The main purpose of the book is to dispose of the theory that human language is rooted in a universal grammar, itself rooted in genes; in its short hand, language as instinct as opposed to language as use. A theory said still to be advocated by the very eminent Messrs. Chomsky, Fodor and Pinker, amongst others. So eminent, that endorsement from any one of them is more or less sure to get one's research grant or PhD waved through...
A book which is, I think, directed at both a general audience and those with a more serious, professional, not to say professorial interest. But perhaps with the first audience in mind, Evans has adopted a rather populist style, which I found very irritating. Rather jokey but rather rudely adversarial at the same time. He shows little respect for or courtesy towards his opponents. He has sexed up a large proportion of his examples, with much talk of supermodels and kissing. Maybe it is OK to talk like this in the seminar rooms of Bangor, but I like my printed word to be in a more sober & neutral style.
Disposition is achieved disposing of six subordinate, albeit interconnected, theories.
One: language is not related to the communication systems of other animals. Observations: some non-human animals do communicate with each other and even more with humans. Some such animals can even talk about things which are not there, not in the present. Some of them can lie. So false.
Two: there are language universals. Observations: this despite the diversity & irregularity of grammar exhibited across the world, never mind that of vocabulary and sound. The evidence does not support either of the following propositions:
There are universal ingredients and universal rules for combining ingredients, words, into sentences. Around 20% of languages, for example, manage without word order - although this 20% can be a little misleading in that it does not copy to 20% of humans and I have no idea whether this last is too big or too small.
There are universal ingredients and universal rules for combining ingredients, sounds, into words.
There always seem to be at least some exceptions to any proposed universal rule. So not proven.
Three: language is innate. Observations: along the way he knocks out a favourite biblical phrase of mine ‘in the beginning was the word’, heard in a church no less than 36 hours ago. It seems reasonably clear that things, deeds and the corresponding concepts came into mind before the words we then used to describe them. It also seems reasonably clear that young humans learn language from those around them over a period of years. For ordinary mortals at least, no those around them then no language. So false.
Four: language is a module. Observations: a module would need to satisfy three conditions, none of which hold up too well in the light of what we know today, what has been learned in, say, the last thirty years:
A specialised part of the brain, doing just language. The search for bits of brains doing just language has not been very successful. Brain processing now seems to be distributed across brains, with any one function of interest appearing to involve lots of different regions.
Dissociation: people exist with language but not cognition and people exist with cognition but no language.
Distinctive development trajectory. But the learning of vocabulary – clearly not instinctive – and grammar seem to be mixed up together. And the learning if both is continuous; changes in speed perhaps, but not jumps.
So not a module in anything like the way of a module in a computer system.
Five: there is a universal in-brain language, here dubbed ‘Mentalese’. Observations: how can we disentangle symbols of mentalese from the cloud of images needed to give them meaning? Clouds of images which are the result of experience. (Aside: we often talk of abstracts in terms of concrete analogs: so we talk of love in terms of journeys). A sentence in any language can be well formed, but still meaningless when the meanings of the individual words are taken into account. But this sort of information is not going to be coded in, by genes or otherwise, at birth. And how does does mentalese deal with the very different strategies for doing things in languages across the world?
There is interesting talk of the way in which words activate bits of brains. So, for example, a good word might activate good emotions, a bad word bad emotions. A motor word like 'running' might activate motor pathways, sometimes to the point where there is twitching where the motor action might be. I wonder whether people differ in this regard. So perhaps some people can process language, as if it were a computer or mathematical language, just symbols to be manipulated. While other people are much more rooted in reality and have difficulty generating language about things of which they have no experience. Or which they dislike.
I associate to that fact, which interests me from time to time, that children can talk about something, reasonably sensibly, just on the basis of learning the appropriate language of discourse, without having much, if any, real life experience of the subject of that discourse. I guess this is possible because so much knowledge has been coded into that language of discourse that applying generative rules to make conversation generally generates sensible stuff. But I also think that one of Evans's points is that while this may be true, the rules will also generate nonsense. Language which is not rooted in experience is not enough - at least not unless it is about something reasonably ethereal like the groups of mathematics.
So probably false, certainly not proven.
Six: thought is independent of language. Observations Evans describes various experiments which demonstrate interactions between the quirks of particular languages and the behaviour of its speakers. So false.
Then, very impressed to find 27 pages of references at the end of the book. To think that to produce a book like this you have to read & digested, or at the very least to be able to give the appearance of same, of so much stuff. So much time exploring what has gone before which might have been spent in other ways. That is not to say that one shouldn't do it, but I think I would find wading through the treacle of so many others rather tiresome.
In sum, despite the irritations mentioned above, the book is convincing. Language as instinct might have been a reasonable working hypothesis when it was first articulated sixty years ago , but the accumulated weight of evidence dug up since then is against it. But there is then the thought advertised at reference 1: so what? The journey has been interesting but what do we take away from it?
Oddly, amazon thought that I would like to buy a second copy of the book, sending me an advertisement for it, along with a number of others. Including a second book by the same author, a treatment of the treatment of time in languages across the world, so sober that it had been remaindered from £60, which was a bit strong for me, to the £10 or so (plus P&P) which I fell for, having been interested by his treatment of colour in the first book.
With thanks for the illustration of an early version of the modularity of mind myth to wikipedia.
Reference 1: http://psmv2.blogspot.co.uk/2014/12/a-lapse-into-error.html.
No comments:
Post a Comment