The Articulate Mammal (2 page)

The work of Noam Chomsky provided another vital springboard for psycholinguistics, arising initially from his review of a book
Verbal Behaviour (1957)
, whose author B.F. Skinner was a leading behavioural psychologist. Skinner had argued that, just as rats and pigeons could be trained to do a
series of complex tasks by means of ‘operant conditioning’ (trial-and-error learning), so human language learning could be explained in a similar way.

Chomsky wrote a witty and devastating review of this book, pointing out that the behaviour of trained rats is irrelevant to human language. Anyone who made ambitious assertions about language needed to know more about its basic nature, he argued.
The Articulate Mammal
begins (
chapter 1
) by summarizing the key points of Chomsky’s review, and agreeing that he was essentially right when he said that anyone who made strong claims about language needed to understand it.
The Articulate Mammal
tries to do this by explaining both its biological nature (as explained above), and also by summarizing some of Chomsky’s basic ideas.

In the influential ‘classic’ version of his work (1965), Chomsky put forward a new conception of a ‘grammar’. In the past, he observed, linguists had written descriptive grammars, which had tried to describe an accumulation of already uttered sentences. But a person who has acquired a language has not simply memorized past sentences. Instead, he or she has internalized a set of ‘rules’ that allow him/her to produce and understand an indefinite number of novel utterances. Chomsky was interested in the workings of this internalized rule structure. In short, he hoped to encapsulate a speaker’s knowledge of his/her language, rather than just their usage. This knowledge, he suggested (1965), might be captured best by a type of grammar he labelled a ‘transformational’ grammar’, which had two levels of structure, deep and surface.

Chomsky’s ground-breaking transformational grammar began (at that time) with a set of basic phrase structure rules. These outlined the essential underlying sentence structure. Then so-called transformations changed this ‘deep’ structure into the surface structure of a sentence.

Chomsky had promoted a new way of approaching the study of language. Yet some students found his writing difficult to understand. This student reaction to Chomsky prompted a chapter in
The Articulate Mammal
called ‘Celestial unintelligibility: Why do linguists propose such bizarre grammars?’ (
chapter 8
). This chapter, much enjoyed by students, was written as a fairy tale, about a mythical Emperor of Jupiter who became intrigued by the ability of a space-ship full of English speakers to communicate with one another. He arrested their captain, a man called Noam, who clarified how language worked. Noam explained by recounting in a simplified way how he had reached his idea of a transformational grammar. This chapter is still included in the latest edition of
The Articulate Mammal
, though with minor updating, including some extra comments about why transformational grammar has appeared to be abandoned by Chomsky in his latest work.

Chomsky himself has always denied that transformational grammar was related to sentence production. In his view, it encapsulated sentence relatedness, not sentence processing procedures.

Yet to psychologists, this type of grammar appeared (at first) to follow a sequence of steps which, they argued, (wrongly) might be viewed as the way in which speakers prepared a sentence for utterance. The rise and fall of transformations in the minds of psychologists is described in , ‘The white elephant problem’.

In spite of Chomsky’s warnings that transformational grammar was primarily about sentence relatedness, George Miller, a prominent psychologist and a professor at Harvard University, conducted a series of experiments in the 1960s to test the relevance of transformations to speech processing. He explored whether the sequence of rules used in the grammatical derivation of a sentence corresponded to the psychological steps that are executed when a person processed that sentence.

Miller reasoned that if transformations affected processing time, this could be measured. He therefore checked how long it took to match a simple active declarative sentence such as
Joe warned the old woman
with its passive
The old woman was warned by Joe,
or its negative
Joe didn’t warn the old woman,
or, lastly, passive and negative together. Just as he had hoped, he found (initially), that it took twice as long to match a simple sentence such as
Joe warned the old woman
to one which differed by two translations (passive and negative) such as
The old woman wasn’t warned by Joe,
as it did to match a sentence with either a negative or passive alone. He and his fellow psychologists were jubilant, and the so-called ‘correspondence hypothesis’ – the idea that a transformational grammar corresponded to a person’s processing of language was (briefly) enthusiastically embraced.

But then disillusion crept in. Numerous transformations were found which certainly did not take up processing time. In some cases, the version closest to the deep structure took longer to process. For example, a sentence J
ohn runs faster than Bill
took less time to process than a similar meaning sentence which was nearer to the deep structure,
John runs faster than Bill runs
. Eventually, the correspondence hypothesis was abandoned. Reluctantly, psychologists accepted that transformational grammar was not relevant to speech production

Chomsky has changed his mind repeatedly over the way in which linguists should handle language, and in the last two decades of the twentieth century proposed several new versions of transformational grammar, each one more abstract than the last, and (to the average student), harder and harder to understand. His latest version even abandoned just about all transformations! He claimed that he was no longer interested in looking at individual constructions, instead he was trying to find basic laws of nature, the linguistic equivalent of the law of gravity.
The Articulate Mammal
in its latest edition (
chapter 5
) has tried to explain (in outline) Chomsky’s newer views.

But just because Chomsky’s ideas have become more wide-ranging and abstract, this is not necessarily true of the views of all linguists. Others have
argued that language is a complex interweaving of linguistics with other cognitive abilities. The origin of language has become a trendy topic and multiple books are beginning to be published, showing how the various human cognitive abilities are interwoven.

