Comparative Perspectives on Language Acquisition

A Tribute to Clive Perdue

Marzena Watorek, Sandra Benazzo, Maya Hickmann

• Publisher: Multilingual Matters
• Year: 2012
• ISBN:
  9781847696052
• Topic: Languages & Linguistics
• Subtopic: Psycolinguistics
• Index: Languages & Linguistics

This volume aims to provide a broad view of second language acquisition within a comparative perspective that addresses results concerning adult and child learners across a variety of source and target languages. It brings together contributions at the forefront of language acquisition research that consider a wide range of open questions: What are the precise mechanisms underlying acquisition? How can we characterize learners' initial state and predict their degree of final achievement? What role do specific (typological) properties of source and target languages play? How does fossilization occur? How does the relative complexity of cognitive systems in adult and child learners affect acquisition? Does language learning influence cognitive organization? Can language learning shed light on our general understanding of human language and language processing?

Part 1

Second Language Acquisition:
From Initial to Final Stages

1 A Way to Look at Second
Language Acquisition

Wolfgang Klein

Thinking Back

It is peculiar, but often it is neither the most recent nor the most important events that are most vividly remembered. I must have met Clive Perdue for the first time at a workshop of the GRAL group in Vincennes 30 years ago. But while I clearly remember Larry Selinker pouring a glass of red wine over my jacket, there is no recollection whatsoever of speaking with Clive. Still, the impression must have been deep, because when, soon afterwards, the European Science Foundation in Strasbourg decided to fund a Europe-wide research project on the second language acquisition of immigrant workers, and the Max Planck Society in Munich granted me money for a coordinator for this project, it was Clive Perdue who immediately came to my mind. So, I called him in Paris, he listened, and after one of those long and thoughtful pauses that were so characteristic of him, he asked ‘Ah, ah, are you offering me a job?’ And I still hear the slight tone of disbelief and the distinct rise on the word ‘job’ in his voice, as if it had been yesterday. I said ‘yes’, and this was the beginning of a wonderful cooperation and of a wonderful friendship spanning almost three decades. In fact, friendship as well as cooperation last well beyond his death, because once in while, I note to my own surprise that I keep asking him questions, and he keeps answering them.

When you learn a second language, be it in the classroom or in the wild, you often face substantial problems because a particular feature of this language – a sound, a word, a construction – has no immediate counterpart in your earlier linguistic knowledge. French learners of English often struggle with the first sound in this because there is no such sound in French; and English learners of French often struggle with the last sound in tu, because there is no such sound in English. But you may also have problems because a particular feature has a counterpart which is very similar but not completely identical. French as well as English have an unvoiced dental stop/t/; but in English, it is aspirated, and in French, it is not; such little differences belong to the hardest pièces de résistance for ultimate achievement. German learners of English regularly have problems with the choice between the progressive form and the simple form: He was singing – He sang. There is no grammatical distinction of this sort in German; both meanings are expressed by the same verb form, which in shape and history corresponds to the English simple form: Er sang. But they also have problems with the present perfect He has sung; while the German counterpart Er hat gesungen is very close in composition and meaning, there are subtle differences, reflected for example in the fact that in English, the time of the event, although clearly in the past, cannot be specified by a past time adverbial, such as yesterday or a while ago. In the early 1980s, when I still tried, or was trying or have tried, to improve my English, I once asked Clive: ‘Can you tell me in a single sentence how the English continuous form is used?’ I do not remember exactly in which year this was, but I do remember exactly that he first began to roll a cigarette, and just before lighting it, he answered: ‘You use it when you are 100% in the action.’ This is not what I had read in the grammar books, and it had never occurred to me when I tried to speak English. But I thought it was a perfect answer, for at least three reasons:

• It is simple.

• It helps the learner because it takes his perspective.

• It captures the crucial semantic intuition.

So, this one-sentence explanation is not a perfect linguistic account of the meaning of the English progressive form, but it contains the decisive empirical germ for such an account.

