William Labov: Orderly Heterogeneity, Not Anecdote
Labov turned “messy speech” into evidence, theory, and a research programme
William Labov is bigger than two famous classroom stories
If you have studied sociolinguistics, you already know the set pieces: Martha’s Vineyard and the New York department stores. They are important and these papers should be read and analysed by anyone interested in language variation. However, Labov did not choose them because they made for memorable fieldwork narratives. He chose them because they let him do something linguistics was not, at the time, prepared to accept.
He treated variation as evidence for linguistic theory, not as “performance noise”. And he pushed a claim that still unsettles tidy models, namely that a a speech community can be systematic because it is variable, not despite it.
That is, to me, is the interesting Labov.
Labov’s wager: stop pretending language is homogeneous
Labov’s foundational move was not the department store staircase. It was the insistence that the “clean” object of analysis (a single, uniform grammar) is often a convenient fiction, and that linguistics pays a price for it when it tries to explain change.
Weinreich, Labov, and Herzog state the point bluntly:
“it will be necessary to learn to see language … as an object possessing orderly heterogeneity.” (Weinreich, Labov and Herzog, 1968)
It says it all! It says:
Variation is structured, not random.
Structure and heterogeneity coexist.
A theory of language that cannot represent this will struggle to explain change.
They also frame the “actuation problem” as the hard one: why this change, in this place, at this time.
Labov spent decades building partial answers, and better ways to test them.
Martha’s Vineyard: not “a sound change story”, but a theory template
Labov opens his 1963 paper by announcing what he thinks the job is:
“the direct observation of a sound change in the context of the community life from which it stems.” (Labov, 1963, p.1)
Even in the first page, the structure of his method is clear. He does not simply “find a correlation”. He sets up a package deal:
A linguistic variable (variants of /ai/ and /au/)
Social distribution (age, occupation, locality)
Reconstructable history (change in apparent time)
A mechanism linking social alignment and phonetic choice
The classroom takeaway often becomes “identity affects pronunciation”. The deeper takeaway is methodological:
Labov shows how we can move from tokens to pattern, and from pattern to an argument about change in progress.
This is why the Vineyard study stays central. It is not a museum piece. It is a template.
The New York department stores: a lesson about style, monitoring, and inference
The department stores study deserves its fame because it does something rare: it makes social stratification audible with a tiny, repeatable design.
But its real contribution is not only “/r/ correlates with class”. It is that Labov demonstrates:
we can sample speech without interviews
we can engineer contexts that shift attention to speech
we can show style-shifting inside the same speaker as evidence that social evaluation is active
This was a step change from the generalised opinion that maybe groups just have their own, immivovable, different, underlying grammars. Labov shows the same person can move, in seconds, when the context changes.
He set a standard as to what what counts as evidence for social evaluation, and what we can infer from a controlled, minimal interaction.
Harlem and African American English: theory, not “deficit correction”
Labov’s work on African American Vernacular English (AAVE) is often taught as a political or educational intervention. It is, but it is more than just that. It is also a theoretical intervention aimed at the core of grammatical analysis.
A good example is his work on variability in the English copula, where he uses formal tools to argue that “absence” patterns are constrained, not sloppy (Labov, 1969).
A copula is a verb (or verb‑like element) whose primary function is to link the subject of a clause to a complement, rather than to express an action. It creates an equation or classification between two elements.
Think of it as the grammatical equals sign.
In English
The main copula is “be” (am, is, are, was, were).
Examples:
The cat is hungry.
Here, is doesn’t describe an action; it links the cat to hungry.
He Ø tall is possible in AAVE, but
He Ø the one is not.
Labov shows that the pattern mirrors the hierarchy of contraction in other English varieties: the copula can be absent only where it could also be contracted (He’s tall → He Ø tall), but not where contraction is impossible (He is the one → *He’s the one → He Ø the one).
This demonstrates that “absence” follows a grammatical constraint system rather than being random or sloppy.
* = gramatically incorrectThis line of research helped reframe “nonstandard” varieties as systems with:
rule-governed distributions
predictable conditioning
learnable constraints
Labov also stepped into public institutions when language ideology caused measurable harm. The 1979 Ann Arbor case is a clean example of linguistic knowledge being treated as relevant to schooling. The judgment warns:
“unless those instructing in reading recognize … the existence of a home language … great harm will be done.” (Martin Luther King Jr. Elementary School Children v. Ann Arbor School District, 1979)
Labov treated linguistic description as accountable to consequences. That stance still divides academics over whether a linguist should stay “descriptive and neutral”, or use descriptive work to argue publicly for policy change (schools, courts, media), even if that pulls you into politics and ideology.
