More harm has been done in science by those who make a fetish of skepticism, aborting ideas before they are born, than by those who gullibly accept untested theories. V. S.Ramachandran, Creativity versus Skepticism within Science. Skeptical Inquirer 30(6); 48-51, 2006
Why do I write unpublishable things? What it would be most desirable for writing to accomplish, I suppose (beyond what is accomplished in the process of writing itself), would be to have everyone who might be inclined to read a particular writing know about it and have easy access to it. Although formal publication would seem to be the most obvious path, even that wouldn't necessarily bring the work to the attention of every potentially interested party nor would it provide easy access to everyone. Still, formal publication might hope to reach some potential audience. What alternative could offer even that hope? I haven't really found an answer for that question.
Furthermore, I should point out that although unpublishability is possibly the most conspicuous characteristic of the Ethnolinguistic Notes, it, of course, was never an objective. In fact, for a long time after I began writing the Notes I had--in the back of my mind in any case--the idea that I would try eventually to find a way to publish at least a good part of their contents. And in fact two books based on their contents have been published (although another two were not). However; these books resulted from a deliberate re-working of the material of the Notes--a re-working that was intended precisely to try to make them acceptable to publishers. Moreover, in the two cases where I did succeed in getting books published, the books were hardly successes either commercially or in their impact on any audience. In any case, the books clearly don't undermine my general characterization of the Notes as unpublishable.
But the point I want to make here is that I didn't really have a choice between making the Notes publishable or unpublishable. There would have been no way to adapt what I wanted to write so as to make it publishable without fundamentally changing its character (or at least that's how it seemed, and still seems, to me). When I began writing these Notes, my main concern was to formulate questions that were disturbing me, and to look for answers. I had been feeling more and more that I didn't really understand what I was doing. I felt a need to try to figure out what assumptions I was making--first of all what assumptions underlay the research I'd been doing, but also what assumptions underlay the things I'd been teaching.
I felt (and in fact I still do) a strong need to try to work through these questions in writing, and I have no regrets. Of course I would like it better if people read the Notes and found something useful in them. But a reviewer of one of my books said that he suspected that I was writing for myself, and I think that is somewhat true. Certainly an important part of what I'm doing is trying to arrive at some understanding of things that puzzle me. Still I hold the fond hope that someday people will find and read some of what I've written and find it helpful--that's why I keep trying to find ways to make the Notes as generally accessible as possible.
Therefore, my answer to the question: "Why write unpublishable things" is that one presumably does it when the things one wants to write are not the kind of things that are publishable.
And that leads me to the question that I really want to consider here: What makes something publishable or unpublishable? What kinds of writings are publishable? Of course, manuscripts of perfectly appropriate kinds are constantly being rejected by journals and publishing houses because they don't meet standards of quality. I'm not concerned with these here (although the standards themselves can't really be defined until we know what kind of work they're to be applied to); I'm concerned with what kinds of writings are publishable--how these kinds are determined.
Ursula LeGuin has been quoted (Wood 1982: 18) as having said, "You must either fit a category or 'have a name' to publish a book in America." (which is why she chose to write works that would fit the category "science fiction"). I think the requirements for publication in professional academic journals are somewhat analogous, with the categories corresponding approximately to what may be called the "research paradigms" that are active at the time. In a word, my Notes are unpublishable because they do not contribute in any apparent way to any ongoing research paradigms. People in linguistics (and occasionally related fields) have read them, and sometimes agreed with what I've said, but these people's business is research, and there has been no apparent way that what I've said could be incorporated into their research--what I've said is (in that sense) irrelevant (even to someone who might find it intrinsically interesting).
The thing is: what I say in the Notes is pretty much irrelevant (as far as I'm aware) to all of the presently-ongoing research paradigms. Even though there has been a proliferation of specialized journals in recent years, there are none (as far as I can see) to which the Ethnolinguistic Notes might appropriately be submitted.
I want to make it clear here that this is not a complaint about the policies of journal editors and book publishers. I'm not trying to say that the world is treating my writings unfairly. What I am trying to say is that there's a conflict between our responsibilities for, on the one hand, the production of new knowledge and, on the other, the exposition and dissemination of existing knowledge. The production of new knowledge must (apparently) be governed by research paradigms. On the other hand, the exposition and dissemination of existing knowledge should be as free of them as possible.