A few years later, our findings on how second language learners express temporality in the absence of inflectional morphology increasingly excited my interest in how time is encoded in general by natural languages. And more than any tradition of tense studies (which one must not ignore, of course), it was this simple and direct way of looking at the phenomena which guided my interest. If you are 100% in the action, this means that the time about which you make a claim when you speak must be fully contained in the time during which the action lasts. And if you use the past tense I was singing, this means that the time about which you talk (the ‘topic time’, so to speak) (a) precedes the moment of speech, and (b) is fully included in the time of singing – the ‘situation time’; you are 100% in the action; but it does not imply that the action itself cannot include the moment of speech. Hence, tense does not express a relation between the time of speaking and the time of the event, as you read in every text book; instead, it expresses a relation between the time of speaking and the topic time. The aspectual dimension, as brought in by the choice of the progressive rather than the simple form, reflects the fact that the topic time is properly included in the time of the action (see Klein, 1994). If you follow the grammar books and teach your students that tense serves to place the action before the moment of speech, then this is not entirely false; nevertheless, the learners will often be misled, and if they make mistakes, it is the teachers and the linguists who are to be blamed, not the learner. So, if one wants to understand how language works and how people learn languages, one should not, at least not as the first stage, ask what tradition tells one, but just have a fresh, simple and unbiased look at the phenomena at hand.

This is the way in which we – Clive Perdue, myself and many others around the original ESF (European Science Foundation) project and various follow-up projects – began to look at second language acquisition in the early 1980s. This is not the place to recapitulate this work (see e.g. Klein & Dimroth, 2009; Klein & Perdue, 1997; Perdue, 1984, 1993). But in what follows, I will try to characterise the perspective which underlies a great deal of this work in the form of four simple maxims, each of which will be illustrated with a few examples. This perspective is not a theory of language acquisition; it is just a way to approach a field, no more, no less.

Don't Lose Sight of the Obvious!

Here is a small list of facts which, I believe, no one can seriously deny.

(1) No one is born with a specific language in his or her head. But we are all born with the capacity to learn any language. We pass through this long and tedious process which we call language acquisition once, and frequently more often, in life. It can take very different forms, depending on factors such as

• age,

• learning conditions,

• previous linguistic knowledge

and perhaps many others. The course, speed and eventual result of this process vary considerably according to these factors. This is what language acquisition researchers aim to find out.

(2) Language acquisition is not the only possible transition between ‘language’ as a biologically given language faculty, which is more or less the same for all human beings (though minor variation is not excluded), and ‘language’ as specific linguistic systems, which are quite different from each other, and of which there are several thousands on earth. There must have been a time, very long ago, in which the language faculty already existed in the brain of our ancestors, due to some genetic changes in our species, but in which there was no single linguistic system. At that time, no one was exposed to a language: there was no input that our ancestors could learn from. They could not copy an existing system – they had to invent, to create, to construct such systems. And they did.

(3) There is a net conclusion: the genetically given human language faculty involves at least two quite different capacities: (i) the faculty to copy a linguistic system, and (ii) the faculty to construct a linguistic system. Apparently, these faculties correspond to two types of transition from language faculty to linguistic system: language learning and language creation. And the crucial difference between them seems to be that one of them works on input, and the other one does not. But is this really the true case? No. At best, it is only half the truth.

(4) Nobody really knows how our ancestors developed the first linguistic systems. But it seems obvious that the creation of a language also involves a lot of copying. It seems unlikely that, in some idle hours, a particularly talented person among our ancestors thought up the first linguistic system in his or her head and then passed it on to his/her family and his/her best friends. The creation of the first linguistic systems was the result of a fundamentally social process, which of necessity involved a massive amount of mutual copying. On the other hand, the learning of a language, under whichever conditions, is a long process which does not only require copying others, but also involves a lot of construction. Otherwise, we could not explain how rules are learned, at least those rules which are language specific. These rules do not show up in the input, although they underlie it; rather, they are constructed on the basis of input.

(5) Again, there is a net conclusion: both language acquisition and language creation are essentially social processes, and both involve the copying faculty as well as the construction faculty. How do these two faculties come together? The answer is very easy in one sense and very complex in another. These faculties come together in communication: someone says something with a certain aim, others understand it and in doing so, gain access to the other's knowledge. Our ancestors constructed a linguistic system together be...

Table of contents

• Coverpage
• Titlepage
• Copyright
• Contents
• Contributors
• Introduction: New Comparative Perspectives in the Study of Language Acquisition – Clive Perdue's Legacy
• Part 1: Second Language Acquisition: From Initial to Final Stages
• Part 2: L1 and L2 Acquisition: Learner Type Perspective
• Part 3: Typological Variation and Language Acquisition