Language in Society

Historical Linguistics and Language Families 1 listed social prestige among the causes of language change, with the loss of the /r/ in Received Pronunciation as its example. Phonetics and the IPA 7 then showed that English has several accents, and that an accent can carry a social meaning. This course takes that social side seriously. The question it asks is not "what is the rule?" but "who says what, to whom, and why?"

What Sociolinguistics Studies

The field combines dialectology with the social sciences to identify regional dialects, sociolects, ethnolects and other styles. It studies variation between social groups and under cultural norms, expectations and contexts, including how variation contributes to language change. The social variables the source names are ethnicity, religion, status, gender, education and age, as well as geographic barriers.

Variation: More Than One Way of Saying It

Wikipedia's article on linguistic variation defines it as the feature of language where "there is more than one way of saying the same thing in a given language". It can occur in pronunciation, vocabulary and grammar.

• Inter-speaker variation: different speakers choose differently.
• Intra-speaker variation: the same speaker chooses differently on different occasions.
• The page uses variable for Labov's sociolinguistic variables and variants for the competing forms. It does not give a formal definition of either, so this course uses them as the page does.
• Some variation is part of language change, while other competing forms can persist in "stable variation" for a long time.
• Variation is bounded: speakers across dialects tend to keep the same word order and fit new sounds into existing phoneme inventories.
• Social factors (class, region, gender, age) and formality (register) condition variation.

Examples of variables

Age gives a clear case. The same page reports that in 1979 about two-thirds of 12-year-olds in Toronto said "zee" for the letter Z, against 8% of adults, and that in 1991 39% of 20-25-year-olds and 12% of those over 30 used "zee". The page hypothesises that an American children's alphabet song influenced this pattern. That is a hypothesis in the source, not an established cause.

How Researchers Collect and Count It

• The sociolinguistic interview: a long, loosely structured conversation meant to elicit the vernacular style. The observer's paradox complicates it: the speech being studied is what would occur without the interviewer present. Techniques include eliciting narratives of memorable events and interviewing subjects in groups.
• Variationist analysis: quantitative study of variation and change, associated with Labov, which assesses social and linguistic variables in a speech corpus. Labov's ideal variable is high in frequency, resistant to conscious suppression, part of larger structures and easily quantified on a linear scale. Analysis counts the tokens of a variant against the number of times it could have occurred (Tagliamonte's "Principle of Accountability").
• Other methods: ethnography and participant observation, analysis of recordings, matched-guise tests, dialect surveys, and analysis of existing corpora.

A Short History

• Social aspects of language were first studied by Indian and Japanese linguists in the 1930s, with forerunners in Denmark and Switzerland around 1900.
• Thomas Callan Hodson's 1939 article "Sociolinguistics in India" is the first attested use of the term.
• Charles A. Ferguson and William Labov (US) and Basil Bernstein (UK) pioneered Western sociolinguistics. Labov is regarded as the founder of variationist sociolinguistics.
• Labov's 1963 paper "The social motivation of a sound change" helped establish sociolinguistics as a subfield. The Martha's Vineyard page describes his 1963 study as "widely recognized as a seminal work in the foundation of sociolinguistics".

The Founding Example: Martha's Vineyard

Labov wrote his master's thesis on changes in the dialect of Martha's Vineyard, an island off the coast of Massachusetts. The Wikipedia pages I read give only that outline. The details below come from a LINGUIST List announcement of Meredith Josey's 2004 NYU dissertation, which summarises Labov's study before reporting her own data, so they are a second-hand account:

The lesson is the one the field is built on: a small sound difference was not random. It marked who the speakers took themselves to be.

Try It Yourself