Friday, February 28, 2014

Consilience in Linguistics

[This note is a spin-off from our initial essay, Consilience, along with its sidekick Consilient Connections (distinguishing vertical from horizontal consilience);  and is sibling to such other special mini-studies as Consilience in Psychology, Consilience in Mathematics [currently closed for reconstruction], and the purely satirical Consilience and Cognitive Science.]

The whole notion of “Consilience [with]in …” a specific field  is contrary to the Wilsonian/vertical-reductionist conception of the grand principle.  Instead we are noticing similarities in various regions of a field, which may have no analogues outside that field, nor follow synthetically as consequences of other principles outside that field.  Accordingly, the various discipline-specific mini-essays may have little in common, let alone contribute to a grand TOE of the sciences.  They are more in the spirit of the various volumes of the “Bobbsey Twins”.   I do plan to stop, though, somewhere short of “Consilience in Stamp-Collecting” and “So!  You Want 2 B a  Consilientist ”.

Indeed, much of the thrust of linguistics, over the past nigh-on one hundred years, has been an assertion of its disciplinary autonomy -- not a branch of sociology, or history, or individual language studies -- even with respect to its ancestor philology, and further with regard to various components of linguistics itself, which one hoped  neatly to delineate.


… the need felt by many  to keep linguistics an autonomous discipline, to prevent it from slopping over  into psychology, sociology, neurophysiology, etc.
-- Roger Lass, On explaining language change (1980), p. 121

This depreciatory rhetoric of “slopping over” is diametrically at variance with Wilson’s utopian holistic vision for Unified Science.

Indeed, over the past  close-on to a century, or at any rate  over a good fifty years during the heyday of the growing prestige of the field, other disciplines have looked to linguistics more than it to them.  As:

The appeal of linguistis methodologically  to anthropology is in part because of its achievement of units at once concrete and universal.
-- D. Hymes & J. Fought, American Structuralism (1975), p. 153.

In terms of the sociology of science, this influence is reminiscent of the Physics Mesmerism earlier spawned by the pizazz of the field that invented The Bomb.

That, in fact, may explain a lot.   For the study of syntactic and semantics structures plainly finds nothing of ready use for it from current physics (“The Propositional Island Condition:  a String-Theoretic Approach”), whereas the neighboring scholarly fields of psychology, anthropology, sociology, etc., have been -- not to put it too unkindly -- not especially successful of late.  They do not, so to speak, have The Bomb;  so linguistics has been free to go its own way.

The linguistic scene in Europe during the nineteenth century, and up to about Weimar, had a decidedly less isolated or self-sufficient feel, with the same Grimm brothers who wrote straight sound-law Germanistik and who launched the massive Wörterbuch, likewise collecting their equally-celebrated work of folklore,  Kinder- und Hausmärchen.   Saussure, with his penchant for dichotomies and abstractions, culminating in his purely cerebral reconstruction of the Indo-European laryngeals  not visible in the extant texts (a ghostly forerunner of the key notion of “empty categories” in the Chomskyan school), were atypical (notice that he published precious little in his own lifetime, his main influence being posthumous, his penchants  more to the liking of a later time).   There were, to be sure, occasional Declarations of Linguistic Independence similar to those which later characterized American structuralism and its progeny --

Linguistics must be regarded as an independent science, not to be confused with either physiology or psychology.
--  Baudouin de Courtenay (1871)

but more characteristic were the plumbing of ancient literatures, the quasi-folkloristic dialectological spelunkings in the hills and hamlets of the hinterland, and so concrete, sleeves-rolled-up a tendency as that of Wörter und Sachen.   As a Romance philologist once put it, emphasizing the Unentbehrlichkeit of the sister-disciplines for linguistics,

Man komme also nicht mit dem beliebten « ne supra crepidam »;  das gilt wirklich nur für Schuster.
-- Hugo Schuchardt, (1892), in Leo Spitzer, ed., Hugo Schuchardt-Brevier (1921; 2nd edn. 1928), p. 105

These humanistic traditions were carried on at Berkeley by my Doktorvater Yakov Malkiel, a White Russian refugee and time-capsule of Mitteleuropäische Kultur, whose journal Romance Philology (and it really was as deeply his, as The Rambler was Doctor Johnson’s) featured both linguistic and literary articles side by side.   It was this tradition I relished, and which informed my dissertation (later published as The Semantics of Form in Arabic, in the Mirror of European Languages), to the virtual exclusion (alas, for professional advancement) of the then-current trends.   It is a Kulturgut deeply to be savored, in centuries to come, long after the squabbles of Amerian Vietnam-era linguistics shall have faded into the mists of history.