In December 1996, the Oakland Unified School District passed a resolution stating that many African-American pupils’ home speech (“Ebonics”, broadly AAVE) should be recognised as a legitimate language system and used as a bridge to teach Standard English.
It proposed staff training and a contrastive approach (explicitly comparing home speech and classroom English), and it also sought recognition and support under bilingual education funding logic.Labov took the view that mis-describing African American English in schools causes real harm, so linguists should intervene. He backed that kind of intervention publicly, including testifying (Standford, 1998) in support of the Oakland educators’ position at the US Senate hearing on 23 January 1997.
Other academics criticised the Oakland resolution and the wider “linguists as social-change actors” stance, arguing that (a) public advocacy can blur the line between evidence and prescription, and (b) “error correction” does not straightforwardly translate into social change, so the intervention logic can be overconfident.
Change as a system problem: constraints, embedding, evaluation, actuation
A serious Labov article should not stop at “variation correlates with social factors”. Labov’s programme targets a chain of problems that are harder than correlation:
Constraints: what conditions a variant?
Embedding: how does it sit in the wider grammar and the social order?
Evaluation: what do speakers think it signals, consciously or not?
Transition: what are the intermediate stages?
Actuation: why does it start here and now?
This is the reason Labov remains difficult in the best sense. He is not offering a single neat theory. He is building a research programme with multiple moving parts, and demanding that explanations cash out in observable distributions.
Here is an honest challenge for linguists today: actuation. The actuation problem has not gone away. We have better statistics and bigger corpora, but “why this change now” still resists easy answers.
Dialectology at scale: the Atlas and the return of phonology
Labov also changed dialectology by bringing variationist rigour into large-scale mapping and sound system analysis. Sankoff’s overview puts it plainly: his work on American vowels and the Northern Cities Shift “profoundly changed dialectology”, culminating in the Atlas of North American English.
This side of Labov often gets under-taught because it is harder to compress into a lecture anecdote. But it is central to his overall contribution:
change shows up as systemic movement (chain shifts, mergers, splits)
social patterning interacts with phonological structure
dialect geography becomes an engine for theory, not a stamp collection
hat later sociolinguistics kept, and what it pushed back on
A fair reading of the field after Labov looks like this:
What many researchers kept:
the idea that variation is structured
quantitative accountability
the link between variation and change
What many pushed back on:
treating demographic categories as sufficient explanations
under-modelling stance, interaction, ideology, and local practice
Here is the point to hold onto: later “waves” of sociolinguistics did not replace Labov so much as spend Labov’s inheritance in new directions. We cannot do third-wave work responsibly if we think Labov was only “class and /r/”.
The Labov you should remember
If we reduce Labov to Vineyard plus department stores, we miss the thing that made him a problem for linguistics in the first place.
Labov’s core contribution is a discipline-level demand:
treat everyday speech as data worthy of theory
treat variability as part of competence, not an embarrassment
link description to mechanisms that can be tested
accept that explanation must survive contact with distributional facts
Labov died in 2024, but the argument he forced on linguistics is still live.
©Antoine Decressac — 2026.
Labov, W. (1963). The Social Motivation of a Sound Change. WORD, 19(3), 273–309.
Labov, W. (1966). The Social Stratification of English in New York City. Washington, DC: Center for Applied Linguistics.
Labov, W. (1972). Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press.
Labov, W. (1972). Language in the Inner City: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press.
Weinreich, U., Labov, W., and Herzog, M. I. (1968). Empirical foundations for a theory of language change. In W. P. Lehmann and Y. Malkiel (eds.), Directions for Historical Linguistics (pp. 95–195). Austin: University of Texas Press.
Good reads:
Ball, M. J., Mesthrie, R., and Meluzzi, C. (eds.) (2023). The Routledge Handbook of Sociolinguistics Around the World (2nd ed.). London: Routledge.
Beaman, K. V., and Buchstaller, I. (eds.) (2021). Language Variation and Language Change Across the Lifespan: Theoretical and Empirical Perspectives from Panel Studies. London: Routledge.
Kerswill, P., and Wiese, H. (eds.) (2022). Urban Contact Dialects and Language Change: Insights from the Global North and South. London: Routledge.