I talked about this matter at greater length in a paper I published (if that's the right word) in 1979 in the University of Hawaii Working Papers in Linguistics, and rather than repeat myself, I'm attaching that paper here.
The paper is entitled "The dilemma of the disciplines". The "dilemma" is that between the responsibility of a discipline such as linguistics to conduct research, which apparently requires a sprouting of more or less independent research paradigms, and the responsibility for supplying knowledge about a whole sector of the phenomenal world--a whole sector of the frontier of knowledge (in this case, the sector occupied by human language).
My being able to include that paper here was made possible by Byron Bender, who scanned the original paper for me. The graphics are not as clear as I would wish, but I think they can be made out. I've retained the paper as far as possible in its original form. Some people objected to my term "knowledge management", which they felt implied manipulation. I'd probably change the term if I were re-writing, but I'm trying here just to reproduce the original as faithfully as possible.
To go to "The Dilemma of the Disciplines", click here.
Wood, Susan (ed.). 1982. The language of the night: Essays on fantasy and science fiction by Ursula K. LeGuin. New York: Berkeley Books. Back up
Return to George Grace's home page
to Ethnolinguistic Notes
to my personal page
Last updated 1 February 1996
THE DILEMMA OF THE DISCIPLINES(1)
George W. Grace
University of Hawaii
This paper is concerned with what may be called the management of knowledge within our society. In the general perception, knowledge management in the sense which I intend falls mainly into the domains of two institutional systems (as we may call them), science and the educational system (in this case, particularly the universities).
One might think of a schematic representation where the domain of the educational system is represented as a sausage shaped area with the least "scientific" kind of knowledge management (labeled "humanities") at one end, the most scientific (labeled "pure" or "hard science" or something of the sort) at the other, and an intermediate domain (which I will call "human sciences" and otherwise not attempt to characterize). If we superimpose a schematic representation of "science" such that academic science overlaps the academic sausage and extra-academic science extends beyond it to the left, we get something that somewhat resembles a distorted map of Africa.
Although there are formal institutions at other levels, knowledge management is entrusted primarily to a set of specialties-called (here, at least) "disciplines" (i.e., academic disciplines, scientific disciplines).
I take it to be apparent that all is not well with the knowledge management business. Within our own discipline of linguistics an uncertainty about directions can be observed. In the words of W. Haas (1978: 293), "Anyone surveying the contemporary linguistic scene will be struck by a curious wave of unrest and uncertainty, a softening of oppositions, a greater readiness to listen to one another on the part of 'schools' formerly exclusive, and an eagerness for new suggestions, however vague, as if the one thing sure were the inadequacy of the extant school-solutions."
However, the problem is more general. It seems to me apparent that our society--and the world at large--is in or near a moment of crisis, that fundamental changes of direction, reorderings of our institutions are in prospect--and the universities are quiet! Wherever the intellectual ferment is, the universities seem largely irrelevant to it.
I assume that the solution to this problem of limited relevance does not lie either in political activity (in attempting to employ our professional status [whatever it is worth] or our professional organizations as political instruments) or in "applied" research (shaping our endeavors to the questions that are framed by government institutions in their efforts to govern), but must ultimately be found in our knowledge management role itself if it is to be found at all. That is, the role of the knowledge management sector of society in solving the problems that face society today is to do what it can to see that a better understanding of the conditions of human life can be applied to the decisions that will ultimately determine the future of human life. As far as linguistics is concerned, I take it to be obvious that the possession of language is a very important aspect of the human condition, that an adequate understanding of the phenomenon language will be required for any adequate understanding of human nature, and that these understandings are prerequisite to the definition and implementation of realistic social objectives.
What, then, is it that is not well? The particular idea that I want to put forward here was suggested to me in part by an article in Newsweek by Mortimer J. Adler (1978). Adler was an associate of Robert M. Hutchins in his redesigning of the undergraduate program at the University of Chicago. Of their aims for undergraduate university education, Adler says, "...he and I thought that the undergraduate college could be emancipated from the paralyzing clutch of the graduate and professional schools. They, like major-league baseball clubs, tend to regard the college as little more than a bush-league feeder station."
