Links to pages: 568, 569, 570, 571, 572, 573, 576, 577, 579, 580, 581
Comments Welcome
George W. Grace
University of Hawaii
Revised 28 November 1989
"When I took the first survey of my undertaking, I found our speech copious without order, and energetick without rules: wherever I turned my view, there was perplexity to be disentangled, and confusion to be regulated; choice was to be made out of boundless variety, without any established principle of selection..." (Samuel Johnson, Preface to the Dictionary).
"I agree with Joseph that standardization dramatically affects the course of evolution of a variety in an unnatural way by deliberate human intervention, thus rendering standard languages as artificial objects. I use the term 'objects' here deliberately to emphasize even more their special ontological status (see Le Page and Tabouret-Keller 1985; and Popper's 1972 remarks on Third World objects). Because standard languages are taught in schools, their acquisition proceeds in a way which is nothing like that of natural languages." (Suzanne Romaine 1989: 577)."
To begin with, I believe that the world is coming more and more under the dominion of a single culture. The principal roots of this culture are in Europe; its rapid spread beyond Europe was begun under European colonialism. However, even though colonialism of the classical kind has largely ceased to exist, the dominion of the culture which it unleashed continues to grow. I have been calling this culture "the modern-world monoculture" (cf. Grace 1988: 487ff.). Although it is still often referred to as European or Western, it has now spread through the entire world. For example, the nation-state is an institution of the monoculture, and the entire world is now divided up among nation-states. In fact, what is called "international law" is also a thing of the monoculture. In other words, the world is ruled (to the [considerable] extent that any unified rule may be said to exist) by the monoculture. The monoculture prescribes the rules by which international dealings of any sort--e.g., diplomatic, economic, even warfare--are conducted. Moreover, the material culture of the world as a whole is dominated by monoculture technology, and science in the modern sense is one of the most important institutions of the monoculture.
In short, no one in the world today can entirely escape the dominion of the monoculture. However, most, if not all people have other cultural resources. In most of the world the monoculture may be thought of as superimposed on a traditional culture, with the traditional culture still surviving in certain roles.
Now, one of the requirements of the monoculture is a language capable of performing certain functions which are indispensable to that culture. More precisely, the requirement does not concern what the language can actually do as much as what people can be made to assume that it does. Probably the most necessary assumption is that the language produces autonomous text. That is, it is assumed to produce linguistic expressions such that each expression has its own meaning which remains constant irrespective of the context in which it may be used ("the meaning is in the text"). It is assumed to be "hooked onto" the world".
There seem to be a number of further assumptions, at least many of which follow from this most fundamental one. For example, it seems probable that both the assumption that the language in question conforms to the grammar-lexicon model in its structure and the assumption that the language is a "language of universal translation" (see Grace 1987: 445ff.)(1) are necessary corollaries of the autonomous-text assumption. In any case, what is needed is what I have called "school language" (cf. Grace 1989a: 523ff.). A school language is a language for which standards and rules have been deliberately designed and imposed by some authority. The aim in the imposed design is to create a kind of language capable of specifying in a very precise way situations of the most previously unimagined kinds. One might think of it as a kind of erector set which can be used to invent (by characterizing them) situations of the most novel, and the most abstract, kinds within the scope of the human imagination.
The designation "school language" was suggested for it because it must be taught in schools. I believe that this "school language" ("erector-set language"?--what Joseph 1987 calls simply a "standard language") is always superimposed on what I called (in Grace 1989a: 523ff.) a "vernacular language" base. In fact, just as most individuals have cultural resources other than those of the monoculture, so this school language probably does not constitute the entire linguistic resources of any individual. But the vernacular which is the native language of the students is often seen as a pernicious influence and the teaching of the school language may be accompanied by direct attacks on it.
To put it differently, I assume everyone to have access to what I have called (Grace 1989b: 557ff.) "recognition strategy" (which I take to be the normal strategy used for vernacular language), whether or not they also have been taught to use "analysis strategy" (the strategy associated particularly with the use of school language).
