cold weather 

Data Set 2:

he stuffed the turkey

she seemed funny 

Data Set 3:

most of the time

ground attack 

Data Set 4:

he seemed odd 

she passed a test

Syllable stress

(ING): [ɪŋ], [ɪn]

• I am swimming a lot these days - verb

• That boring man is here again - adjective

• Her swimming is much better now - noun

• This morning was very busy – non-morphemic

• Have got is favoured* with concrete complements (i.e. ‘have got’ will occur at higher levels where the complement is concrete than when it is abstract - I have got a shop > I have got a problem)
• Personal pronouns prefer have got (I‘ve got time v the boys have got time)
• Have favoured with generic subjects (Every kid has one v He has one)

• he’s not feeling very well: AUX
• he isn’t feeling very well: NEG
• he ain’t feeling very well: SEC
• She’s not found it: AUX
• She hasn’t found it: NEG
• She ain’t found it: SEC
• BE: AUX contraction most common he’s not feeling very well: AUX >
• he isn’t feeling very well: NEG
• Preceding vowels > consonants favour AUX contraction (cf he‘s not v it‘s not)
• HAVE: NEG contraction most common
• She hasn’t found it: NEG > 
• She’s not found it: AUX

I said ‚that‘s strange‘

I went ‚that‘s strange‘

I thought ‚that‘s strange‘ I‘m like ‚that‘s strange‘ I‘m all ‚that‘s strange‘

I‘m ‚that‘s strange‘

I shouted ‚that‘s strange‘....

• Baker examind t glottaling in turn-final and non-turn final tokens in a middle class brownie leader (girl scouts) from Essex.
• The word 'right' was examined as discourse marker and as non-discourse marker
• Right as a discourse marker was less frequently pronounced with a glottal stop
• Can function as a discourse marker as well as adjective, noun etc
• She also distinguished between different kinds of ‘right’ as discourse markers:

• • Perspective maintenance right 

  • Perspective altering right

• I will do my marking tonight
• I‘m going to do my marking tonight I gonna do my marking tonight

Going to prefers:

• Past point of reference: I knew it was gonna hurt
• Subordinate clause: He told me he was going to sing
• Non-1st person subjects: he‘s going to do my marking
• Not verbs of movement: he‘s going to sing
• Immediate proximity: he‘s going to leave in one minute

As talkers, we engage routinely in conversational management, involving the management of turn-taking, turn relinquishing, story telling, topic shifting, and so on

Does conversational management structure shape language variation?

(ing): [ŋg] – hopping, skipping and jumping 

HappY Tensing – merry, happy, lively

New Zealand English –t/d deletion 

% deletion given a following...:

Obstruent/nasal 56% 

Liquid 55% 

Glide 53% 

Vowel 26% 

Pause 17% 

(N=3706)

1. ‘Turn-finality’ appears to behave rather prominently as a linguistic constraint, very often appearing as one or the context in which one variant of a variable is found most (or least);

2. More often than not, turn final contexts encourage a less lenited, less reduced, often more standard variant of consonant variables and very often avoid deletion more than other contexts

• Variability is highly shaped (but not necessarily blocked) by a number of factors;
• What appears to be messy and variable on the surface hides a governing and underlying set of ‘contraints’, factors which hinder or encourage particular variable linguistic forms to occur;
• These constraints can be linguistic (and these are usually the strongest), but also social, contextual, psycholinguistic, interactional, etc.
• For many variables some of the linguistic constraints appear universal (e.g. the greater likelihood of non- standard [in] for (ing) appearing in verbal forms than noun forms).

...Orderly heterogeneity

• interest in how you should speak
• Set of externally set rules that speakers should follow
• Grammar/dictionary writers, government institutions (media, education, …), employers, parents, ‘complaint tradition’

• interest in how people DO speak
• Historical reasons: e.g. the development of certain ‘Englishes’
• Linguistic reasons: all varieties have fully functioning grammars acquired by native speakers

• One sees a set of rules imposed on usage from outside, say, by some authority on correctness…
• One sees complex and abstract human system, not imposed by overt prescription…
• former learnt, latter acquired. The latter is in place before former is taught

• Our frameworks of analysis conform to the views that we have learnt at school rather than to the variety we acquired at home and use in face to face interaction. 
•  developing a sense of what was ‘right’ and ‘wrong’ as well as what is and is not appropriate in the written mode 
• We became programmed there, to see language prescriptively, to prioritise the written over the spoken, and to see fundamental language structure as conforming to the rules of written language (cf. data transcription).
• Sociolinguists have emphasized a very clear distinction between standard and non-standard forms of the language, concentrating on the non-standard and the spoken

… in many ways artificial

… Socially and historically created

… Derived from national elites and promoted through institutions of government and power

… Codified in attempt to eradicate variability

… Often held as a symbol of national unity, distinctiveness, prowess and prestige

… Considered, because of formal exposure to it, to be correct and proper, and the representation of a language (end result of norm enforcement)

… Trigger a complaint tradition

… A symbol of political power (e.g. Yugoslavia)

… Linguistically odd (suppressing language variation)

… Most tidy in a written form, but writing is learnt not acquired. Many languages aren’t written.

… lack native speakers as they are ideological

… often fossilizing rather strange forms

… oppressing non-standard varieties

Continued use of standard language in formal, institutional contexts by people with power and authority leads speakers to: 

• combine use of standard with non-standard forms in their speech 
• absorb attitudes of linguistic inferiority about the non-standard form 
• assume the standard form denotes correctness, civilization, education, authority, truth etc...
• internalize ‘standard ideologies’. 

• “humans are normative”
• “people attempt to ‘clean up’ language and make its structure or its use conform more closely to their ideas of beauty, truth, efficiency, logic, correctness and civility” -> may express deeper anxieties
• Highlights the meddling in language engaged in by prescriptivists
• linguists are also engaging in verbal hygiene (e.g. language planning & management: saving Gaelic)

• Campaigning for the use of plain English
• Dialect/minority language preservation societies
• Taking courses in assertiveness or better conversational English
• Editing text for a newspaper
• Language planning (e.g. Cornish)
• National Curriculum for English (however dialect-friendly)

Guidelines for non-sexist language àPC