Chapter 10: Language Variation and Change

10.3 Language changes

Language is constantly changing. Speakers of English today do not speak like the authors of Beowulf (c. 700 CE) or The Canterbury Tales (c. 1400 CE) or Hamlet (c. 1600 CE) just in the same way that speakers of Japanese today do not speak like the authors of the Kojiki (c. 700 CE) or the Genji Monogatari (c. 1000 CE). In some ways, English and Japanese speakers today do not even speak the same way that people spoke English or Japanese a century ago or even just a few decades ago. English, Japanese, and really all languages have changed and continue to evolve.

Language change is important for variationist sociolinguistics because language variation will always be present during language change. It’s not like one day in the early sixteenth century all English speakers woke up and went “hey, you know what? I think I’ll start putting my negative marker BEFORE my verbs like they do in Danish embedded clauses instead of after like they do in Icelandic embedded clauses!” Rather, the linguistic change from move not to do not move happened gradually. Over time people began using the new do not VERB option more and more and using the old VERB not option less and less. During this period both options were possible – the two options were variants of a linguistic variable. Sometimes we have stable variation where two or more variants are present but one isn’t replacing the other. So while not all examples of linguistic variation involve language change in progress, all examples of language change in progress involve a period of sociolinguistic variation. Studying changes in progress is sociolinguistically informative because changes in progress guarantee the presence of linguistic variables. But linguistic change is interesting in its own right because language change is also intimately linked with social factors and with social change.

Analyzing language change. Perhaps the most obvious way to analyze a linguistic change is to consider language use at one period of time and compare it to language use at a different period of time. If we notice differences in the frequency of use of variants of a linguistic variable between the earlier data and the more recent data, this is a good indication that a change has taken place or is taking place. This approach, examining data that represent the same community at two different times, is called real time analysis. This approach is great when we have older data available to us. But what about when we don’t? Good news: There’s still a rigorous way to analyze change in data that comes from a single time period! We can compare older and younger people! This is called apparent time analysis and it rests on the observation that individuals’ grammars stabilize in late adolescence. This means that (typically) we use language in basically the same way we did when we were about 18. We can certainly learn new words after this age, and we might adjust some aspects of our grammar in the direction of the community we live in, but by and large, the patterning of linguistic variables we had at 18 will stick with us through our lifespan. By considering the pattern of linguistic variables in the language use of people of different ages, we can make inferences about linguistic change.

In addition to the distinction between stable variation and language change, sociolinguists also distinguish two kinds of language change. Changes from above are linguistic changes that take place above the level of social awareness (i.e., language users are aware of them). A change from above typically takes the form of the adoption of a prestigious or standardized variant from outside of the community. A classic example of a change from above is the importation of ‘r-fulness’ to New York City English (Becker 2014). From the 18th century into the early 20th century, NYC English was generally r-less. Words like cart and star would have standardly been pronounced something like [kʰɒət] and [stɒə]. However, by the middle of the 20th century, the norms of General American English, including its r-fullness, began to influence New Yorkers’ speech. The new, prestigious r-full variant (like [kʰɒɹt] and [stɒɹ]) began to compete with the older (and increasingly stigmatized) r-less variant, slowly spreading and advancing through the community. On the other hand, changes from below are changes that represent the operation of articulatory or grammatical pressures within a linguistic system that people are generally not aware of. For example, in Canadian English the vowel in the word goose, which would be transcribed as the high, back, rounded vowel [u] in a dictionary, has been gradually moving toward the front of the vowel space to something more like [ʉ] or even [y]. Chances are, any given speaker of Canadian English would be unaware that their goose vowel is more front than older Canadians’ goose vowel!

Just as all languages exhibit variation, all languages also change over time. Because change involves variation, variationist sociolinguists often examine changes in progress in addition to stable variation.


Becker, K. (2014). Linguistic repertoire and ethnic identity in New York City. Language & Communication35, 43-54.


Icon for the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

Essentials of Linguistics, 2nd edition Copyright © 2022 by Catherine Anderson; Bronwyn Bjorkman; Derek Denis; Julianne Doner; Margaret Grant; Nathan Sanders; and Ai Taniguchi is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.

Share This Book