The fiercest assertion of linguistic independence -- what we might dub the thesis of the Idiosyncrasy of Human Language (it obeys principles peculiar to itself, not generally shared among other features of animate beings) -- has come from the Chomsky camp (that bastion of the intellectually fierce and rhetorically ferocious)

The proper perspective of linguistic independence is given a trenchant formulation by Chomsky:

How can a system such as human language arise in the mind/brain, or for that matter in the organic world, in which one seems not to find anything like the basic properties of human language?  That problem has sometimes been posed as a crisit for the cognitive sciences.   The concerns are appropriate, but their locus is misplaced:  they are primarily a problem for biology  and the brain sciencies, which, as currently understood, do not provide any basis for what appear to be fairly well-established conclusions about language.
-- Noam Chomsky,  The Minimalist Program (1995), p. 2

[Sidenote:  This stance exactly parallels our own with respect to the (possibly not entirely unrelated) question of Free Will, in our diatribes against the eliminative materialists, who, failing to discover Free Will in a test-tube, declare that it must therefore not exist.  Morality they seem to have discerned cringing at the bottom of  an Erlenmeyer flask, and identify it with oxytocin, or oxycontin, or some damn thing.   Our retort:  Free Will (and with it,  possible prolegomena to morality) stands unspattered;  good luck with whatever it is you’re doing, and ever saying anything useful about these.]

But similar American declarations of disciplinary independence go back a ways:

The history of structural linguistics in the United States  can be readily interpreted in terms of three successive (and overlapping) movements:

(1) the movement for autonomous study of language, i.e., for a profession of linguistics;
(2)  the movement for autonomous study of linguistic structure;
(3) the movement for autonomous study of grammar (syntactic structure)

Chomsky’s classification of linguistics as a branch of cognitive psychology, like Hockett’s classification of linguistics as a branch of cultural anthropology, does not imply any less of autonomy.
-- D. Hymes & J. Fought, American Structuralism (1975), p. 227-8

Back when I began at university,  linguistics as a discipline  was still largely at stage (1).  At Harvard College, they didn’t allow you to major in Linguistics simpliciter -- that was apparently perceived as being too narrow, like majoring in sausages or non-Abelian groups.  It had to be a combined major, say, Linguistics and Math.  (That is not necessarily a bad thing:  such disciplinary spread virtually forces you to think ‘consiliently’.)  I actually did briefly consider doing that, owing to the personal magnetism of two visiting luminaries:  George Lakoff and Noam Chomsky.  And while this-all takes us right off the topic of Consilience,  and back to the sociology of the wacky Sixies, perhaps the anecdotes may be worth retailing for their own sake.

It was my sophomore year (in many ways, my sophomoric year).  The campus was in turmoil, partially and fitfully shut-down.  The center had not held,  and we were whirling.
            Into this maelstrom  stepped  Noam Chomsky, meeting with Harvard undergraduates,  not to lecture, but to hear us out.   Already he was better known as a political commentator than as a linguist (I had eagerly read his foresightful American Power and the New Mandarins);  a small but respectful student audience  gathered at his feet -- or rather, he at ours:  for the students had occupied the available sofas and chairs,  and he himself sat down on the floor with the rest of the overflow.  He spoke softly, when he spoke at all;  listening, thinking, listening …
What he said that day, I don’t remember (how often are our memories  wordless, alas!)  but his sense or essence of open-mindedness, yet centeredness (most of us  were neither)  powerfully remains.    I left the meeting  feeling  that we student activists were truly juvenile, even if  largely right on the central issue of the Vietnam War;  and have him to thank that, thenceforward, I may have been  marginally less of an imbecile  than would otherwise have been the case.

Where Chomsky was all calm -- Jovian -- Lakoff was all energy -- Mercurial -- and tended to create a fevered atmosphere wherever he bustled -- not unlike that at campus protests and SDS meetings.  It was attractive, in a way, to an undergraduate whose veins flowed full with the sap of the Zeitgeist. 
However, I had, before my freshman year, renounced my original intention of majoring and English with a view to becoming a writer, penitentially donning the sackclothlike labcoat of a chemistry major,  in part because I knew myself to be subject to the Siren-calls of pride and cliqueishness  that went with the literary life.  And in Lakoff and his entourage (for there is no other word for the cloud of virtual particles that surrounded him), that factor was much in evidence.  So I drew back.  (Years later, at Berkeley, our paths would once again anastomose.)

What really settled it, though, was the practical reality of what it would mean to take courses from the permanent Harvard faculty (Lakoff was just passing through, and Chomsky taught over at M.I.T.)  I lasted just two meetings of the introductory course taught by droning Professor Undertaker (as I thought of him;  the real surname lies not too distant in phonological space).

