How old is language?

This morning I listened to the first in a new series of a BBC Radio 4 programme called Fry’s English Delight in which the Stephen Fry discusses the development, design and functions of the mouth. According to an evolutionary biologist interviewed on the programme, the human mouth attained it’s modern form between about 80 and 50 thousand years ago, and this allowed our ancestors to make the full range of sounds used in languages. These dates also fit with some estimates of when first language emerged.

They also mention that art and related activities also start to appear in the archaeological record roughly between those dates. This is not conclusive evidence that language is between 50 and 80 thousand years old – there is some evidence of possibly older artist activity, but is very interesting nonetheless.

Language evolution

Some interesting experiments on language evolution are being undertaken in the University of Edinburgh’s Language Evolution and Computation Research Unit, and one thing they’ve found is that some aspects of language can develop in an afternoon.

They believe that language evolves culturally through being learned and used by people. They have demonstrated aspects of this process with computer simulations and with an experiment with real people. For the experiment they used pictures of alien fruit with names in a made up language which the participants were asked to memorise. They were then tested on what they could remember and their answers were used with the second group of participants, and so on.

The first participants found it very difficult to learn and remember the words, but with each subsequent ‘generation’ it became easier to learn them and they developed regularities in their structure, and eventually the participants were able to understand words they’d never seen before.

The researchers believe that many aspects of languages can arise through the evolutionary process of cultural transmission and do not need to be genetically encoded – the brain provides scaffolding for language but not necessarily all the specific details.

Language mainly a cultural phenomenon

A report I found the other day suggests that language is most likely to be mainly a cultural phenomenon and that any genetic underpinnings for language probably pre-date the emergence of human language.

Researchers in the UK and USA modelled how aspects of language might have been encoded genetically and concluded that this was very unlikely to have occured given the way human culture changes. For linguistic traits to become encoded in the genes they have to provide a selective advantage and there has to be a stable language envirnoment. Human languages change too quickly for this to happen, and modern humans haven’t existed for long enough either. Therefore they believe that language evolved culturally and that the existence of a genetic language module is unlikely.

They also argue that if human populations in different parts of the world had evolved separate, incompatible language modules, they wouldn’t be able to learn one another’s languages. This is obviously not the case.

I can hear you

A report I found today in Science News suggests that early hominids had hearing capabilities similar to modern humans, and paleoanthropologists at the American Natural History Museum in New York believe that this could indicate that they had some form of language.

Analysis and reconstruction of the auditory bones in skills of Homo heidelbergensis dating from 530,000 years ago have demonstrated that their hearing was probably similar to that of modern humans. They could hear best between 2 khz and 4 khz, the frequency range within which much of the sound of speech is transmitted, and researchers believe that such an ability must have been used, as maintaining such sensory systems is neurologically very expensive and they are unlikely to evolve and not be used. Whether they were connected with speech or something else is not known.

Language Evolution

Language Evolution

Yesterday I finished reading Language Evolution by Morten H. Christiansen and Simon Kirby. The book contains 17 chapters written by scholars from a range of fields, including archaeology, biology, cognitive science, linguistics, neuroscience and psychology, and discusses the latest theories and current controversies in the field of language origins and evolution.

It’s very interesting and I’d certainly recommend it. However, quite a few of the chapters contradict previous ones or seek to prove them wrong, and by the time I’d finished it, I wasn’t at all sure which of many theories to believe.

They’re definitely a lot more sophisticated than the old bow-wow, ding-dong and yo-he-ho theories, which suggest that we got language by imitating animal noises or other natural sounds; that language began as instinctive responses to stimuli; or that it began as a way to facilitate cooperative labour.

Language evolution

According to an article I found today, languages tend to evolve in short bursts. The happens when groups of people coin lots of new words to describe their world.

Researchers have used computer programs normally used to study biologically evolution to study the development of basic vocabulary in 490 languages in Europe, Asia and Africa. They found that many new words appear in a short burst over a few generations when a new language starts to develop.

The new words could arise as way of differentiating new languages from related languages, and also to reinforce group identity. The researchers also suggest that small groups sometimes develop now forms of language based on linguistic idiosyncrasies of their founder members.

Language change

According to a New Scientist article I came across the other day, frequently-used words tend to be more resistant to change then words that are used less often.

A team at the University of Reading lead by Mark Pagel, an Evolutionary Biologist, compared the words used to express 200 different meanings in 87 different Indo-European languages. They found that the more frequently a word is used in speech, the less likely it is to change over time. They also found the conjunctions and prepositions tend to change more readily than numbers, pronouns and question words like who, what, where, etc. The team calculated a ‘mutation rate’ for each of the words the studied and predicted that frequently-used words are likely to resist change for over 10,000 years.

Another study at Harvard University demonstrated that the most frequently-used English irregular verbs have tended to remain stable over time, while most of the least frequently-used ones have become regular.

Tones and genes

According to an article in the New Scientist, researchers at the University of Edinburgh have demonstrated using statistical analysis that two genes, ASPM and Microcephalin, that govern aspects of brain development tend to differ between regions where tonal languages are spoken, and regions where non-tonal languages are spoken.

The article also mentions that there are some differences in brain structure between English speakers with facility for learning tonal languages and those who find such languages difficult. So if you are struggling with the tones of a language like Mandarin or Thai, maybe it’s because your brain has evolved to cope best with non-tonal languages.

Another article on this subject in Scientific American gives more details of the research:

Ladd and Dediu compared 24 linguistic features — such as subject-verb word order, passive tense, and rounded vowels — with 981 versions of the two genes found in the 49 populations studied. Most of the language contrasts could be explained by geographic or historical differences. But tone seemed to be inextricably tied to the variations of ASPM and Microcephalin observed by the authors. The mutations were absent in populations that speak tonal languages, but abundant in nontonal speakers.

Further details are available on this blog, which is written by the son of one of the researchers.