psycholinguistics and plato's paradoxes of the menopsycholinguistics and plato's paradoxes...

Psycholinguistics and Plato's Paradoxesof the Meno

WALTER B. WEIMER Pennsylvania State University1

Two fundamental problems in both psychology andphilosophy concern the nature of knowledge andthe nature of our acquisition of knowledge. Nomatter how "pure" it may be in research interestand theoretical intent, psychology ought to be veryapplied in the sense that it should seriously attemptto answer our fundamental questions concerning thenature of human knowledge and the process bywhich it is acquired. Yet, despite the spectacularsuccesses experimental psychology has had in itsattempt to understand the phenomena of its domain,it remains to be seen whether the answers advancedto fundamental problems such as the two above areadequate, or whether they are indeed answers at all.

This article explores two problems of knowledge,which take the form of paradoxes, from their originsin Plato's Meno to their reemergence in contempo-rary philosophy and psychology. I trace thependulum swing of intellectual fashion from Plato'sattempt to solve the paradoxes with some ingeniouspostulations concerning the nature and workings ofthe human mind, through to Aristotle's (and themajority of contemporary thinkers') attempts todeny the Platonic machinery and the solution itenvisages, and conclude with the resurrection ofPlatonic doctrine in psycholinguistics. Runningthroughout is the rather disheartening theme thatwe have not learned much about these problems insomewhat over 2,000 years of reflective thought.That is, my task is to convince one that the Platonicsolutions, inadequate though they may be, are still

1 A number of colleagues and students have greatly im-proved this article by their constructive criticism. Thoseto whom the author is most indebted are C. N. Cofer, T.Halwes, D. S. Palermo, and N. P. Young. Obviously, theycannot be held responsible for the final result.

Requests for reprints should be sent to Walter B. Weimer,Department of Psychology, Pennsylvania State University,417 Psychology Building, University Park, Pennsylvania16802.

the best available. Thus, I shall defend Plato in-directly (a) by showing that his doctrines are bothintelligible and "scientific" today and (b) by indi-cating that opposing theories are quite incapable ofproviding adequate solutions.

Plato's Problems in the Meno

It has long been a favorite philosophical pastime topropose the true problem or paradox that Plato in-tended the Meno to portray, and then to supplythe true resolution of that problem. It is not mypurpose to engage in this fruitless game of truePlato exegesis and scholarship: there is a case tobe made for many legitimate problems in the Meno,and undoubtedly some are more important, andmore difficult, than others. My concern is ratherwith two epistemological problems that are funda-mental in the sense that their proposed solutionshave never been less paradoxical than the problemsthemselves.

Although they may or may not be the "true"paradoxes of the Meno, I wish to argue that theones examined are genuine problems for both con-temporary philosophy and psychology, and thattheir resolution, in one or another manner, must bea prime task of both disciplines. These problemsare, first, the nature and role of abstract entitiesin knowledge and the learning process, and second,the "productivity" or "creativity" of behavior. Letus develop them by considering a stylized versionof their presentation in the Meno.

The dialogue begins with the questions, "Canvirtue be taught? Or is it acquired by practice?If neither, then how?" This is traditionally ajumping-off point for inquiries into ethics and valuetheory, or into the problems of education. But itshould be noticed that virtue, or justice, etc., are"abstract entities." And Plato, speaking through

AMERICAN PSYCHOLOGIST January 1973 15

Socrates, protests that he does not know the mean-ing of such abstract terms. Yet, both Socratesand Meno feel that they can recognize instances of,say, virtuous behavior. For any given behavior,they can tell whether it is or is not virtuous. Al-though they do not know what the abstract conceptmeans, and cannot define it, they can identifyparticular instances of it.

But now the perplexity arises: How can one knowa part of virtue (an instance of virtuous behavior)without previously knowing virtue in the whole(abstract)? How can people know a part of some-thing they do not know? Socrates and Meno admitthat such a result cannot obtain. And yet theiranalysis demands that such a result obtain, forotherwise we could not know anything. The di-lemma takes this form: A man cannot inquireeither about what he knows, or about that whichhe does not know; for if he already knows, thereis no need for inquiry, and if he does not know, hedoes not know the very subject about which he isto inquire. This is the paradox of knowing in theMeno: We cannot learn (come to know) anythingunless we already know (have learned) it.