Meanwhile, language processing has not been forgotten, and psycholinguists have continued to explore speech comprehension, as well as speech production. These topics are outlined in and
11
of
The Articulate Mammal
. And a key to much of this is turning out to be the lexicon, the human word store. Humans, it seems, rarely acquire words as single packages, apart from a few names of people and objects. Instead, they often learn them alongside words frequently found with them.
The Articulate Mammal
refers to some of this work, but also refers readers to my own book on the topic:
Words in the Mind: An Introduction to the Mental Lexicon (
3rd edition 2003, 4th edition in preparation), Oxford: Blackwell.

Overall, psycholinguistics is in a healthy state. Study of the biological basis of language is still in a vigorous state of development, both via studies of the brain, and the origin of language. Chomsky is no longer the key force he once was, though it is still important to understand his contribution to linguistics, as it broadened the topic out into a general study of human linguistic ability, and how much of it might be preprogrammed. These days, researchers are moving into even wider questions, trying to understand how our various cognitive abilities link together. The next century will be an exciting one, as all these various strands (hopefully) come together.

P
REFACE TO THE
F
IRST
E
DITION

Some years ago, I gave an evening course entitled ‘Psycholinguistics’. I was quite amazed at the response. A large, eager and intelligent group of people arrived, many of them with a serious reason for wanting to know about the subject. There were speech therapists, infant school teachers, an advertising executive, a librarian, an educational psychologist – to name just a few of those whose jobs I noted. There were also parents interested in understanding how children acquire language, and one student who wanted to know how she might help a relative who had lost her language as a result of a stroke. In addition, there were a number of men and women who said they ‘just wanted to find out more about language’.

The Articulate Mammal
was written for the members of that class, and for others like them: people like me who would like to know why we talk, how we acquire language, and what happens when we produce or comprehend sentences. The book is also intended for students at universities, polytechnics and colleges of education who need an introduction to the subject. It cannot, of course, provide all the answers. But I have tried to set out clearly and briefly what seem to me to have been the major topics of interest in psycholinguistics in recent years, together with an assessment of the ‘state of play’ in the field at the moment. I hope it will be useful.

I am extremely grateful to a number of scholars who made helpful comments on the manuscript. In particular, and in alphabetical order, Michael Banks of the London School of Economics, David Bennett of the School of Oriental and African Studies, Paul Fletcher of Reading University, Jerry Fodor of the Massachusetts Institute of Technology, Phil Johnson-Laird of the University of Sussex, Geoffrey Sampson of Lancaster University, and Deirdre Wilson of University College, London.

The book would probably have been better if I had taken more notice of their comments – but as the suggested improvements were often contradictory, it was difficult to decide whose opinion to accept. In cases of doubt, I preferred my own, so I am wholly responsible for any errors or over-simplifications that the text may still contain.

My thanks also go to Irene Fekete, the evening-course student (and Hutchinson’s executive) who persuaded me to write this book.

Let me add a brief note on style. In English, the so-called ‘unmarked’ or ‘neutral between sexes’ pronoun is
he
. Had I used this all the way through
The Articulate Mammal
, it might have given the misleading impression that only male mammals are articulate. I have therefore tried to use an equal number of
he
s and
she
s in passages where a ‘neutral between sexes’ pronoun is required.

Jean Aitchison
London, 1975

P
REFACE TO THE
F
IFTH
E
DITION

In the thirty plus years since this book was first published, psycholinguistics has increased considerably, both in popularity and in the amount written about it. It has expanded like a young cuckoo, and is in danger of pushing some more traditional interests out of the nest. Or, to take another metaphor, it has behaved like an active volcano, belching out an increasing lava-flow of important findings which have poured out over almost all areas of linguistics and psychology, and have – to some extent – changed the shape of the landscape.

Luckily, many of the questions asked remain the same, though many more answers have been proposed. It is clearly impossible to include all the new developments in this revised edition. I have, however, attempted to outline those which seem most relevant to the issues discussed in this book. No chapter remains unaltered, and some have undergone substantial additions and/or changes. For example, human ‘mind-reading’, the ability to understand the intentions of others, is turning out to be a key property underlying language (
Chapters 2
–
3
). Huge steps forward have been taken in understanding the brain, largely due to the increased sophistication of modern brain scans (
Chapter 3
). Chomsky’s ideas are still recognized as playing a foundational role in modern psycholinguistics, but are these days being pushed out of the limelight by the work of a younger generation of scholars (
Chapter 5
). Verbs have continued to take centre stage in children’s acquisition of language (
Chapter 7
) and in speech comprehension (). And so on, and so on. In addition, numerous new references have been added. I hope this new edition will enable readers to keep up with what is happening in the field at the moment.

As before, I am grateful for the skill and help of those at Routledge, especially (for this edition) Nadia Seemungal.

Jean Aitchison
London, 2007

I find my position as an articulate mammal bewildering and awesome Would to God I were a tender apple blawssom

Ogden Nash

I
NTRODUCTION

Psycholinguistics is sometimes defined as the study of language and the mind. As the name suggests, it is a subject which links psychology and linguistics. The common aim of all who call themselves psycholinguists is to find out about the structures and processes which underlie a human’s ability to speak and understand language.

Both psychologists and linguists are involved in studying psycholinguistics. As one group of researchers has noted:

The name says it all … it is simultaneously psychology and linguistics. At the heart of the discipline, therefore, is the relationship between these two fields, each of which can boast centuries of research tradition … By contrast, psycholinguistics itself is relatively young … psycholinguistics as we understand it today and as a discipline with its own name has only been in existence since the mid-twentieth century.