Thus, when someone speaks of "the English language", he/she may be speaking of the standardized school language (this is what Joseph Greenberg has called the "par excellence" interpretation--i.e., the school language is English par excellence), but what he/she has in mind might also include some vernacular variety or varieties as well as (or even on occasion to the exclusion of?) the school language.
What concerns me is a growing suspicion that we monoculture intellectuals are, as it were, creating man (i.e., human beings qua speakers of languages) in our own image--that is, that we are projecting onto humankind at large the kind of linguistic competence that comes only from intensive training in the manipulation of a school language (note that, in the case of many "mainstream" Americans for example, this training may begin long before school age). I have come increasingly to believe that this kind of competence may be different in very important ways from the competence of the typical native speaker of the typical language in human history. And I am concerned that this difference may be such that it would better be thought of as a difference of kind, rather than one just of degree.
I believe, in fact, that there is a whole package of insufficiently thought out assumptions about "natural" languages now being made by linguistics, and I believe that they should be brought into the light and subjected to more careful scrutiny. One concern is that these assumptions, and other assumptions which derive largely from them, have been very widely disseminated beyond the linguistic profession with the authority of the profession behind them. It is hard to estimate their social consequences. If they are false, and I think that they are false in large part, it is hard to estimate what harm can result. In any case, I think it is not a matter about which the linguistic profession (or other affected professions) should be indifferent.
Of course, what I am saying is that standard languages (what I have been calling "school languages") are to a significant extent the product of deliberate human design. But how does this affect the program of linguistics, the object of which is said to be "natural language"? What is meant by "natural language"? What is meant when a linguist speaks of a language as "natural"?
The distinction between natural, on the one hand, and artificial, on the other, is clear in the case of physical objects. It is clear what contrast is being drawn when we say that the agencies which have shaped a particular stone were natural ones (water or wind erosion, percussion when being rolled in a stream, etc.), and that another stone is an artifact because it has been deliberately worked into its present (artificial) shape by human craftsmen.
The use of the terms "natural" and "artificial" in reference to language is intended to imply the same distinction; a distinction, that is, between systems (the "natural" languages) whose development and present characteristics can be explained in terms of the operation of some kind of natural laws, and systems whose characteristics cannot be explained without reference to deliberate decisions made by human beings who designed them. More specifically, the intent has been to exclude such things as computer "languages" or the so-called "languages" set up by logicians and mathematicians as well as so-called "artificial languages" such as Esperanto.
Thus, the assumption underlying the notion of natural language is that the characteristics of individual natural languages are attributable to the operation of natural laws. What kind of natural laws? It has been proposed that there are natural laws governing semiotic systems as such, and again that there are natural laws governing linguistic change. But above all today, there is the idea that the human brain is in part a language machine--that the design of the brain contains detailed specifications of much of the structure of human languages. Study of the structural properties of languages, therefore promises, according to this idea, to lead to very specific and otherwise unforeseeable insights into the laws governing the most essential aspects of human nature.
I mentioned above a "package of assumptions" made by contemporary linguistics about which I felt we should be concerned. Two of those assumptions were mentioned in the foregoing discussion. They are:
1. that there are languages which are "natural" in the sense that the characteristics of individual languages are attributable to the operation of natural laws, and
2. that the human brain contains detailed specifications of much of the structure of human languages.
But there is a third assumption which I want to add immediately. It is:
3. that the familiar human languages which have been isolated and given names, such as English, are natural languages in the sense with which we are concerned.
Maybe I should mention in passing that I'm somewhat skeptical about any languages being natural in the sense intended, and that I believe less and less that the brain functions associated with language are separate from those employed for other purposes. However, I'll leave such questions aside for the present. What I do want to discuss is the status of what I have called "monoculture-dedicated languages"--"standard languages" in the sense which Joseph (1987) gives to that term. According to Joseph, "...'standard languages' represent a specifically European concept, whose defining criteria are based on the attributes of European languages and on European cultural values" (1987: x).
What I want to say is that assumption 3 is false--that whether or not there are some languages that are natural in the intended sense, English is not one of them. Or at least standard English is not--the English of Basil Bernstein's "restricted code" might be another story.