The second problem arises from the famous pas-sage in which Socrates demonstrates seemingly in-nate and unknown capacity or knowledge in Meno's"uneducated" slave boy. This is the passage inwhich Socrates, the original master of the Socraticmethod, succeeds in getting the slave to prove atheorem in geometry (that the square twice aslarge as a given square has a side equal to thediagonal of the given square), despite the boy'sprotestations that he does not know the answer.By reiterating simple facts that the slave admitshe does already know, Socrates has him prove thetheorem by himself, a feat the boy had thought wasbeyond his powers. But now the problem arises:how can one be aware, or have knowledge, of thingsof which one is not aware? How can one exhibit aknowledge of things for which one's prior learningexperience has given no preparation? This is theparadox of learning (or productivity or creativity):the novel but appropriate behavior that constitutes"creativity" is not based on past experiences.Whenever we learn something new, that knowledgecannot be based on learning (prior experience).

The two paradoxes dovetail in a distressing man-ner, because of the obvious relation between learn-ing and knowing. We cannot learn anything newunless we already know it (by some other means),

and we cannot know anything unless we have al-ready learned (come to know) it. How, then, arehuman knowledge and learning possible?

Plato's Solution: The Doctrine of Formsand the Doctrine of Anamnesis

Plato considered these "paradoxes" spurious. Hehad a formulation of the nature of knowledge anda theory of learning that rendered these paradoxicalresults explainable. His theory of knowledge ac-corded a fundamental role to abstract entities, the"essences" of things (Platonic forms), and it heldthat the mind is acquainted primarily with abstractconcepts, and acquainted only derivatively withconcrete particulars. His theory of learning pro-posed that all so-called learning is actually remem-bering, that there is no real "learning" at all. Thisis the doctrine of anamnesis (recollections).

The doctrine of jorms is presupposed in the Menorather than formulated or explicated. The Platonicepistemology is essentialism (Popper, 1963): trueknowledge is a description of the ultimate natureor "essence" of things—the reality which lies be-hind appearances. Behind every appearance (thingin the material world) lies an ultimate reality (es-sence) which is "known" by the intellect in pos-session of its truth. Plato distinguished threeworlds or realms of being: the first world, theworld of material objects; the second world, theworld of psychological awareness; and the thirdworld, the world of forms, essences, or ultimaterealities. True knowledge and primary being residein the third world; the material things of the firstworld have derivative or secondary existentialstatus. If we, as creatures of the second world,happen to exercise our faculties correctly and hap-pen to be lucky, we will glimpse the essences in thethird world. For a Platonist, the function ofscientific theory (as true explanation) is to denythe status of reality to appearances, and to derivethem (lawfully) from the underlying level of forms,which alone constitutes reality.

Anamnesis dovetails nicely with the doctrine offorms. The doctrine of recollections states thatthere is no learning (of essences), only remem-brance. It hinges on the immortality of the soul.For if the soul, as a denizen of the third world, isimmortal, then it already knows everything there isto know. This is the text of the Meno:

16 AMERICAN PSYCHOLOGIST January 1973

The soul, then, as being immortal, and having been bornagain many times, and having seen all things that exist,whether in this world or in the world below, has knowledgeof them all; and it is no wonder that she should be ableto call to rememberance all that she ever knew about virtue,and about everything; for as all nature is akin, and thesoul has learned all things, there is no difficulty in hereliciting or as men say learning, out of a single recollectionall the rest, if a man is strenuous and does not faint; forall inquiry and all learning is but recollection [Sesonske &Fleming, 1965, p. 17].

Thus the illusion of learning results from the frailtyof memory.

For the theorist in possession of the doctrines ofanamnesis and forms, the "paradoxes" of the Menoare indeed spurious. The paradox of knowing sim-ply disappears. The forms and essentialism com-bine to guarantee that "knowledge" is of essencesij we are aware of (remember) anything. Theproblem of productivity disappears because crea-tive behavior could not be dependent on priorlearning, for there is no learning: we are "creative"or "productive" because we remember well. Thisparadox exists only if it is assumed that knowledgeis based on learning, a thesis that the doctrine ofrecollections explicitly undermines. Thus for Plato,there is a solution for these prima facie paradoxesthat renders their conclusions perfectly plausibleand desirable. If these doctrines are correct, thenthe "paradoxical" situations are only to be ex-pected. It remained for Aristotle and the nominal-ists to dissolve the paradoxes, because they rejectedboth of these Platonic doctrines.

