Re: Any studies on language change rate vs. population density?

From: giaforni
Message: 71361
Date: 2013-10-03

Thanks for your answer. I'm particularly interested in language change in remote prehistoric times (say, 10,000 to 60,000 YBP), which cannot obviously be studied experimentally.

 

However, a reasonable proxy might be a set of studies of language change among hunters-gatherers (say, in South America, Papua, Australia, or the like) - if any such studies have been made.

 

Actually, I've just come across one such study (though rice farmers and slash-and-burn "farmers" are mainly involved): Francois Jacquesson - The speed of language change, typology and history - in Sanchez-Mazas A, Blench R, Ross M, Peiros I, Lin M, eds. (2008) Past human migrations in East Asia: matching archaeology, linguistics and genetics. Routledge Studies in the Early History of Asia, London”. Conclusions on page 307 include the following sentence: "low density languages will change slowly, while high density languages will change far more quickly”

 

Estimates of world population in remote prehistory obviously vary significantly. By some estimates (e.g http://en.wikipedia.org/wiki/World_population, http://www.prb.org/Articles/2002/HowManyPeopleHaveEverLivedonEarth.aspx), world population grew from 1 to 200-300 million (i.e. by two orders of magnitude) between 8’000 BC and 1 AD, whereas it only grew from a few tens of thousands to 1 million (i.e. by one or two orders of magnitude) between 70,000 BC and 10,000 BC.

 

If language change actually was slower when population density was very low, then languages spoken before 10k YBP (ten thousand years before present) must have changed much, much more slowly than in historical times. More precisely: if the conjecture that language change grows with population density is correct, then languages may have changed much more slowly between the first out of Africa migrations (usually placed around 70,000 BC) and 8,000 BC, than they changed from 8,000 BC to present. Intriguingly, current proto-language reconstruction are believed to bridge (most of) the gap  from 8,000 BP to present, i.e. such reconstructions have already recovered the state of proto-languages before rampant changes started. Language change prior to such reconstructed proto-languages was probably so slow, that e.g. PIE would be much closer to its ancestor spoken, say, 35,000 BP, than it is to its currently spoken descendants.

 

Gianfranco Forni

 

---In cybalist@yahoogroups.com, <cybalist@yahoogroups.com> wrote:

This is a hard question to address, since most population expansion and the greatest increases in population density have come in the past century. World population was relatively stable before the invention of the steam engine. In addition to this, it is very difficult to obtain accurate population data before the 19th century in most places.


To make things more difficult, you have the problem of population mobility. This has also been much more common in the past 100 years, and was extremely rare before the 18th century.


The historical evidence suggests the opposite. Language change rate is greater when population is less dense. But this is primarily due to lack of mobility. When two villages are 10 miles apart and it takes an entire day to get to one and back, they will develop differences in language due to isolation.


This is why, historically, diversity within a language or language family is greatest in the area of origin. There is more variation of English dialects within England than in the rest of the world because there have been separate populations of English speakers experiencing language change independent from nearby populations due to lack of interaction.


Historically, languages tend to mix two ways - warfare and commerce. But this doesn't necessarily increase diversity. Warfare often results in the extinction of the languages of the defeated people. The number of languages in currently in use in North and South America and Australia is tiny compared to what there was before European arrival. But we also have historical records of languages like Etruscan being wiped out by Romans over 2000 years ago.


At the same time, though, service in the military of an empire such as Rome or China, was one of the most common ways for a person to travel and interact with people from different parts of the world, importing new vocabulary.


Commerce often results in a koine or creole that combines simplified elements of the languages found along the trade route to facilitate commerce. Also, it may result in a situation where the dominant language of commerce in an area becomes the dominant language in the area. If you combine commercial power with military power, you have today's map, where a few languages are used in vast areas - English, Spanish, Arabic, Mandarin Chinese, and Portuguese are spoken by billions of people. All of these languages, though, incorporate elements from the languages they replaced, and from each other.


Population density does not necessarily mean there is much greater interaction. Historically, class and caste systems greatly limited the types of interactions people were permitted to have with those outside their social group. The most powerful restrictions were placed on women and slaves, both groups often being forbidden from speaking to anyone outside the family they were part of or were owned by.


Even today, a subway train from Flushing Queens to Midtown Manhattan may contain people from 100 countries. But they aren't necessarily talking to each other at all, and when they do, it's probably about dealing with the train they are sharing.


In companies and universities there is generally low interaction between people in different divisions and departments. This produces costs both in duplication of efforts and in lack of opportunities for development. Large corporations have spent much money trying to capture this "institutional knowledge" in a manner that can be accessed by more workers within a company, and universities are opening up "cross-disciplinary" committees all the time.


I think there is an element of truth in what you say, but it is a new issue that is difficult to study, and is a very recent phenomenon. Generally, though, people interact with those near them and those with common interests or goals. And many interactions are limited to specific topics, such as ordering coffee.



---In cybalist@yahoogroups.com, <giaforni@...> wrote:

I suspect there may be a correlation between population density and language change rate. In other words, I suspect that language change might be slow when population density is very low - and vice versa (higher density causes faster changes).

 

This may seem intuitive: the more people you have a chance to bump into, the more language diversity you're likely to be exposed to.

 

However, I have not been able to find any studies that:

1) try to verify whether this correlation exists or not

2) try to measure this correlation quantitatively, in case it does exist

 

If you know any research on this topic, can you please point me to it?

 

Thanks!