How and Why Language Changes

How and Why Language Changes

This post was written by Mark Davies, HC Fellow, Linguistics Department

Why do languages change? The answers that some linguists tended to give 100-150 years ago strike us as being quite absurd nowadays. For example, they sometimes looked to the physical environment as a motivation for language change, such as the fact that the Germanic peoples in the Alps in 2000-3000 BC huffed and puffed so much as they were going up and down the mountains that they turned the “stops” (p, t, k) from Proto-Indo-European (spoken about 3000-5000 BC) into “fricatives” (f, th, h; a change known as Grimms Law), as in [p]a[t]er > [f]a[th]er, [t]res > [th]ree, [c]ornu (copia); “horn of plenty”) > [h]orn, etc. No one would suggest that as a motivation nowadays.

During the last 100 years or so, there have been two major camps when it comes to language change. The first are “structuralists” and “typologists”, who see internal motivations for change in the language. The other are “sociolinguists”, who see external / social motivations for change.

As noted, structuralists emphasize the role of competing factors in a language as a motivation for change. To give a concrete example, standard Spanish distinguishes between the two verb forms () tiene/s/ (“you have”; informal) and (usted) tiene/-/ (“you have”; more formal). In most varieties of Spanish, it’s not necessary to use a subject pronoun ( / usted), because the verb ending already indicates who the subject is. But in some varieties of Spanish like Puerto Rico, the final /s/ is often lost, leading to the same form for both informal and formal: tiene/-/. In these varieties, the subject pronoun is almost obligatory (as it is in English), to compensate for the loss of the verb ending, and to let us know who we’re talking about. (Something similar happened in English between about 1000 and 1500 AD).

The preceding is an example of the interplay between phonology (sounds), morphology (word forms), and syntax. Sometimes there is also interplay between semantics (meaning) and word forms. For example, if someone says “My wife is so hot; I need to get her a drink”, there is an awkward ambiguity between the two meanings of hot (at least since the early 1990s). While such ambiguity can be tolerated for some time, there are often limits, such as the ambiguity of gay (= “happy”, or sexual orientation) being resolved with the loss of gay = “happy” in the 1950s-1970s.

As mentioned, the second major group of historical linguists focus on social motivations for change. This is something that really began to be studied systematically in the 1960s, and it was pioneered by the sociolinguist William Labov. To give a concrete example, Labov noticed that some people on the island of Martha’s Vineyard pronounced words like light and house with an intervening “schwa” sound, as in /l-uh-ite/ and /h-uh-oose/. It turns out that this was a “phonetic marker” that showed that these speakers identified with Martha’s Vineyard as it used to be, before the influx of rich “out-of-towners” from the mainland, who were running up the price of land. And when he studied this phenomenon in the early 1960s, such use appeared to be on the increase, as more and more people were being squeezed out of their ancestral farmlands on Martha’s Vineyard. Since the 1960s, sociolinguists and historical linguists have found hundreds of other interesting examples of how group identification can affect the adoption (or non-adoption) of linguistic features, much the same way that groups of people adopt new styles in clothing or hairstyles.

The two schools of “structurally-motivated change” and “socially-motivated change” might seem to be at odds with each other. In the first case, the “invisible hand” of language is moving things along to keep language understandable and workable. In the second case, change seems to be as random and unpredictable as the most recent changes in fashion (bell-bottoms in the 1960s, preppy clothes in the 1980s, or the Kardashians in the 2010s; ouch!). But careful research has shown that often there is interesting interplay between the two types of motivations. The invisible hand of language change (structurally-motivated change) can create an environment for change, which is then helped along by a certain social group adopting that change, and using it as a social marker.

I’m a historical linguist, and I’m fascinated by how and why language changes. But I’m also a corpus linguist, meaning that I use large collections of texts to see what’s going on in the language. I’ve created some of the largest historical corpora of English, such as the Corpus of Historical American English (400 million words, 1810s-2000s) and the Corpus of Contemporary American English (520 million words, 1990-2015). Both of these corpora (and many more; see can be used to look at language change 15 or 150 years ago.

But I’m also fascinated by what’s going on in the language right now – this week; this month – and how that affects (and is affected by) longer term changes in the language, which often take place over decades and even centuries. To look at changes right now, I’ve created a corpus called NOW (News on the Web; It grows by about 5-6 million words each day (or about 150 million words each month, or 1.8 billion words each year). With this corpus, researchers can look at the creation and spread of new words (Brexit, gig economy, normcore, precariat, etc) and see how current events are affecting the language. Or they can look at syntactic change (e.g. the ever-increasing rise in the like construction, “and he’s like the most awesome guy I’ve like ever gone out with, like for real”), morphological change (new words with the “scandal” suffix gate, e.g. deflategate, or Russiagate), or semantic change (new uses for existing words, e.g. trigger warning, green tech or astroturfing).

By spending time “in the trenches” with the language – day by day – we can gain insight into longer-range trends and changes. And in turn, this allows us to amass large amounts of data that can help to answer basic questions about how and why languages change.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>