It is not my purpose in quoting Adler to exposit or defend his philosophy of education. However, his remarks do point to a kind of schizophrenia which seems to afflict the knowledge management industry. This schizophrenia seems to have its roots in two conflicting models of the organization of knowledge management--of the role of the disciplines. It is to the conflicting demands of the two roles that I was alluding in entitling this paper "The dilemma of the disciplines".
One of the two models is what we may call that of "the frontier" or of "assigned sectors". It might be represented thus:
According to this model each discipline is assigned a certain sector along the frontier of knowledge. An implied corollary is that the entire frontier is partitioned among the disciplines so that any question that might arise falls within the domain of some discipline. Each discipline would have the responsibility to manage knowledge within its assigned sector and to make it available to what we may call its "clientele", viz.:
(1) people responsible for policy decisions in the society,
(2) representatives of other disciplines, and
(3) the public at large, but especially the rising generations (through the educationa1 system).
Managing the knowledge would involve exploiting every opportunity to obtain new knowledge pertaining to anything within the sector assigned to the discipline. It would also involve a serious effort to incorporate into some sort of integrated system all of the knowledge for which the discipline was responsible. This system should be designed to make readily available the best answers possible on the basis of existing knowledge to those questions which were most likely to be of interest to the clientele of the discipline.
The second model may be called the "paradigm" model. Both the model and the term are inspired by Thomas Kuhn's The Structure of Scientific Revolutions (2nd edition, l970). Although Kuhn was particularly concerned with what I am calling the "hard sciences", the model seems also to represent knowledge management practices found in other disciplines as well. Kuhn writes (1970: 45), "Effective research scarcely begins before a scientific community thinks it has acquired firm answers to questions like the following: What are the fundamental entities of which the universe is composed? How do these interact with each other and with the senses? What questions may legitimately be asked about such entities and what techniques employed in seeking solutions? At least in the mature sciences, answers (or full substitutes for answers) to questions like these are firmly embedded in the educational initiation that prepares and licenses the student for professional practice." He asserts further that "normal scientific research" is "a strenuous and devoted attempt to force nature into the conceptual boxes supplied by professional education" (1970: 5).
I picture the paradigm model as having the form of a Portuguese man-of-war, the visible nucleus of which is made up of a bundle of assumptions and what might be called a "work-style" (characterized by a stylization of questions thought worthy of investigation, styles of research activities, kinds of inferences thought justified by different kinds of evidence, writing style, etc.). Depending from the nucleus are the results of the research that derives from it.
The paradigm model differs from the assigned sectors model in several ways. (1) The domain of knowledge which it actually deals with is more arbitrarily defined, and is likely to be much narrower. (2) The emphasis is on research to the point that the other knowledge management functions sketched above are largely neglected. (3) In fact, at least in the human sciences the assumptions of such paradigms (which are often quite ingenious) tend to be perceived as acquired knowledge and to be represented as such to the clientele of the discipline.
The dilemma, then, is to decide which of the two models actually defines the role and responsibility of the disciplines. Is the discipline to be conceived of as responsible for a sector of the frontier, or is it to be identified with the research flowing from a certain paradigm ( or a small number of paradigms)? The two models have quite different implications, and, in the case of linguistics at least, there is evidence of a measure of schizophrenia.
It is the assigned sectors model that is tacitly being invoked when territorial claims are asserted. For example, I have known linguistic departments to oppose the introduction of courses dealing with some aspect of language by other university departments on the ground that, since they dealt with language, they belonged to the domain of linguistics. Similarly, one sometimes encounters the expectation--as in book reviews--that if an author who lacks linguistic credentials writes a book dealing in some way with language, it is most likely that serious flaws lurk somewhere within it whether or not they are immediately apparent. In short, it is possible to discover an assumption on the part of some that only a linguist is competent to discuss language.
