Language variation and change

ANT/LIN/SOC 506
FEB 1
Weinreich, Labov and Herzog (1968) “Empirical foundations for a theory
of language change”

Five issues crucial to understanding language change:

• ACTUATION: “What factors can account for the actuation of changes? Why do changes in a
structural feature take place in a particular language at a given time, but not in other languages
with the same feature, or in the same language at other times? This actuation problem can be
regarded as the very heart of the matter. It is thus apparent that we want a theory of language
change to deal with nothing less than the manner in which the linguistic structure of a complex
community is transformed in the course of time so that, in some sense, both the language and
the community remain the same, but the language acquires a different form.”
• CONTSTRAINTS: What are the constraints on the transition from one state of a language to an
immediately succeeding state? How can we determine a possible change in a language?
• TRANSMISSION: How are changes transmitted from one individual to another? How does a
language change spread through a speech community?
• EMBEDDING: How does a change related to other aspects of grammar? How does a change alter
the overall structure of a language? How are changes related to prior (or future) changes?
• EVALUATION: How do speaker attitudes towards variation influence language change? How does
the social evaluation of forms of variation contribute to the spread of language change?
(un)conditioned sound change
Unconditioned sound changes apply to every instance of a sound,
regardless of where it falls in a word.

Conditioned sound changes apply only to sounds in specific

phonological contexts [ = where it occurs in terms of sound]

Conditioned change: example

[ɛ]  [ɪ]/ [nasal consonant] (m, n, ng)
pen > pin, meant > mint, ten > tin, hem > him
Conditioned sound change: [front vowel] > [ɛ]/ [r]
Merger of (front) vowels before /r/:

Example: Merger of vowels before nasals:

“pit” pɪt “pin” pɪn

“pet” pɛt “pen” pɛn  pɪn

Conditioned sound change: [front vowel] > [ɛ]/ [r]
Merger of (front) vowels before /r/:

ey BAIT

ɛ BET ə BUT

æ BAT
Sound change:
Merger of (front) vowels before /r/:

ey Mary

ɛ merry

æ marry
 Sound change:

Merger of (front) vowels before /r/:

ey BEAR
ə
ɛ BERRY ə BURR

æ BARE
actuation:

Merger of (front) vowels before /r/:

TRANSMISSION: How are changes transmitted from one individual to
another? How does alanguage change spread through a speech
community?

EVALUATION: How do speaker attitudes towards variation influence

language change? How does the social evaluation of forms of variation
contribute to the spread of language change?
 Observer’s paradox
People adjust their speech when they know they are
being listened to by strangers.

Sociolinguistic interviews will cause people to adjust

their speech.

How to observe people speaking when they don’t

know they are being observed?

9/23/2019
 Collecting data
for sociolinguistic
analysis
Overcoming the observer’s paradox
• Ask people to tell an emotional story so that they
will be distracted from the context (“Danger of
Death” stories – Labov)
• Record your own friends to make the situation less
strange – (so many studies of Thanksgiving dinners)
• Throw away the first so-many minutes of the
recording (assumes people become more relaxed
over time)
• Record now, ask later (unlikely to pass the IRB)
• Analyze publicly available data
• Hope the interview is interrupted by someone close
so that the speaker shifts their speech to match
someone more familiar
(Labovian) linguistic variable – a language
structure (sound, word or sentence structure)
with multiple forms with the same referential
meaning but different indexical meanings.

C~Ø/C # (test ~ tes) “test”

r~Ø/ {C, #} (mowər ~ mow) “more”
ð~d (ðis ~ dis) “this”
Labov’s position on language variation and change
 Speakers vary the frequency with which they use linguistic variables.

 Different frequencies are associated with different social identities.

 The most important aspect of ‘identity’ in cases of language change
is social class.

 Speakers alter these frequencies depending on context.

 Variation across contexts is primarily due to attention paid to speech

Linguistic variable –

some linguistic variables are stable for long periods of

time (e.g. in vs. ing).

other variables are part of language change in progress

Language variation and change

Example:
d/t ~ ð, ɵ

dis vs. this

Tupac v. Biggie

Biggie: d/t = 336, ð, ɵ = 27 (92.5% d/t)

Tupac: d/t = 11, ð, ɵ = 395 (97.4% ð, ɵ)

Note: Tupac’s pattern is more typical (compare d/t for:

Jay-Z =13%, Dr Dre=13%, Outkast=17%, Snoop=8%)
Post-vocalic /r/ = /r/ before a consonant or
at the end of a word

Kanye West

Jay-Z
 Language variation and change
Social class is the central variable in Labovian paradigm.

Wolfram (1969): Variation across social classes

Five phonological variables:

1. Word-final consonant cluster reducation

(eg. des ~ desk)
2. Interdental fricatives (non-initial ɵ ~ t, f, ø)
3. post-vocalic /r/ (r ~ ø in coda position)
4. word-final alveolar stop devoicing (d ~ t, Ɂ)
5. word-final alveolar stop deletion (d ~ ø)
 Contextual styles
• Casual speech – This was when the person was speaking to their family or
friends, or when they were telling a very emotional story (“danger of
death”)
• Careful speech – This was when the person was interviewed by a linguist in
their home.
• Reading passage – The person was asked to read a passage containing the
variable.
• Reading list – A list of words, some with the variable, some without
• D’. Minimal pairs – The person was asked to read a list of words that would
make the variable more obvious because they only different in terms of /r/:

bead/beard, bee/beer, etc.