Aristotle's Rebellion and the Beginningoj Modern Dogma in Philosophy andPsychologyTo most modern readers the doctrines of forms andanamnesis are absurd. Such talk of the immortalityof the soul, of a realm of essences, and of innateideas is a quaint relic of an earlier, less sophisticatedera that was easily satisfied with glaringly inade-quate and unscientific explanations. Such a readertraces his heritage to Aristotle, who also rejectedhis master's preachment on these issues. Aristotlecommonly is regarded as the founder of the doc-trines of nominalism (which holds that there areno nonconcrete, nonparticular entities), and as-sociationism, and he first used them to dissolvePlato's paradoxes.

To combat nativism, Aristotle needed a mecha-nism of the mind that rendered memory, or recol-lection, explainable without resorting to a trans-cendent soul. This he found in associationism,which he "borrowed" from Plato. Aristotle claimedthat the sequence of cognitive events is not ran-dom, but rather sequentially organized: habit andnecessity of connection are the organizational prin-ciples. In both habitual or spontaneous connectionand purposive or directed thought, the same threefundamental laws are operative: the laws of simi-larity, contrast, and contiguity. These principlesof association provided him with a mechanism ofthe mind which could explain recollection withoutrecourse to the Platonic doctrines. Consider H. C.Warren's (1921) rendition of Aristotle's view:

Recollection occurs inasmuch as one experience naturallysucceeds another. If [the succession be] necessary, it isplain that when the one is stimulated it will stimulate theother; if not necessary but habitual, then it will stimulateit only in most instances. It is a fact that some personsare more habituated after being stimulated once thanothers after many times; just so we remember some thingsbetter after seeing them once than other things after manytimes. Hence, when we are recollecting we keep stimulat-ing certain earlier experiences until we have stimulated onewhich the one in question is wont to succeed. And justso we hunt through the sequences, thinking along from thepresent or some other [thing], and from similar or con-trasted or contiguous. By this means the recollectioncomes; for the experiences are in some cases identical with[the one in question], in others simultaneous with it, inothers they involve a portion, so that the remainder issmall and is thereupon stimulated. . . . It is not necessaryto inquire how we remember the remote, but only thecontiguous; for it is plain that the procedure is the same—I mean, that of sequence—when we [recollect] withoutpreliminary effort and even when we fail to recollect. Forexperiences habitually follow one another, this succeedingthat. And so, when a person wishes to recollect, he willdo this: he will endeavor to get some initial experience,which the one in question succeeded [pp. 25-26].

Associationism, then, provides a mechanism of themind that explains recollection without recourse tothe doctrine of anamnesis.

Nominalism enters Aristotle's system to combatthe forms as soon as associationism is admitted.One has only to note that the senses (which areassumed to be the foundation of our knowledge)present us with particulars which are associated bythe mind, rather than with universals or other ab-stract entities. We never perceive redness, or man-kind, or the law of gravitation; rather we see in-dividual red objects, this or that man, and falling


objects. Why then say that the mind is funda-mentally aware of forms or any such abstractentities? Aristotle, with his "scientific" tempera-ment, could see no reason for the primacy offorms at all.

But Aristotle did recognize the conceptual utilityof generic concepts. How then is one to derivethem from particulars? His answer was a theoryof abstraction based on the principles of association,which we may call the subtractive theory oj ab-straction. This theory says that our generic con-cepts result from the association of common ele-ments (or features) of particular objects. Man-kind, for instance, means nothing more than thesum of properties found in common in particularmen. There is nothing more to redness than thecommon features of all of the individual red thingswe experience. For Aristotle and the nominalists,the abstract or universal is merely that which isshared in common by concrete particulars.

Examining the question of the nature and roleof abstract entities in the learning process showsthat from the Aristotelian point of view the "solu-tion" to Plato's paradox consists in its dissolution.Abstract entities play no essential role in ourknowledge, for all learning is the learning of par-ticulars and their recombination, and abstract con-cept formation is a derivative result of this processof recombination. Learning and knowledge havetheir basis in concrete particulars only: .abstractspecification occurs only as a result of the learningof particulars. The problem of productivity like-wise vanishes: it is an illusion resulting from over-looking the recombination of old particulars. Pro-ductive thinking is nothing but the association ofnew combinations of elements. Aristotle solvedPlato's paradoxes by dissolving them entirely.

The Centrality oj Aristotelian ThoughtTodayA moment's recollection (if you will pardon thatterm) shows that Aristotle's doctrines are at theheart of contemporary thought in epistemology andthe psychology of learning. The centrality of as-sociationism as the mechanism of the mind is sowell known as to require only the observationthat not one single learning theory propounded inthis century has jailed to base its account on as-sociative principles. Associationism enjoys incred-ible currency, even among otherwise mortal enemies

such as neobehaviorism and the Gestalt approach(seeNeisser, 1967, chap. 11).