On the other hand, it is the paradigm model that is being invoked when Martin Joos writes (1950: 703), ''...all phenomena, whether popularly regarded as linguistic...or not, which we find we cannot describe precisely with a finite number of absolute categories, we classify as nonlinguistic elements of the real world and expel them from linguistic science." Again, the following statement by Roger Brown (1958: ix) also depicts linguists as adhering to the paradigm model, "For many years linguists have absolved themselves of responsibility for the most difficult questions concerning language behavior by relegating such problems to the province of psychology, sociology, and anthropology."
In fact, it seems quite clear that the paradigm model presents a much more accurate picture of the discipline of linguistics as it exists today than does the assigned sectors model, despite the occasional expression of sentiments of territoriality by linguists. Although I believe this to be clearly true for linguistics, the situation in other human sciences varies considerably, and I do not intend to suggest that the same judgment would apply to others.
The assumptions involved in a paradigm are likely to be difficult to isolate and to make explicit. However, I believe that an effort to examine the assumptions on which present-day linguistic discussions rest would be very useful. I hope that the very rough and general remarks which I am about to make may stimulate more accurate and detailed analyses by others. Linguistics is described as the science of language. In the paradigm of linguistics, language is customarily assumed to be exhaustively partitioned into a set of entities called "languages". It is assumed that these entities are precisely defined, or, at least, that it is reasonable for most purposes to proceed as if they were. These languages are treated as if they were autonomous objects which maintain their distinct identities over time.
The central concern of linguistics has been the description of these postulated entities. More precisely, the central concern of linguistic theory has generally been taken to be the formulation of the metalanguage to be used in such descriptions and the elaboration of criteria by which particular empirical observations may be judged to authorize particular statements in the descriptive metalanguage to be made about some recognized language.
Although there have been a number of different descriptive traditions, all seem to agree in assuming that (apart from phonology) the description of each language divides into two parts, a dictionary and a grammar, so that each fact attributable to the language (as determined by the rules of inference) is to be construed as either a lexical or a grammatical fact. [Elsewhere (Grace, to appear) I have questioned that anything like so sharp a division can rightly be attributed to the language and also that a description in terms of just a lexicon and a grammar, in any reasonable interpretation of those words, can rightly be seen as coming anywhere near to being an exhaustive description of the language. However, those considerations are aside from my main point here.]
For many linguists, languages are "mental" phenomena, and their actual physical locus is in the brains of their speakers, with each individual having in his/her brain some sort of replica of each language that he/she speaks. In fact, the only existence which the language has is in the form of these replicas. Even linguists who avoid references to the mind are obliged to recognize that languages are transmitted by learning and that therefore the nature of languages must be subject to some kinds of constraints imposed by the operations of the human organism. It also seems clear that aside from aspects of pronunciation, most of the operations involved are the kind that we are accustomed to calling "mental". Therefore, it is to be expected that any tacit assumptions which we may have absorbed from our cultural tradition or elsewhere about the nature of the human mind (or "thought processes", "cognitive processes" or whatever label we want to use) are likely to influence our ideas about how speakers use languages, and following that, in what form languages are learned and held in memory.
If the foregoing observations are at all on the right track, it would seem to follow that an investigation of the assumptions about the nature of the mind and mental processes in modern Western culture might make an important contribution to describing and explaining the paradigm of linguistics.(2)
If we were to consider language in the frame of the assigned sectors model, what questions would take shape? Obviously, any answer at this point can at best be quite speculative. However, even some preliminary hints will lend some substance to this otherwise abstract discussion. In my book (Grace, to appear) I suggest three main topics. The following brief discussion of each may be indicative of the kinds of questions that might ultimately come up.
1. "Content form", or how ideas are represented by language. This would involve attempting to approach form through meaning, rather than starting from form. That would, in turn, involve some kind of attempt to characterize what is (to be) said independently of the way in which it is said (e.g., what is it that is common to an original statement and its translation or paraphrase? What is it that we know when we know what we have read or were told but do not remember in what language we read or were told it? How can a description correspond to what is described?, etc.). This problem is at the foundation of epistemology and any adequate grasp of the conditions on human interaction with external reality. It is the question par excellence of how language works.