I thus agree with Suzanne Romaine's statement quoted at the head of this paper that standard languages are artificial objects. (I included the quotation from Samuel Johnson as an indication of how the not-yet-sufficiently-standardized English of two centuries ago appeared to one observer).
In order to consider what is at stake here, it is important to recognize some of the other assumptions in the package of which I spoke at the outset. I propose that the following, at least, constitute part of the package:
4. that a natural language takes the form of a grammar (where "grammar" is understood as including a phonology and a lexicon),
5. that any "normal" person who is exposed to a natural language during his/her early years (it's not clear just how much exposure is required, but it has been emphasized that the requirement is not high) will learn (or "acquire") the language--i.e., that he/she will become a fully qualified "native speaker",
6. that to learn (or "acquire") a language--that is, to become a fully-qualified speaker--is to come to know (in a particular sense of "know") its grammar (i.e., to have internalized a linguistic description of pretty much the sort that linguists write),
7. that the reason why "knowing" the grammar is necessary is that the components of the grammar must be used in order to speak the language--i.e., to come up with acceptable linguistic expressions and to understand what is said by others,
8. that the language machine in the human brain includes a "language acquisition device"--i.e., that all normal individuals are born with built-in strategies for analyzing and otherwise processing the input data so as to infer the grammar of the language being "acquired".
It is not hard to find grounds for disputing any of these assumptions, but I won't attempt to discredit all of them here. In fact, some of them--at least if properly reformulated--seem quite defensible. For example, in defense of assumption 4, I think we may say that languages do have grammars in the sense that the linguistic expressions of the language show patterning from which generalizations can be abstracted. The grammatical patterns which are revealed by linguistic analysis are surely there. In fact, I assume that the richness of language as we know it could not exist without such patterning. The patterning is surely an important aid in interpreting individual linguistic expressions as well as in learning the language in the first place (cf. Grace 1989a: 532-33). However, to say that a language has a grammatical structure is not to say that that structure is known by its speakers.
Assumption 5 can also be defended. I see no objection to the assertion that every normal child who has any kind of reasonable exposure to a language during his/her formative years will become a speaker of the language. However, to say that is not to say that all native speakers are intersubstitutable--that all have the same linguistic competence. It seems apparent to me that the linguistic abilities of different individuals vary widely. The idea that there is a single kind of "competence" in a language that takes the form, for all speakers, of a kind of replica of a linguistic description of that language seems very wrong to me.
It is assumptions 6-8 that seem to me to represent the heart of the problem. I do not see any reason to believe that learning a language is equivalent to internalizing a linguistic description of that language. Of course, the formal learning of a school language, and the similar approach to learning a second language (which, if it is formally taught, will generally be taught like a school language), comes much nearer being of that kind. But that is precisely what I am arguing is not natural.
In Grace 1989a (esp. 513-18) I discussed some of my reasons for believing that it is not necessary, at least in some cases, to go through any process which involves taking grammatical structures into account, either in understanding speech or in coming up with acceptable linguistic expressions to utter.
And of course, since I do not believe that natural language speakers "know" grammar, I find no reason to suppose that a "language acquisition device" or any other kind of language machine exists in the brain. In fact, really do not believe that the brain has many functions at all which are dedicated exclusively to language. It seems to me that most of the abilities that we display in using language are very closely related to abilities that we display in other activities.
To review, what I believe has happened is this. A single cultural system has assumed overwhelming dominance in the world today. An essential cornerstone of this culture is a particular assumption about the nature of language. The assumption is that language can be used to produce text which has a precise, permanent, invariant "meaning"--texts in which facts, laws, agreements, etc. can be captured and preserved in language. This assumption is supported by the assumption of individual language systems (languages) made up of lexical items with specified meanings and grammatical rules which are used to produce sentences in which different lexical items are placed in particular logical relationships to produce statements which in turn have specified meanings.
At times an assumption has had currency that the kind of language which I have just described--the kind required by the monoculture--did not occur naturally, but had to be deliberately designed (see the Samuel Johnson quotation at the head of this Note). Related to this was the assumption that some languages were evolutionarily more advanced than others--that various languages were "primitive" in one or another degree. However, neither of these positions seems at all acceptable among what might be called "right-thinking people" today.