And contemporary associationists mirror Aris-totle's thinking with respect to creativity and serialordering almost exactly. Learning theories arecopying theories: if the present situation heldnothing in common with past situations, learningwould be impossible, for there would be nothing tolearn from. Learning cannot occur unless someconcrete, particular elements remain invariantwithin the overall variable situation. If the samestimulus situation is not somehow instantiated inapparent novelty, learning cannot be exhibited, forthere would be nothing from which to transfer priortraining. Consider the following "explanation" ofnovelty:

If the correct response must always occur before it can berewarded, what novelty is added by learning? The newfeature is that the particular response rewarded now occursregularly to a specific cue, whereas previously its occurrenceat just that time and place may have been exceedingly in-frequent. The connection between cue and response is thenew product of learning. Often a number of different re-sponse units are connected to cues so that they all occurtogether, either simultaneously or successively. Thus anew pattern of responses is produced; the responses areold, but the combination is new [Dollard & Miller, 1950,p. 371.

Serial order is also a matter of the association ofelements of behavior. Hull (1930) provided theclassic stimulus-response approach with mediationtheory (generic r0), and his basic thought pervadesthose formulations addressed to complex phenom-ena, where the problem is most acute (see Berlyne,1965; Jenkins & Palermo, 1964; Osgood, 1953,1957, 1963).

These same theorists follow Aristotle in dissolv-ing the problem of productivity. Consider thesecomments for their denial of genuinely creativebehavior:

Productive thinking takes place when "past learning issubject to modification and reorganization," for example,when an individual who has never "reacted to the weightaspect" of a pair of pliers conceives of its possible use as apendulum bob "during the stress of the problem situation.". . . Productive thinking is typically a product of stimulus-response generalization, and it invariably permits of wideextension to future problem situations through furtherstimulus-response generalization [Berlyne, 1965, pp. 315,317].

This same theorist echoes the subtractive theoryof abstraction in a particularly clear fashion:


Abstraction, on the other hand, consists of making overt be-havior depend on certain properties of a stimulus patternwhile disregarding other properties. . . . Abstraction, then,means making behavior take account of some of the signalsin the message and disregarding the remainder [p. 45].

In philosophy, the dream of constructing the en-tire complexity of the world on a basis of concreteparticulars constantly recurs. A recent milestonein this tradition is Nelson Goodman's (1951) at-tempt in The Structure of Appearance to constructabstracta out of concreta. W. V. O. Quine is alsoof professed nominalist leanings. In general, onemust admit that nominalism, seemingly implied bythe empiricist world view, dominates the philosophi-cal scene. Generic concepts, classes, etc., are ad-mitted, albeit grudgingly, but the majority of lipservice is still to nominalism.2

But despite the plausibility and pervasiveness ofthe "received view" doctrines stemming from Aris-totle, it is easy to see that they are indefensibledogmas. The destructive burden of this article isto show that nominalism and associationism areutterly incapable of providing adequate accounts forour two problems. They are, indeed, considerablymore implausible than the seemingly unscientificdoctrines of Plato. Let us now examine somerecent "heresies" in philosophy and psychologythat expose the "received view" doctrines as theegregious blunders that they in fact are. Considerfirst the untenability of nominalism.

The Conceptual Primacy of the AbstractA number of developments in recent philosophyforce the abandonment of nominalism in its tradi-tional forms. They may be grouped convenientlyunder the heading of the epistemological, or more

2 There are prominent cracks in the nominalistic bulwark,however. For example, Quine (1960) now defends nominal-ism as an admitted bias, on pragmatic grounds only. Con-sider these comments in Word and Object:

The case that emerged for classes rested on systematicefficacy. Note it is certainly a case against nominalism'snegative claims, but still it is no case against a preferentialstatus for physical objects. In a contest for sheer sys-tematic utility to science, the notion of physical objectstill leads the field. On this score alone, therefore, onemight still put a premium on explanations that appeal tophysical objects and not to abstract ones, even if abstractobjects be grudgingly admitted too for their efficacy else-where in the theory [p. 238].

