Is urbanization correlated with language innovation?

Question

In Brazil, the Portuguese dialects spoken in rural areas preserve, despite their own innovations, several features of the language that were common in the 16th century. This phenomenon is particularly evident in the so called “caipira dialect”. These conservative features include phonological, lexical and semantic aspects, and are considered archaisms in the modern standard dialect. On the other hand, a number of innovations in Brazilian Portuguese originated in the more urbanized areas (southeast region, especially São Paulo and Rio de Janeiro) and then spread to the rest of the country.

On another geographic scale, Brazilian Portuguese as a whole is considered rather conservative, when compared to European Portuguese. Historically, until the 19th century, Brazil remained as basically an agrarian economy, while Portugal was the urbanized colonial center.

Evidently, this is not a scientific observation (so it may be strongly biased), but it seems that urbanization and the rate of language change are somehow correlated. Is that a fact?

Answer 1

Urbanization and language change are, no doubt, somehow correlated; the question is whether change spreads faster/slower in urbanized areas or small communities. Both arguments have been proposed. Here is a (very brief) summary of the main points in Crowley and Bowern 2010 (13.3.4 Variation in small communities). On the one hand, it's easier to enforce language norms in small communities. On the other hand, social networks are denser in small societies (i.e. change will spread easier/faster), not in towns. Thus, Crowley and Bowern, while discussing language change in small communities in Melanesia and Aboriginal Australia, conclude that "linguists do not yet know the answer."

Also, social prestige is a very ambiguous term in the sense of what is considered prestigious is different for different social groups. Living in town is not universally "desirable" in all societies.

William Labov is the "god" of sociolinguistics. You should read his works on this problem.

Answer 2

I'll answer according to a theory of language change mostly associated with Croft, according to which change comes from intra-speaker variation. In an evolutionary model of language change, changes may be propagated via neutral selection (linguistic variants randomly come in and out of existence, but more frequent ones should be the most durable) and via replicator selection (where variants produced by certain prestigious groups people have a better chance of enduring).

We expect innovation (upheaval in customary forms) to happen more frequently when there is greater variation in the community, and when there are groups of people with whom prestige is associated. An urban setting doesn't necessarily have to have both of these, things but they tend to. We should expect innovation in urban settings especially when (i) the residents are coming in from various parts of the country where different varieties are spoken (more variation), (ii) within the urban center itself, there are people who are considered prestigious, and whose speech tends to get disproportionately emulated. If these two conditions are met (and additionally we assume relatively free interaction between speakers), then, according to the theory, we should expect the relatively rapid emergence of a new urban speech variety.

So I think the answer to the question is that this is a valid assumption, since the factors encouraging language change tend to be very frequently present when new urban centers form.

Some downloadable papers:
Croft (2010): The Origins of grammaticalization in the verbalization of experience

Baxter et al (2009): Modeling language change: An evaluation of Trudgill's theory of the development of New Zealand English

Answer 3

Urban language change may be distinctive due to different patterns of language acquisition. If in hunter-gatherer and agrarian societies children take parents/grandparents as their linguistic models right through the acquisition process, then this would produce a pattern of language change in which slow, steady incremental change would dominate. (Though this basic pattern would be later subject to processes of linguistic diffusion through contact with other groups.)

In urban environments, however, there are larger groups of children and adolescents living in the same settlement. In some cases they may take their age-peers as linguistic models from a relatively young age, which might be expected to accelerate changes in process.

On change by incrementation and diffusion, see: Labov (2007), “Transmission and Diffusion.”

For an example of urban language change breaking with the normal incrementation pattern, because of different socialisation patterns, see: Cheshire et al. (2011), "Contact, the feature pool and the speech community: The emergence of Multicultural London English"