Jump to ratings and reviews
Rate this book

New Horizons in Linguistics

Rate this book
New Horizons in Linguistics

367 pages, Paperback

First published January 1, 1970

Loading...
Loading...

About the author

John Lyons

131 books31 followers
Librarian Note: There is more than one author in the GoodReads database with this name.

Sir John Lyons, LittD, FBA, is an English linguist, working on semantics. [http://en.wikipedia.org/wiki/John_Lyo...]

Ratings & Reviews

What do you think?
Rate this book

Friends & Following

Create a free account to discover what your friends think of this book!

Community Reviews

5 stars
10 (34%)
4 stars
10 (34%)
3 stars
9 (31%)
2 stars
0 (0%)
1 star
0 (0%)
Displaying 1 - 2 of 2 reviews
Profile Image for Liam Porter.
194 reviews51 followers
February 3, 2014
This is an interesting collection of essays written for the general public, written at a time of great excitement and innovation in the field. Though the essays cover all the subdivisions of the field, syntax dominates the book since the Chomskyan Revolution had still not settled into the orthodoxy by this point, and great excitement for its project had by this point decisively won over the skepticism of the older stucturalist generation.

The editor of the collection, John Lyons, relates this internal state of affairs to the general reader, including this criticism of the narrow range of linguistic evidence in use by syntacticians:

It must be appreciated that a considerable amount of research is going on all the time into the grammatical structure of various languages, which, though it does not command the same general interest or have the same immediate effect as the more obviously theoretical work, must eventually be taken into account in the further development of syntactic theory. The current concern with "deeper" and more "abstract" syntax on the part of many theoreticians makes it all the more important to examine as wide a range of languages as possible. Much of the work so far published which purports to demonstrate the "universality" of certain "deeper" analyses of English makes only the most cursory reference to the structure of other languages [...secondly,] the differences between the various "versions" of post-Aspects[...of Language, by Noam Chomsky] generative syntax are now at least as striking as those that exist between "Chomskyan" syntax as a whole and other "non-Chomskyan" systems of syntatic analysis. Chomsky himself has remarked on the work that derives immediately from his own: "At present the field is in considerable ferment, and it will probably be some time before the dust begins to settle and a number of issues are even tentatively resolved" (1968:54, fn.6) If and when the dust does settle, it may be even less useful to distinguish between "Chomskyan" and "Non-Chomskyan" grammar than it is now.,/i> p.139 John Lyons, Dept. of Linguistics, Edinburgh


M.A.K. Halliday, a more conservative voice, writes a chapter on "language's function." Seeing the functional of language as self-evidently concerned with the communication of thoughts (inseperably linked to their internal representation), he dissents from the Chomskyan distinction between linguistic "competence" and "performance", at least, drawing more attention to issues of Pragmatics which show that "language serves to establish and maintain social relations." But even though he is a more conservative voice, he still allows that language habituates the reception and formation of concepts:

Language serves for the expression of "content": that is, of the speaker's experience of the real world, including the inner world of his own consciousness. We may call this the ideational function , though it may be understood as easily in behavioural as in conceptural terms. In serving this function, language also gives structure to experience, and helps to determine our way of looking at things, so that it requires some intellectual effort to see them in any other way than that which our language suggests to us. p.142 M.A.K. Halliday Department of General Linguistics, University College London


The phonetician D.N.Fry describes here the concept of allophonic variation within a language-determined phonological system. I think that he converys the mystery of the facts behind speech perception very well in just a few illustrative observations:

The particular varieties of sound which have to be processed in [one of] two ways, that is assigned to a single class or to different classes, depending on the system of the given language. In English, for example, there are at least four or five varieties of the [t] sound which have marked acoustic differences, but the ordinary English listener is not aware of these differences. On the other hand in some Indian languages, there are six classes of sound which must be kept separate if the phonological system is to function properly, and some of the acoustic differences to which the English listener is insensitive are the features which the Indian speaker uses to sort the [t] sounds into their right classes. By contrast, no English listener would be likely to confuse the two English words red and raid, and yet the vowel sound in these words are indistinguishable to many Indian speakers. The listener's processing of acoustic information in speech is therefore carried out according to the patterns established by language learning. Just as visual; recognition depends on the perception of visible features, so recognition of speech sounds depends on the perception of audible features of one kind or another.