There are several matters which have been raised in recent work or otherwise are of particular current interest that might be treated under this topic, e.g.,
(a) What are the cultural and cognitive consequences of (1) writing, (2) specifically alphabetic writing, (3) general literacy within a society, (4) printing? How does the competence required to read and write skillfully differ from the competence required for oral language use? Is language structure in the strict linguistic sense affected?
(b) How are languages changed in the process of "modernization" (which is one of the principal concerns of "language planning")? Is language structure in the strict linguistic sense affected? (A hasty check of the literature suggests that there are regularities, e.g,, (1) more abstract terms, (2) changes in derivational morphology one aspect of which is addition of or expanded use of affixes which nominalize verbs, (3) more conjunctions, (4) introduction of, or extended use of, passive constructions).
(c) Is it not the case that most new concepts are given designations that are "motivated"i.e., are constructed of meaningful elements in such a way that they have a "literal" meaning that characterizes the new concept in a particular way [e.g,, "prolife" society, oil "production", language "acquisition"]? How does this widely overlooked but very essential aspect of language--that it uses meanings to designate meanings--get used in propaganda and other attempts to mislead?
2. The second problem is that of linguistic competence, or the "idiolect" (as I have used the term; cf. Grace, to appear). What kind of theory of human beings accounts for what they can and cannot do as language-bearing animals? As I have indicated (cf. esp. fn. 2), although linguistics sometimes suggests that linguistic work casts some light on this matter, I believe that there are many misconceptions that we must shed before we will have much to contribute.
We should particularly ask the question of what the bilingual knows that is not a part of the description of either language.
3. The problem of the various kinds of patterning manifested in actual speech or writing. One such kind of patterning provides the basis for the identification of individual languages and the evidence upon which they are described. Other kinds of patterning have not been systematically sought out and studied. They might include those common to a (non-monolingual) linguistic community (e.g., cf. Gumperz and Wilson 1971) or to a linguistic area, those characterizing the "elaborated" and "restricted" codes of Basil Bernstein (e.g,, Bernstein 1964), or those characterizing "discourses" in the sense of Michel Foucault (1972), for examples.
The discipline of linguistics as it exists today is fairly accurately represented by the paradigm model. Partly as a result of its having chosen this style it has attained a high degree of sophistication. Its concepts are often not comprehensible to the uninitiated, its questions not those which could be formulated by the man in the street, its assumptions complex and themselves sophisticated. As a consequence the discourse of theoretical linguistics remains on a plane quite remote from everyday human concerns.
I would like to propose that there are other knowledge management needs which may be thought of as deriving from the assigned sectors model. In particular, I believe that there should be a continuing effort made by someone to identify and articulate the great questions about language as it relates to the human condition and a continuing effort to provide access by our clientele to our current knowledge as it relates to these questions.
For such efforts to be effective, I believe that we must be careful not to make the identification of questions contingent upon their "researchability". That is, it should be permissible to recognize and discuss problems even though we do not yet (or will not ever) know how to design a research project around them. Secondly, such efforts, if they are to be effective, must not be constrained by any requirement that they adhere to the assumptions of the current paradigm of linguistics. On the contrary, novel perspectives should surely be encouraged.
If we agree that such needs do exist, the dilemma that confronts us is whether the task of the discipline of linguistics itself should be reconceived so as to meet them or whether, on the other hand, we should, at the price of renouncing any claim to being the science of language, invite others to meet them.
Although I have focused upon linguistics through most of the discussion, I intend to suggest that the dilemma in one form or another exists for all of the disciplines--at least for all of those of the human sciences. It is also my perception of the matter that the assigned sectors model does fairly accurately represent the view of many of the public as to what the disciplines are supposed to be doing.
Finally, I do believe that major decisions confront human society, and that they will be influenced by people's understandings of the nature of humankind, of our natural environment, and of the conditions of our interaction with that environment. Surely, to the extent that our institutions of knowledge management throw useful light on these questions they will deserve not to be ignored.
1. This is a revised version of a talk given to the University of Hawaii Tuesday linguistics seminar in November 1978. (Back up)
2. I expect that it will eventually prove to be possible to identify a more or less coherent set of assumptions shared pretty much throughout the human sciences about thought processes. This may be thought of as our "model of the mind." I expect that our model of the mind will turn out to have had its main origins in prescriptive, rather than descriptive formulations--that it represents the rationalist view of approximately the 17th century of the kind of thinking that was called for by the new age of science that was in prospect. Subsequently, it has somehow come to be thought of as a model of ordinary mental operations (but with the now rather ancient cliché that children, mad persons, and primitives do not think properly).