Senior year, I applied to graduate school in math at Stanford and Berkeley.   When both accepted, I asked Lynn Loomis which he would recommend.   He was very practical:  no finicky attempts at comparing this professor with that, just:  Berkeley is bigger.  That proved wise advice, for exactly the same meta-reason for which I turned down Yale (whose English department had been making beckoning advances) and chose Harvard because, well, what the hell.  Once again, I wound up pursuing a course of studies entirely different from that in which I had applied.  After dropping out of math from impecunity, I stumbled upon a much more inviting introductory lecture-series by Charles Fillmore of Berkeley linguistics, and there was a thriving department ready to take me on;  whereas Stanford, at that time, didn’t even have a full-fledged Linguistics Department, just a Program within English.   So it looked as though, providentially, I had landed in the right place.

Yet how often have I dwelt in retrospect, how different my intellectual life (and with it, life tout court) might have been, had I instead found myself across the bay at Stanford.  For, despite its shoestring beginnings, by the time I received the doctorate (1981), Stanford had burgeoned into the M.I.T. of the West, whereas the Berkely department -- like the surrounding town -- was falling victim to centrifugal forces, partly political, and partly personality-driven, but more particularly, in the case of Linguistics, because the Rota Fortunae had made another half-turn, resurrecting MIT-style linguistics to a second run of pre-eminence, and sending the fortunes of Berkeley-style Generative Semantics down, down, down…   Fuit Ilium!

~ ~ ~

Below, a miscellany of consilient influences.


The syntactic cycle (cyclic ordering of rules in syntax) has been called “analogous” to the phonological cycle (e.g. Noam Chomsky, Language and Mind (1968), p. 39).   This, however, is  from a disciplinary standpoint  quite a domestic affair, and need not be ‘consilient’ with anything outside linguistics.   And indeed, as Chomsky remarks later in the same pamphlet (p. 66), “One cannot expect structuralist phonology, in itself, to provide a useful model for investigation of other cultural and social systems.

A word of caution.  By saying that a theory of sense ‘breaks down’ the complex ability of linguistic productivity,  I do not mean to suggest that the explanation must be reductive, but only that the theory of sense must articulate [unpack, make plain] the structure of complex ability.  If the ability is best explained by reducing it to other abilities, the theory of sense says what these are … If the ability is best explained holistically, … then it explains how the ability functions holistically …
-- Norbert Hornstein, Logic as Grammar (1984), p. 127


The whole thrust of modularity is that, while tout might more or less se tenir , within  a given module, the modules themselves are connected, in some cases, only the way the wheels and the engine and the windshield of a car are connected, in subserving a larger whole.

Ctr. a sort of ideological purity:

By strongly syntactic, I mean  not only that the theory is developed using syntactic notions,  but also that the principles invoked, and the distinctions advocated, have no natural semantic analogues.
… In fact, a stronger point can be made:  the natural distinctions and generalizations that a natural semantic theory of interpretation would [consiliently] make available  lead us to expect configurations of data that are simply not attested.
-- Norbert Hornstein, Logic as Grammar (1984), p. 102, 4

And,  more polemically, emphatically rejecting any extradisciplinary aid:

An extreme position was taken by Kurylowicz:  “One must explain linguistic facts by other linguistic facts, not by heterogeneous facts. … Explanation by means of social facts is a methodological derailment.” (1948).  For Kurylowicz, even the influence of other languages was irrelevant:  “the substratum theory has no importance for the linguist”.
-- U. Weinreich et al.  in W. Lehmann & Y. Malkiel, Directions for Historical Linguistics (1968), p. 177

All of this -- taken as  a program rather than a description -- is strictly unexceptionable, apart from the final “the” in front of the word “linguist”.  With that restrictive article, the passus passes over into the overweening;  as who should say,  “The wearyingly multifarious  and mostly insensate behaviors  of non-penguin species  is of no concern to The Zoologist.”
Contrast, indeed, in the very same volume, the opinion of one of its editors (Y. Malkiel, p. 28): 

In languages where “progressive” and “conservative”, “aristocratic” and “rustic” variants are suggested by differences in form, no truly satisfactory interpretation is conceivable  without an equal share of attention  granted to the social matrix.”

Along with many other stances that could be cited. (“Pike’s tagmemics  expressly linked to an analysis of cultural behavior as a whole”).


In post-Chomskyan linguistics, hardcore practitioners have usually been at pains to distinguish concepts and processes peculiar to the human language faculty, from those obtaining elsewhere, be it in bee-langue, or the “language of flowers” or “the Brain’s Inner Language” (this, from an article in this morning’s New York Times), or neuranatomy or chemistry or physics or -- horrendo referens -- Psychology with its “general learning strategies.”  By contrast, generativists revel in any apparent structural or procedural harmony within their discipline -- striving, for example, to see how much material can be gleichgeschaltet under the sway of the Empty Category Principle -- the awe-inspiring ECP (the Great and Powerful).