But the point remains—one cannot notice cracks in the wallof nominalism unless the solidity outweighs the proportionof crumbling doctrine.

generically, the conceptual, primacy of the abstract.We can dismiss nominalism by considering (atleast) these areas: the conceptual nature of factsor "factual relativity"; the abstract nature of scien-tific explanation or the "deductive unification ofexperience"; and finally, the primacy of the abstractin the nature of both common sense and scientificknowledge, and in scientific theory construction.

Consider first factual relativity. Philosophers ofthe "empiricist" persuasion that Aristotle cham-pioned hold that our knowledge is founded or basedon "facts." The given in experience for the nomi-nalistic empiricist is the particular, the concrete,and the atheoretical. The deliverances of sense arethe touchstone on which scientific knowledge restsand is validated, and those deliverances have nomi-nalistic properties "built in," according to the re-ceived view. But there is a problem with "facts":they cannot have either the nominalistic or thefoundational properties that this position demands.Facts cannot be the "raw data" of observation.

A little reflection shows that observation is some-thing over and above the raw data of perception—it requires an active process to construct facts outof the raw data. Observation requires skill: it isan achievement term in the sense of Austin or Ryle.As Turner (1967) rightly noted: "An observer whocould claim no other credentials than those ofnaivete would see very little that would be of in-terest to the scientist [p. 190]." But this hassimply not been noticed by empiricist philosophers.

The reason such philosophers have ignored theconceptual nature of facts is obvious: it destroysthe foundation (all knowledge is a deliverance ofsense) on which their world view was erected. Forthe empiricist, reality consists of the given, theconcrete, and the particular. Factual relativitysays, in effect, that the given cannot be "taken" atall, that what there is is not concrete as the em-piricist construes the term, and that the particularcannot be known at all without prior or simultane-ous knowledge of the general.

But theorists do not, in viewing the same "ob-jective" situation, perceive the "same" facts. AsT. S. Kuhn (1970) has indicated in discussingparadigm clashes, theorists on opposite sides in aclash literally live in different worlds. This gulfbetween observers of different theoretical persua-sions is well captured in a historical example:

If in the brilliant disc of which he is visually aware Tychosees only the sun, then he cannot but see that it is a body


which will behave in characteristically "Tychonic" ways.These serve as the foundation for Tycho's general geo-centric-geostatic theories about the sun. . . . Tycho sees thesun beginning its journey from horizon to horizon. He seesthat from some celestial vantage point the sun (carryingwith it the moon and planets) could be watched circling ourfixed earth. Watching the sun at dawn through Tychonicspectacles would be to see it in something like this way.Kepler's visual field, however, has a different conceptualorganization. Yet a drawing of what he sees at dawncould be a drawing of exactly what Tycho saw, and couldbe recognized as such by Tycho. But Kepler will see thehorizon dipping, or turning away, from our fixed local star[Hanson, 1958, p. 231.

Factual relativity guarantees that one cannotsimply go out into the world and neutrally collectfacts. Without a prior conceptual framework,that is, a point of view from which to impose orderupon reality, there is only the changing phenomenalflux of experience, the "blooming, buzzing con-fusion" of William James. The data of sensationdo not come with little tags attached proclaimingtheir factual status. Observation is not merelyfocusing one's attention on the data, but rather as-similation of data into the conceptual scheme of theobserver.

Factual relativity, emphasizing the theoreticalcontamination of the allegedly solid and neutraldata base of science, is the first step in the over-throw of nominalism. The nominalist holds thatwhat is given to the mind, the basis of all its fur-ther constructions, is atheoretical and neutral, orin a nutshell, concrete. Factual relativity removesthis alleged concreteness from the deliverances ofsense in a most convincing fashion: the "data base"for either science or everyday human endeavor istheoretical, shifting, dependent on nonobjective fac-tors, and thoroughly nonconcrete.

Thus, what is given to the mind is not concrete:is it also not particular? If it is not particular, thennominalism, as the thesis that there are no noncon-crete nonparticulars, is untenable. Let us now lookat the perceptual experiences that are said to con-stitute the ultimate data base of scientific explana-tions, and see that not only are there no "perceptualexperiences" whatsoever at the "basis" of science,but also that the propositions of explanatory dis-course deal with abstract and ideal entities.

Concordant with the view that science does not"collect" its facts but must instead "manufacture"them, this claim can be advanced: science cannever deal directly with particulars or individualthings at all, but only with "abstract" entities or

thing-kinds. Most contemporary "philosophy ofscience," insofar as it deals with substantive the-ories at all, is concerned with the purely formalstructure of the subject-matter theories examined(in particular, those that are or can be axiom-atized). But one factor that has been quite ne-glected is the restriction which the structure ofscientific axiomatic systems imposes on the sub-ject matter of such theoretical systems.