It seems to me that to analyze this model into its constituent assumptions and to trace them to their sources would be of great value to any analysis of paradigms in the human sciences in general and in linguistics in particular. As one example, Chaim Perelman and L. Olbrechts-Tyteca (1969) attribute specifically to Descartes the responsibility for narrowing our conception of the nature of reasoning until only what has been called more geometrico reasoning (essentially modern Western 'logic') is recognized. In their words (1969: 1), 'It was this philosopher who made the self-evident the mark of reason, and considered rational only those demonstrations which, starting from clear and distinct ideas, extended, by means of apodictic proofs, the self-evidence of the axioms to the derived theorems.'
I have come increasingly to suspect that the notion of 'extension', which seems to have first been clearly set out and labeled in the Port Royal Logic (cf. Bocheñski 1961: 259, Kneale and Kneale 1962: 318), has had an importance in our models of language and of mind that should be brought out into the light.
Is it not a noteworthy fact that a calculus of classes in extension should have received (and in all seriousness) the title: An investigation of the laws of thought (Boole 1854)? Since I want to maintain that the concept of 'intension' is a derivative of that of extension, one might mention as another noteworthy fact that a book devoted primarily to experimentation on 'concept attainment' should a century later have been entitled: A study of thinking (Bruner, Goodnow, and Austin 1956). [Concept attainment is defined (ibid. 22) as ' ...the process of finding predictive defining attributes that distinguish exemplars from nonexemplars of the class one seeks to discriminate.']
The point is that the concept of intension, as it is usually employed, rests on the assumption that in some reasonable sense each lexical item can be said to have an extension, that that extension is definite or at least that it is reasonable to proceed as if that were the case, and that what we want to call the 'meaning' of the lexical item is crucially rooted in that extension. The intension of a lexical item is, in fact, nothing more than a theory explaining the presumed extension--i.e., principles which define the extension at any given moment.
I want to propose that the precise definition of a term designed to make possible a clear discrimination between the things which the term properly applies to and those which it does not is a characteristic of scientific language and not of ordinary language. A crucial part of the scientific enterprise is being able to determine whether particular theorems have been confirmed as true or disconfirmed as false when confronted with the evidence of empirical observations. Thus, a precise definition of the conditions under which a particular proposition holds true is of the first importance in science.
The requirements of ordinary language use are quite different, and I submit that the rationalist model of the mind tends to distract us from the nature of ordinary language functioning. (Back up)
Adler, Mortimer J. 1978. The disappearance of culture. Newsweek, August 21, 1978, p. 15. (Back up)
Bernstein, Basil. 1964. Elaborated and restricted codes: Their social origins and some consequences. American Anthropologist vol. 66, no. 6, part 2, pp 55-69. (Back up)
Bocheñski, I[nnocentius] M. 1961. A history of formal logic. Notre Dame IN: University of Notre Dame Press. (Back up)
Boole, George. 1854. An investigation of the laws of thought. London. (Back up)
Brown, Roger. 1958. Words and things. Glencoe I. Free Press. (Back up)
Bruner, Jerome S., Jacqueline J. Goodnow, and George A. Austin. 1956. A study of thinking. New York: John Wiley & Sons. (Back up)
Foucault, Michel. 1972. The archaeology of knowledge and the discourse on language. New York: Harper Colophon Books. (Back up)
Grace, George W. To appear. An essay on language. Columbia SC: Hornbeam Press. (Back up)
Gumperz, John J. and Robert Wilson. 1971. Convergence and creolization: a case from the Indo-Aryan/Dravidian border. in Dell Hymes (ed.), Pidginization and creolization of languages. London: Cambridge U. Press, pp. 151-67. (Back up)
Haas, W. 1978. Linguistics 1930-1980. Journal of Linguistics 14: 293-308. (Back up)
Joos, Martin. 1950. Description of language design. Journal of the Acoustical Society of America 22: 701-708. (Back up)
Kneale, William and Martha Kneale. 1962. The development of logic. Oxford: Clarendon Press. (Back up)
Kuhn, Thomas S. 1970. The structure of scientific revolutions, 2nd edition. Chicago: U. Chicago Press. (Back up)
Perelman, Ch[aim] and L. Olbrechts-Tyteca. 1969. The new rhetoric: A treatise on argumentation. Notre Dame IN: University of Notre Dame Press. (Back up)
Return to top of Dilemma of the Disciplines
First published 1979 (University of Hawaii Working Papers in Linguistics 11(1): 1-10)
World-Wide Web version: 18 January 1996
Return to George Grace's home page
to Ethnolinguistic Notes
to my personal page
Last updated 9 February 1996
APPENDIX (23 November 1996)
[A short time ago I chanced upon the following brief piece which was, as it explains, intended as a preface to ELN, series 3, no. 2. I'm appending it to the preceding attempt to explain myself in the hope that it will add something complementary. I should point out that the book entitled Ordinary language to which it refers was never published. Some of its content was eventually incorporated into The Linguistic Construction of Reality, which was published in 1987.]