Rules relating S-Structure to LF are of the same kind that operate in the Syntax:  Move alpha.
-- Norbert Hornstein, Logic as Grammar (1984)

Outsiders are unlikely to gasp at this last, nor to marvel about the Unreasonable Effectiveness of Move-alpha.  (“alpha” here stands for:  anything at all.  A slightly more general formulation  would be Stephenson’s Rule:  Do Stuff.)


It is desireable for the overall theory to be articulated into strata, and for certain general notions or procedures to recur at different strata, at least masking their differences in basic material, and perhaps providing a genuine generalization.    Thus:  segmentationcompositionality; and projection.  E.g. (Hornstein):  The Projection Principle:  the Theta Criterion applies at D-structure, ((S-structure)), and LF  (these being the antiseptic rechristening of the old deep structure, surface structure, and logical form, which proved too exciting for the peasantry and had to be withdrawn from general circulation).


The technical tools for dealing with “rule-governed creativity” [basically, a synchronic system of limitless productivity] as distinct from “rule-changing creativity” [diachronic evolution] have only become readily available during the past few decades, in the course of work in logic and foundations of mathematics.
-- Noam Chomsky, “Current Issues in Linguistic Theory”, repr. in Jerry Fodor & Jerrold Katz, eds., The Structure of Language (1964), p. 59


There are books and papers that speculate about the evolution of human language while studiously ignoring all of linguistic research
-- Paul Bloom, reviewing LINGUA EX MACHINA:
Reconciling Darwin and Chomsky With the Human Brain.
By William H. Calvin and Derek Bickerton.


Appendix: Historical references to collaboration or lack of it, between linguistics and neighboring fields:

Summing up the wide-ranging work of the philologist and etymologist Hugo Schuchardt (floruit 1864 - 1927, magnum jubilaeum), Leo Spitzer writes:

Schuchardt hat nicht nur über fast alle allgemeinen Probleme der Sprachwissenschaft  nachgedacht, sondern alle irgendwie zur Linguistik peripherisch gelegenen Gebiete des Lebens  abgesucht, von der Sprachwissenschaft als Zentrum aus  seinen Beitrag der Lösung der drängenden Lebensprobleme  gegeben.
-- Leo Spitzer, ed., Hugo Schuchardt-Brevier (1921; 2nd edn. 1928), p.

Schuchardt himself, however, in places posits the wisdom of keeping linguistics sauber getrennt:

Hamitische Sprachen sind solche die von Hamiten gesprochen werden, oder Hamiten sind die  welche hamitische Sprachen reden.  Jenes ist die anthropologische Erklärung, dieses die linguistische … Mißverständnissen kann nur dadurch vorgebeugt werden, daß Linguistisches und Anthropologisches strengstens auseinandergehalten werde.
-- review (1912) of  C. Meinhof, Die Sprachen der Hamiten; repr. Leo Spitzer, ed., Hugo Schuchardt-Brevier (1921; 2nd edn. 1928), p. 334


Referring to the theoretical-semantic efforts of the pre-eminent 19th-century psychologist Wilhelm Wundt:

Wundt’s failure seems to have discouraged others from taking up the matter in earnest, for during the last 30 years, no one has made public any system of semasiology worthy of serious consideration.  Semantic work from 1900 to 1930  has been characterized by an astonishing and highly regrettable lack of contact and collaboration between psychologists and philologists.
-- Gustav Stern, Meaning and Change of Meaning (1931)


Contra sociological-anthropological structuralisme:

… eager souls who think they can easily borrow  and fruitfully adapt  a formal model  from simpler and perhaps more fortunate fields such as linguistics  or even phonetics.
-- Ernest Gellner, Contemporary Thought and Politics (1978), p. 122

(Here he means rather phonology;  it is phonology, and not phonetics, that is structuralist.)


Where does linguistics fit in -- among the humanties, or the sciences?

Um die Sprachwissenschaft   haben sich bekanntlich  die Natur- und die Geistes- (order Geschichts-)wissenschaften gerissen,  wie in der mittelalterlichen Legende  die Teufel und die Engle  um die Seele des Menschen.  Jetzt pflegt man sie  ihrem Inhalt nach  zu den Geisteswissenschaften,  ihrer Methode nach  zu den Naturwissenschaften zu zählen;  mit fast dem gleichen Rechte  könnte man das Umgekehrte tun.
Ich halte an der Einheit der Wissenschaft  fest,  und vermag  beispiesweise  zwischen Biologie  und Sprachwissenschaft  keine tiefere Kluft wahrzunehmen  als zwischen Chemie und Biologie.
-- Hugo Schuchardt, (1892), in Leo Spitzer, ed., Hugo Schuchardt-Brevier (1921; 2nd edn. 1928), p. 105


No comments:

Post a Comment