 Labov – Isolating contextual styles (NYC)

Casual speech =

Speech on the phone with someone known

Speech to other individuals in the household
Speech when answering the “Danger of Death” question?

Was there ever a time when you thought, “This is it, I’m going to die”
 Labov – Isolating contextual styles (NYC)

Careful speech =

Speech directed towards the interviewer

a stranger who is observing your speech

 Labov – Isolating contextual styles (NYC)

Reading – a text
including minimal pairs
that only differ in terms of
one of the variables being
investigated (e.g.
sauce/source)
Labov – Isolating contextual styles (NYC)

Word list – a text

including word with one
of the variables being
investigated (e.g.
sauce/source)

Comparing tense vs. lax low front vowels

Labov – Isolating contextual styles (NYC)

Minimal pairs – word

lists with pairs who
differ only in the
variable in question
Social stratification in NYC
Labov devised a method for distinguishing between the speech of social
groups across various contexts (and dealing with the observer’s paradox)

• Casual speech – This was when the person was speaking to their family or
friends, or when they were telling a very emotional story (“danger of
death”)
• Careful speech – This was when the person was interviewed by a linguist in
their home.
• Reading passage – The person was asked to read a passage containing the
variable.
• Reading list – A list of words, some with the variable, some without
• Minimal pairs – The person was asked to read a list of words that would
make the variable more obvious because they only different in terms of /r/:
bead/beard, bee/beer, etc.
Social stratification in NYC (th ~ d/t)
Social stratification or /r/ in NYC
Social stratification or /r/ in NYC

Hypercorrection
Social stratification in NYC
Hypercorrection – overuse of a linguistic variable when attempting to speak in the
manner of another social group.

Style (for Labov) = attention paid to speech

Labov (1972) concluded that speakers move towards the standard as they paid
more attention to their speech.

Middle-class linguistic insecurity – pressure from classes above/below lead middle-

class to break the pattern.

Sound change spreads through hypercorrection.

Hypercorrection – overuse of a linguistic
variable when attempting to speak in the
manner of another social group.

Labov (1972) concluded that speakers

move towards the standard as they paid
more attention to their speech.

In a study of hip-hop artists, Alim (2002)

compared speech in interviews with the
speech of performances to compare
“attention paid to speech”:
Attention paid to speech
In a study of hip-hop artists, Alim (2002) compared speech in
interviews with the speech of performances, showing that “attention
paid to speech” may lead to movement away from the standard as
well:

Interview Lyrics
Juvenille 56.6% 75.0%
Eve 5.95% 56.7% Frequency of null copula
Variation and language change
“Apparent time” – different use of variables across generations reflects
change in progress

“Real time” – data from two points of time are used to test apparent time
hypothesis:
panel study – same participants examined at two points in time
trend study – similar set of participants but new research subjects
 Language change

Apparent time hypothesis (Labov)

Differences in the use of variables
across generations
reflects change in progress
 Language change

Patterns of usage across time

show the same “s-curve”

Change may spread through a

community across a single
generation

Change may also spread through

the movement of speakers

Ex: Changes runneth  runs

ye  you
Age-grading:
Variation that correlates with age, but does not change over time.

Typically, age grading show higher amounts of non-standard variables

before ~ 25 and after ~ 64 (= U-curve)
Variation and language change
“Apparent time” – different use of variables across generations reflects
change in progress

Limited data sets may make it difficult to discern between change in

apparent time and patterns of age-grading

Age-grading – speakers use particular frequencies at specific ages and the

age-related pattern is stable across time.

“S-curve” vs. “U-curve”

Change across the lifespan
Long-term studies using speech samples from the same person at
different times.

Do adults shift their speech? (e.g. Queen Elizabeth)

Changes in the length of 2Pac’s /æ/ vowel over time
Changes in the length of 2Pac’s /æ/ vowel over time
Covert
prestige

Trudgill (1988) – gender differences explained by covert/overt prestige

Social class and gender

Labov (1991) – gender differences are tied to types of language

variation – women are more conservative with stable variables (not
undergoing change), men are more conservative with variables
reflecting change in progress

Eckert (2001) – gender differences require more complex analyses

based on ethnographic awareness of the local context
Eckert (2001) – study of language use in a high school
in Novi, MI:

Primary social identities: “jocks” vs. “burnouts”

Jocks: students who gained power and status through

school-sponsored activities and who identified with the
school.

Burnouts: Students who gained power and status

through activities in the “real world” rather than in
school and who identified with the local community.
Social class and gender (Eckert)

Jocks (suburban Burnouts (urban

orientation) orientation)
Girls Being popular and Verbal and
having a good emotional (rather
personality than physical)
toughness
Boys athletic, academic, “toughness” as tied
or activity (choir, to use of controlled
band, school govt.) substances,
achievement encounters with the
law and physical
fights
Social class and gender

Raising of /aj/ diphthong (Canadian Raising)

Burnouts Jocks
Girls .79 .38
Boys .50 .28

Raising of /æ/ (“Dad” moves towards “dead”)

Burnouts Jocks
Girls .61 .53
Boys .38 .46
Language variation and change

Next stop – the fourth floor