PREFACE: TO EXPLAIN MYSELF
George W. Grace
(About December 1981)
[This "Preface" was originally intended to accompany ELN, series 3, number 2.]
I have been told from time to time that nobody can understand the stuff that I have been writing lately. The writings in question consist of things that I have written in the last five years beginning with the original series of notes which I called "Ethnolinguistic Notes" (ELN) in 1976. The original ELN series was finished in 1977 and incorporated into a book (which eventually appeared as Grace 1981a). The second ELN series has been incorporated into the manuscript for another book (Grace 1981b). The present paper constitutes the second note of the third ELN series. That which has proven difficult to understand is, I presume, any or all of the above (plus, perhaps, other things which I have written during the same period).
But what kind of explanation do I need to give to make these writings understandable? What does "understand" mean in this context? I suppose that "understanding'' here should not be understood in the linguistic sense (what did he say?), but rather in the sociological sense of "What did he mean by saying that?" (i.e., what kind of intervention in the ongoing [professional? scientific?] discourse are these acts of writing?). That is, I believe that what is really required is that I provide what has been called a "story" for these acts of writing. The story should give an account of where they are coming from (i.e., what steps and what motives 1ed to them) and where they are headed (what effects they might lead to). Of course, any story may be found not to be really understandable until it is placed in the context of a bigger story, but I will try to get away with providing a little story.
As to where the writings come from, I might begin with how I became increasingly uncomfortable about the fact that there were problems in Austronesian linguistic subgrouping and reconstruction which I could not resolve. I was uncomfortable because it seemed to me that nothing in our assumptions about the processes of change and differentiation could account for the problems I was encountering. For some years I attempted to quiet my misgivings with the pious platitude that the solution lay in improved descriptive data (even though I suppose I always realized that that platitude had not been arrived at through serious disinterested consideration). Anyway, it must surely be clear by now that if the solution to, for example, the position of Yapese lies in descriptive facts about the language which we do not yet possess, we must depend upon pure blind luck for the discovery of those facts, because we certainly haven't the slightest notion what kind of facts we should look for.
Anyway, I was gradually forced to the conclusion that perhaps the problem lay in our assumptions. Reconstruction is based on assumptions about the processes of linguistic change and differentiation. We consider the states of contemporary related languages and attempt to infer how these states could have resulted from a single linguistic system in the past through the operation of linguistic change and differentiation as we assume them to work.
I, therefore, began trying to identify and evaluate my own assumptions--assumptions, however, which seemed generally to be widely shared.
I was led gradually to the general question of what conditions we actually assume for linguistic diachrony and which of these conditions might on occasion fail to obtain. This question led in turn to the question of what we assume a language to be and how the validity of the assumptions we make might be dependent upon factors in the context in which the language is used. Subsequently, I also became aware that assumptions might be valid for some purposes but less valid or even entirely invalid for others. I came gradually to realize that the object of study of a discipline such as linguistics is something which the discipline itself has created-e. g., "language'' as it exists in the professional view of linguists is a socially constructed reality. What we talk about is a model which is supposed to be representative of something in the real world. That is, there is presumably something in the real world which our models of language are supposed to represent, and in our professional work we assume real language to have the properties of our models of language. To put it another way, we talk about (real) language as if it had the properties of our models.