[...]It is somewhat difficult for the ordinary language user to realize just how wide a variety of sounds he puts into a single phonemic class. If we imagine one English syllable, "no", as said to us by [...] southern English, northern English, Cockney, Scots, Welsh, Irish, American, Australian [speakers] and we then imagine these varietites of sound multiplied by the differences for example between a man, woman or child speaker and again by the immeasurably greater number of individual speakers, it becomes clear that the class formed by the various vowel sounds used in this one word is very large. [...] When the listener puts them all into one class he is therefore disregarding these acoustic differences[...]
p.34 D. N. Fry. Dept. of Phonetics, University College London


In this chapter by J. P. Thorne, even the linguistics of literature is touched upon, which is a surprisingly rare combination (at this time Angloamerican literature departments had let structuralism pass them by and were just coming to grips with a Derridean remodelling of De Saussure) and one that I think would have been an interesting genre of literary criticism were it to have continued. This description of the poet's art is as good as a description as any I've heard before:

[Some claim] that what distinguishes poetry from mere verse is that in the case of the former these phonological regularities are matched or reinforced by grammatical regularities. The view is summarized in Mac Hammond's (1961: 482) statement that "syntax is poetic when grammatically equivalent costinuents in connected speech are juxtaposed by coordination or parataxis, or are otherwise prominently accumulated." [...] Implicit in some of these stuides is the idea that not only with regard to the phonological component of the grammar but with regard also to the syntax and semantics the poet works under the restraint of self-imposed rules; that is, rules which do not form part of the grammar of a natural language. [...] One way of representing these facts [of poetic language] is to list those structures in the poem which are deviant, indicating in each case which rules would actually generate these structures. Most of the rules would form part of the grammar of Standard English (that is, a grammar which would generate only well-formed sentences), but it would be necessary to give special rules for the first person pronoun, which in the language of the poem always selects an inanimate predicate [...] and for verbs like laugh and love, which in the language of the poem can select inanimate subjects. There is good reason for preferring the second approach. It suggests a quite interesting hypothesis about what it is poets are trying to do when they write poetry and, therefore, also about what it is that distinguishes some kinds of poetry from prose. Behind the idea of constructing what is in effect a grammar for the poem lies the idea that what the poet has done is to create a new language (or dialect) and that the task that faces the reader is in some ways like that of learning a new language (or dialect). p.194 J. P. Thorne, dept. English Language, Edinburgh



In a tantalizing but, for me, unsatisfying chapter, Janet Fodor (married to Jerry Fodor), talks about the relationship between symbolic logic and generative syntax, which is so close as to be, so far as I can tell, only an application of the former (today Chomsky is called a "linguist and philosopher" or "political activist and linguist" but I've read George Miller, in another pelican book, introduce him as "logican and linguist, Noam Chomsky). However Fodor seems engaged yet skeptically removed (writing very much like Jerry Fodor), writing this passage which seems to pull back from either endorsing or condemning the "apparently growing trend among linguists to ransack textbooks of logic":

Parallels could probably be extended, but they are not especially interesting. Any two axiomatic systems will show formal similarities of this kind. In fact even for chess there is an "axiom" (the initial positioning of pieces on the board) and a set of "transformational rules" which license moves from onme configuration of the pieces to another. More revealing of the relationship between grammars and systems of logic is a comparison of the functions of the representations and rules which the two types of system employ. p.199


I would be inclined to wonder what united symbolic logic, natural lanugage chess games that made them all intelligible only to humans, and whether all three have nothing in common but formal properties; also I wonder what else, other than chess, we can formalize in a symbolic-logic-like artificial language... presumably only systems of which we know all the rules, namely scientific theories of natural systems, which are themselves limited by human intelligence, and perhaps even... language?

Speaking of Fodor, this book contains an account of an experiment into speech perception that Jerry Fodor co-conducted:

Since [the combination of words into meanings] depends in some way upon syntax, linguistic perception must involve a form of syntactic analysis. At the very least, a sentence is likely to be analysed into its grammatical consituents to yield what is called its "surface structure." One aspect of such a process has been examined in an ingenious experiment by Garrett, Bever and Fodor (1966). They constructed pairs of sentences that had identical endings but different beginnings:

1). In order to catch his train George drove furiously to the station,
2). The reporters assigned to George drove furiously to the station.

The identical endings have a different surface structure because the major boundary between clauses occurs immediately before "George" in 1). but iimmediately after it in (2). This may be readily checked by reading aloud the two sentences and trying out pauses at these two points.) Tape-recordings were played to students so that in one ear they heard a sentence and in the other a click timed to occur in the middle of a word, e.g. "George." After each presentation they wrote down the sentence and indicated where the click had occured. There was a reliable tendency for it to migrate from its objective position towards the major syntactic boundary. Thi suggests that in the region of such boundaries the perceptual mechanism is more preoccuped with grammatical anaysis than with monitoring the incoming sounds.
p.262 P.N. Johnson-Laird


In a chapter about communication systems in the animal world, many interesting and thorny issues are touched upon. It concludes that there is nothing like the structure of language in animal communication, yet if we change the question to the existence of "communication," there is no doubt about its manifestation in every other sense that we can imagine:

The most striking differences between animal signs and language behaviour are to be found, then, in the rigid, stereotyped nature of the former and in the fact that they are under the control of independently specifiable external stimuli and internal, motivational states. The vocalizatons of monkeys and apes do not permit greater flexibility of expression, as they vary semi-continuously along acoustic parameters, but their communicative significance, would appear, nonetheless, to be restrcited to expressions of emotional states. As Marler (1965) writes, the graded nature of the cries "permits a subtlety of portrayal of slight shifts in motivation that would be impossible with more stereotyped signals." Even the alarm calls of the vervet monkey which seem, superficially, to be "naming" the type of predator are more plausibly regarded as expressing no more than the relative intensity of the fearful and aggressive emotions aroused by the various predators. Thus Andrew (1963) has suggested that vocaliation in primates is to be interpreted in terms of the patterns of facial expression and muscular tension associated with particular emotive situations. Vocalization is seen, then, as an accidental by-product of bodily "gestures," the structure of which may lead to become ritualized in the way that other movements are. An emotive interpretation of primate calls is also suggested by the observation that vocalizations can be obtained electronically stimulating sites in the limbic system of monkeys. p.234-5

Essentially we need criteria which indicate that an animal recognizes and intends that its behaviour shall have an effect on the behaviour of another animal. The clearest cases are probably those which reveal an "intention to misinform." Anecdotal evidence suggests that some higher animals can mislead by "lying," in various ways, rathe than simply by making mistakes. Lorenz (1954) provides many charming examples, including the following story about one of the his dogs which attempted to concel the fact that it had mistaken his master for stranger:

I had just opened the yard gate, and before I had had time to shut it the dog rushed up barking loudly. Upon recognizing me, he hesitated in a moment of acute embarrassment, then pushing past my leg he raced through the open gates and across the lane where he continued to bark furiously at our neighbour's gate just as though he had been addressing an enemy in that garden from the beginning.

Emperimental evidence for behaviour of this nature would, I think, support claims that an animal understood the notion of communication. Without such demonstrations, the most that could be said is that certain signals are informative. Whilst Lorenz's dog may plausibly be regarded as communicating, one might still hesitate to call his behaviour "language-like." p.235 p.235 J.C.Marshall Speech and Education Unit, Edinburgh



In a chapter called "The Perception and Memory of Sentences," P.N. Johnson-Laird recounts several interesting experimental studies into how recall of linguistic material allows for some tentative claims - at least negative claims - about processes of thought. For one thing, it cannot be claimed that the "form" which memories of a sentence's content take is the same as the "form" that we perceive when we are actively speaking or receiving the sentence;the mistakes that people make in recalling sentences is simply too variable:

If [syntactic] deep structure as I have argued has no independent psychological status, it is difficult to see how it could directly enter memory. And it is evident that it is not normally recalled. Using more complex sentences than earlier studies Clark and Clark (1968) have even confirmed this when subjects knew that verbatim recall was required. However, a simple but stringent test would involve the detection of paraphrases of the original material. [...]subjects are poor at recognizing changes such as substitution of "closed" for "not open" but rather better at recognizing non-synonumous substitutions such as "cold" for "not hold." [...] The most suitable material [for accurate recall] would involve paraphrases of the following variety: "James liked one particular painting very much and the painter sold it to him; One particular painting pleased James very much and he bought it from the painter." If subjects confuse these statements, then it is clear that they have no memory for deep-structure relations but a good memory for meaning.

No one knows how meaning is represented within memory, but there is no evidence to show that any form of syntactic structure is directly involved. This is puzzling because if syntax is not necessary for storing meaning why should it be necessary for communicating it? The answer may be that elements from which meaning is composed (the so-called "semantic markers") are utilized in cognitive processes other than language.
p.269 P.N. Johnson-Laird


Finally, there is a chapter on "historical linguistics," which is optimistic that the older tradition and the newer generative one can work together fruitfully:

It has been suggested by critics of generative grammar (e.g. Hockett 1968a) that it has become caught in the fatal trap of taking its theoretical entities such as rules, ordering, underlying base forms etc. for real entities, whereas they actually, according to Hockett, are only tools to be used for analysing the real entities such as phonemes and morphemes. It is as if, in Hockett's similie, a patholgist were to confuse his slides and microscopes with the bacteria under investigation.

Historical linguistics supports exactly the opposite conclusion in an especially direct way. We have seen that sound change and analogy is change in the grammatical system of language, including precisely the rules and their order. If these real events, then, are changes in the rules and their order, then the rules and their order must be real entities too. [...] Language change is for the linguist, to change the analogy a little, what earthquakes and volcanic eruptions are for the geologist, or supernovae for the astronomer. They add welcome new perspective in a field where the object of study is static and not readily amenable to experimental manipulation. Just as the careful ananlysis fo earthquakes may reveal something about the earth's interior, so careful ananlysis of linguistic changes may reveal otherwise inaccessible aspects of linguistic structure.
p.314 Paul Kiparsky, Dept. of Linguistics, MIT


I found this collection of essays very stimulating and it cleared up some ideas that seemed to be too quickly glossed over when I studied on a modern linguistics course. It's a shame that these "New Horizons" books, or something similar, no longer have any audience, since they are an excellent compromise for the interested general reader who does not want to read excessive technical jargon, or dumbed-down contemporary nonfiction, or pay exorbative prices.
Displaying 1 - 2 of 2 reviews