In endorsing factual relativity, we have deniedthat any form of the given in experience provides abedrock on which to found science; now we mustquestion the other facet of the empiricist "founda-tions" view of knowledge which holds that thatwhich is given is to be taken as particular. Posingthe issue in this form leads back to the problemsof knowledge with which we began: how does thenominalist-empiricist know (or come to recognize,be acquainted with, or recollect, etc.) his concreteparticulars as instances of particulars? That is, ifa "thing" is proffered to me as "an X," where X isany description (i.e., classification) of it whatso-ever, how can I know that to be the case (that itis an X) without first knowing what it is to be aninstance of a thing-kind, namely, of kind XI Inorder to recognize a thing (fact), mustn't one pre-suppose knowledge of, or operation within, a frame-work (theory) of thing-kinds? Reflection indi-cates that classification is fundamentally a processof abstraction. But this goes against the nominal-istic attempt to accord abstracta derivative statusfrom concreta, by making the process of abstractiona sine qua non for determination of concreta. Letus develop this point by considering what is in-volved in "scientific" explanation.

The ideal form of theoretical explanation is usu-ally taken to be the hypothetico-deductive schema.The question we must ask is: Does employment ofthe hypothetico-deductive method idealize the do-main to which it is applied? That is, what doesthe process of deduction require of the empiricalpredicates of the science before it may be appliedlegitimately?

Korner (1966) considers two classes of such con-straints. The first concerns elimination of inexact-ness and indefiniteness of predicates. The logic ofthe hypothetico-deductive framework is an un-modified classical two-valued logic. Thus, strictlyspeaking, the hypothetico-deductive framework ad-mits no inexact predicates whatsoever. So it must,strictly speaking, be the logic of the finished sci-


ence report, that is, admit no open concepts in thesense of Pap (19S8). The point is this: the hypo-thetico-deductive system is not connected to ex-perience directly: the empirical predicates withwhich it deals must be idealizations or abstractions.Raw experience is rendered into "concrete" cate-gories by the process of abstraction.

When applied to perceptual characteristics, the abstractionconsists in distinguishing between relevant and irrelevantdeterminables and, so to speak, removing the deducibility-relation which holds between any relevant and any ir-relevant determinable. It results in the formation of new,"abstract" determinables—every abstract determinable andits perceptual counterpart being mutually exclusive. . . .

This type of abstraction which, in order to distinguishit from other kinds, I shall call "deductive abstraction,"and which replaces perceptual by abstract determinables,reinforces the general effect of the restrictions which thelogico-mathematical framework of every theory imposesupon perceptual characteristics [Korner, 1966, pp. 166-167].

Even this is not sufficient to render an empiricalpredicate fit for a position in a deductive unifiedsystem. Further idealization removes the .predicatefrom perceptual experience entirely:

The disconnection of the theory from its perceptual subject-matter can now be also expressed by saying that no per-ceptual proposition and no perceptual predicate occurs inany deductive sequence. That this must be so is clear. Allinexact perceptual predicates are precluded from occurringin any sequence, and the exact—though internally inexact—determinables have been replaced by non-perceptual predi-cates through abstraction—to say nothing of the further re-placements due to the conditions of measurement, generaland special.

No perceptual proposition will be a last term in a deduc-tive sequence. The theory will be linked to perception notby deduction but by identification [Korner, 1966, pp. 168-169].

This means that the reference of "empirical" con-cepts cannot be given, but is rather idealized, orconstructed, by the active abstraction of our con-ceptual frameworks. An ideal straight line withinan axiomatic system is a far cry from any empiricalline, such as a light ray, the surface of a straightedge, etc. But our theories of reality deal with theabstract, idealized elements such as "straight lines"and not with the lines drawn with rulers or physi-cal objects. With that admission, our theories be-come nonempirical: their reference is to ideal en-tities rather than to concrete physical ones.

We need note only this in conclusion: the nomi-nalistic dream of constructing "abstracta" out of"concreta," thus relegating the former to second-

class status, must admit that particular concretacan be identified only with the help of abstracta.But acknowledging this fact abandons the dreamentirely. Particular "facts" can only be knownas instances of abstract, theory-determined, thing-kinds. What we are "presented" with by the con-struction of experience is neither concrete nor par-ticular. Particulars depend for their very existenceon the primacy of abstract categories. Those cate-gories exemplify the force of Plato's paradox ofknowledge: We cannot "know" anything (as an in-stance of a thing-kind classification) without al-ready "knowing" the abstract concept. Nominalismsimply does not account for the nature of ourknowledge.