Does it matter that models are not identical with realitv? Yes, sometimes. It matters most obviously when a question arises which is not the kind of question the model was designed to deal with. I believe ample exemplification of this problem could be found in the history of attempts to base language teaching on what were at any given time recognized as the latest insights from linguistics. Or consider the concept of the phoneme. This concept has proved extremely useful in the analysis, description, and comparison of sound systems. However, I do not believe that there is any reason to think that people, unless they have been exposed to alphabetic writing, actually segment their speech forms in the way the phonemic model would require (for more discussion, see Grace 1981b, Chapter 4). But when could that actually matter? Well, for example, suppose someone who is trying to compare the competence of a young child or of an animal of another species with the linguistic competence of the human adult appeals to linguists for a characterization of what that competence consists of. In such a case to take the phoneme model at face value and assume that human adult competence involved operating with such letter-like units of sound would be anything but helpful (assuming, of course, that I am right that phonemes do not correspond to anything real).
What I'm doing, then, is trying to dig beneath the assumptions --trying to figure out what we would know, or seem to ourselves to know, about language if we could look at it afresh, setting aside the presuppositions of our models. Doing something like that is not the sort of thing which would be called scientific research, of course. I presume that it is neither scientific nor research, but it is an effort to get at the truth (of course, any truth so arrived at is no more objective truth than anything which preceded, they are all just constructed realities--to succeed is to construct some sort of approximation to the truth).
It is at this point that the second aspect of the story, the account of what effects all of this might lead to, begins. Assuming for the sake of argument that the writings in question are not total nonsense--that their effort to get at the truth has, at least in some limited way, been successful--what is one to do with whatever results they have produced? Whatever truths they may have provided may appear useless, as such, to the profession. It may seem that they are of professional interest only if they can serve as leads for future research, inasmuch as research is the business of the profession. And this, I think, is the question which the second part of the story should attempt to answer if it is to answer the charge of ununderstandability: what implications, if any, do these writings have for research?
My problem is that that is not a question that I feel capable of answering in any very direct way. It is not that I think there are no implications for research; it is, rather, that I do not have any very clear and accurate idea what those implications are. I would prefer to substitute my own question and attempt to answer that. My substitute question would be: How is this stuff relevant to linguistics? My answer would be that what I have been doing is attempting to make contributions to what I will call "linguistic theory". But that also needs explaining.
I tend to think of every discipline as having, implicitly at least, a branch devoted to what I call its "theory". "Theory" in the sense which I intend is concerned with determining what ends the discipline exists for, with imagining and weighing possible strategies for attaining those ends, and with evaluating the effectiveness of the strategies currently being pursued. I would say that such a branch would always exist implicitly even if the practitioners of the discipline never paid any attention at all to it (which has not been the case in linguistics), because it is so obviously one of the responsibilities incumbent upon the discipline.
It is to linguistic theory so conceived that I would assign what I have been writing. In those writings I have tried to raise questions about what linguistics is for. I have tried (especially in 1981a) to show that some of our current strategies are not very well designed for achieving some of the purposes which I think we should acknowledge as ours, and I have tried (especially in 1981b) to develop alternative ways of looking at language (to construct different realities to bear the name "language") which I hope could expand the array of objectives that linguistics could effectively pursue-
I think, particularly, that we must expand our scope. Too many of the questions that people will want to ask about language cannot reasonably be put to our present models. I am not proposing that any of the lines of research currently being pursued under the name of linguistics should be abandoned, but I do propose that the profession as a whole must prepare itself to deal with a wider range of questions or risk losing the status which it has so cherished--the status of leader of the language sciences.
Grace, George W. 1981a. An essay on language. Columbia SC: Hornbeam Press.
--------. 1981b. ordinary language. ms.
|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|
Last updated 23 November 1996
Typo corrected 16 March 2000
Minor modifications made 28 December 2006
More typos corrected 1 August 2009