Now, if the monoculture ideology with respect to its own languages could not safely be questioned, and if was not permissible to suggest that any other languages were different in significant ways from the monoculture (i.e., "school", "standard") languages, then there was no choice but to use the monoculture model of language structure in analyzing and describing other languages.
But surely the standard languages of monoculture societies are the among the most--very possibly the very most--unnatural languages that are to be found anywhere. To interpret other languages in their terms seems a misguided strategy.
With what hypotheses, then, should we approach natural languages? As a starting point, I would like to suggest the following:
1. Learning a natural language in the natural way is primarily a matter of building a very large memory-store.
2. Acquiring this store and using it in speaking and understanding speech requires no intellectual abilities that are dedicated specifically to language.
(Note, however, that this does not mean that the gradual development of increasingly language-like capacities was not a selective factor in the evolution of the brain).
3. No analytic knowledge (that is, knowledge of grammatical or other rules or regularities) of one's language is required for one to qualify as a full-fledged native speaker.
4. However, analytic knowledge can subsequently be obtained by the same methods used by linguists (although presumably used with much inferior sophistication)--i.e., simply by noticing and remembering regularities (rules).
5. Moreover, the possession of analytic knowledge is probably advantageous. Analytic knowledge and the analysis strategy which it makes available are necessary for the use of standard languages, of course, and I suspect that they make possible more skillful manipulation of any language. Therefore, there may be incentives (over and above the aesthetic pleasure derived from discovering and observing the patterning) for individuals to seek analytic knowledge.
6. If analytic knowledge is both readily accessible and generally advantageous, it is not unreasonable to suppose that it is fairly common for individuals to obtain some of it. That is, it seems likely that even native-speakers of the most nearly natural languages frequently acquire some analytic knowledge of their languages and practice analysis strategy at times.
7. It seems likely that some, or all, of those individuals who make good linguistic informants are persons who have already undertaken some analysis of their languages and have acquired some skill at analysis. However, it is possible also that some of them are simply individuals with natural analytical capabilities who, although they have never used these capabilities on their language, find it easy to learn to do so.
None of these suggested hypotheses strikes me as patently false.
1. Joseph 1987 lists intertranslatability as one of the components of the "standard language" concept. To quote (1987: 6), "Standard x must be readily intertranslatable (and regularly intertranslated) with one or more languages recognized as standard." Back up
Grace, George W. 1987. The post-Linnean world-view and languages of universal translation. Ethnolinguistic Notes, Series 3, Number 29. Printout. 3 pp. Back up. Also internet World Wide Web page (Click Here).
Grace, George W. 1988. The idea of a theory of translation: On shared and unshared cultural backgrounds. Ethnolinguistic Notes, Series 3, Number 33. Printout. 10 pp. Back up. Also internet World Wide Web page (Click Here).
Grace, George W. 1989a. The association of situations with linguistic expressions. Ethnolinguistic Notes, Series 3, Number 35. Printout. 27 pp. Back up. Also internet World Wide Web page (Click Here).
Grace, George W. 1989b. Recognition strategy and analysis strategy in language use. Ethnolinguistic Notes, Series 3, Number 37. Printout. 10 pp. Back up. Also internet World Wide Web page (Click Here).
Joseph, John Earl. 1987. Eloquence and power: The rise of language standards and standard languages. London: Frances Pinter. Back up
Romaine, Suzanne. 1989. Review of John E. Joseph, Eloquence and power: The rise of language standards and standard languages. Linguistics 27: 574-79. Back up
| Home Page | The Ethnolinguistic Notes | The Ethnolinguistic Notes, Series 1 and 2 | Ethnolinguistic Notes, Series 3 | The Ethnolinguistic Notes, Series 4 | Reflections: Language Evolution |
| Reflections: Knowledge of Language | Personal Page | The Human Predicament | Why Write Unpublishable Things? | Modest Proposals | Odds and Ends | Pictures |
First put on the Web on 27 September 1996