The third point to be made under the heading ofthe conceptual primacy of the abstract is that ourknowledge is inherently based on abstract specifica-tion, because not only science but also commonsense can deal only with the abstract. If eitherscience or common sense reasoning is in the businessof achieving (by whatever means) our knowledge,then, because of "the way the minds works," allour knowledge is abstract.

Let us consider the ramifications of this conten-tion by exploring two theses which contain basic in-sights into the nature of human knowledge and theorganism which has that knowledge: the thesis ofstructural realism (as a theory about the nature ofour knowledge), and the more psychological thesisof the abstract specification of experience due toF. A. Hayek (1952, 1969). These positions pro-vide slightly different arguments against nominal-ism, and serve both to "overkill" nominalism andto drive home the conceptual primacy of the ab-stract in yet another manner. Consider Hayek'sposition first.

In a fascinating and neglected book entitled TheSensory Order, Hayek (1952) is concerned withperception, and the physiological correlates of ourpsychological abilities. Hayek's thesis is that nosensory input is "perceived" (i.e., inputed throughthe active central nervous system) at all unless itis perceived as one of the kinds of input acceptedby the (innate or learned) classes of sensory order.Sensory perception is always an act of classifica-tion; the input signal is "processed" by any mem-ber (to which it "keys") of the sensory "orders"which impart to the phenomenal event the proper-ties it has. No sensory input is perceived unless itcan be "isomorphically" accepted as a match by


the classes of sensory order. No constructions ofphenomenal existents are possible except in terms ofthe (prior) apparatus of classification inherent inthe operation of the functional nervous system. Un-less an "event" gives rise to the pattern of sensoryinput that fits an organism's preexisting naturalkind classifications, it is not perceived at all. Per-ception is thus never of the attributes of "objects"in the world at all: instead, objects are abstractionsof the organization and memory of the centralnervous system.

The sensory (or other mental) qualities are not in somemanner originally attached to, or an original attribute of,the individual physiological impulses, but . . . the wholeof these qualities is determined by the system of connexionsby which the impulses can be transmitted from neuron toneuron; . . . it is thus the position of the individual im-pulse or group of impulses in the whole system of suchconnexions which gives it its distinctive quality; . . . thissystem of connexions is acquired in the course of the de-velopment of the species and the individual by a kind of"experience" or "learning"; and . . . it reproduces thereforeat every stage of its development certain relationships exist-ing in the physical environment between the stimuli evokingthe impulses. . . .

This central contention may also be expressed morebriefly by saying that "we do not first have sensationswhich are then preserved by memory, but it is as a resultof physiological memory that the physiological impulsesare converted into sensations. The connexions between thephysiological elements are thus the primary phenomenonwhich creates the mental phenomena" [Hayek, 1952, p. S3].

Again, Hayek's thesis is that an "event" is notseen at all unless it is assimilated to a classificationthat already exists in the functioning of the centralnervous system. Everything that we know aboutstimuli, our entire knowledge of the "external"world, consists in the classifications effected by theaction patterns of the functioning nervous system.Human knowledge is a system of rules of determina-tion that indicate equivalences and differences ofvarious combinations of input patterns.

The conceptual primacy of the abstract is that:

We ought to regard what we call mind as a system ofabstract rules of action (each "rule" defining a class ofactions) which determines each action by a combination ofseveral such rules; while every appearance of a new rule(or abstraction) constitutes a change in that system, some-thing which its own operations cannot produce but whichis brought about by extraneous factors.

This implies that the richness of the sensory world inwhich we live, and which defies exhaustive analysis byour mind, is not the starting point from which the mindderives abstractions, but the product of a great range ofabstractions which the minds must possess in order to be

capable of experiencing that richness of the particular[Hayek, 1969, p. 3181.

Hayek's psychological thesis literally is the doc-trine of anamnesis in modern, evolutionary dress.Now, are there philosophical considerations thathave equal affinity to the Platonic doctrines? Thethesis of structural realism fills this bill of "par-ticulars" quite well.

To set the stage for structural realism, considerthe distinction, made famous by Bertrand Russell,between knowledge by acquaintance and knowledgeby description. You are directly acquainted with,for example, the color of your shirt, the warmth ofyour coffee, the pain of your headache, etc. Any-thing that is in the "phenomenal experience" of anindividual is known to him by direct acquaintance:roughly, anything that is within subjective experi-ence, that is, "directly observed," is known byacquaintance rather than by description. But sup-pose you are to meet someone with whom you arenot acquainted, whom you have not met per-sonally. You can still know someone (or some-thing) by description without being acquainteddirectly with him (or it). Suppose you are to meeta new brother-in-law at the airport, and have onlyyour sister's description to go by: he is "about sixfeet tall, has sandy hair, weighs 175 pounds, hasblue eyes, wears blue suits, etc." Once you havepersonally met someone previously known only bydescription, then (given sufficient time) you maybecome directly acquainted with the reference ofmost of your previous knowledge by description.

Knowledge by description is the basis of ourlearning new things, as yet unexperienced. Welearn to assimilate descriptions to known ac-quaintances (e.g., Small boy: "Mommy, what isgray hair?" Mother: "It's like your grandfather's,dear." The boy, knowing his grandfather by ac-quaintance, now knows what "gray hair" refers to).Knowledge by description is inferential and preposi-tional: it is knowledge that, whereas knowledge byacquaintance is knowledge of. The crucial point isthat all scientific knowledge is knowledge by de-scription. Structural realism is a thesis about whatconstitutes that knowledge by description (not ourknowledge of sensory impressions and acquaint-ances).

The thesis of structural realism can be statedquite succinctly. First distinguish between themental and the nonmental world (including ourbodies in the latter), and note that "everything


that we usually consider to be perceived propertiesof external objects must be recognized as existingwholly and only in the mind [Maxwell, 1968a, p.151]." The properties that we ordinarily ascribeto external objects "are just as much in the mind asthose we experience in dreams or hallucinations.. . . In a real and important sense, then, all of theexternal world, including even our bodies is unob-served and unobservable [p. 152]." This distinc-tion between the mental and the nonmental, locat-ing the properties and contents of the latter in themind, is like Berkeley's commentary on Locke'sdistinction of primary and secondary qualities.Berkeley pointed out that the distinction was in-valid: everything is secondary in the sense that allproperties and qualities are in the mind rather than"in" the objects themselves.

But if this is so, of exactly what can our knowl-edge of the nonmental realm consist? The enormityof the issue can be thrown into bold relief by re-flecting on the fact that all of our "natural" and"physical" sciences disclose knowledge only of thenonmental sort. Can science conceive meaningfullyof any properties of the entities of the nonmentalworld? Structural realism answers thusly:

The only aspects of the nonmental world of which we canhave any knowledge or any conception are purely struc-tural (or, in other words, purely formal). . . . The notionof form or structure needed here may accurately be said tobe logical (and/or mathematical) and, in a sense, abstract;characterizations of instances of it will be in terms of logicalone, i.e., the logical connectives, quantifiers, and variables—they will contain no descriptive terms. . . . Structure, inthe sense we require, must be factual and contingent and,at least in its exemplifications, concrete. Furthermore, itcannot be emphasized too strongly—what should alreadybe obvious—that structure is not linguistic nor, even, con-ceptual in character; it is an objective feature of the realworld [Maxwell, 1968a, p. 153].

Structural realism is structural in the sense that ourknowledge of the entire nonmental world, from ourbodies to theoretical entities, is of the structuralcharacteristics of that world rather than of theintrinsic properties of those objects comprising it.

Structural realism may be stated in terms relatedto Kant's distinction of phenomena from noumena.Kantian phenomena, which exist wholly in themind, are the only occurrences of which we havedirect knowledge: we know phenomena by ac-quaintance. But the noumena, the things in them-selves, are completely removed from our acquaint-ance: we have no direct knowledge of them at all.

Yet the bulk of our knowledge is about the nou-mena: but that knowledge is of their structuralcharacteristics. Thus, against Kant, both com-mon sense and science disclose knowledge aboutthings in themselves, but that knowledge is purelystructural. The only things of which we know theintrinsic properties are the phenomenal qualia ofthe mental world. This requires some rather drasticchanges in our usual way of thinking about theworld and our knowledge of it.

One of the concepts most in need of revision(actually reinterpretation) is that of observation.We speak commonly of "observing physical ob-jects." But, strictly speaking, there can be noobservation of public objects at all:

Observation, as usually conceived, is a naive realist con-cept through and through. Therefore, if structural realismis true, then, in any usual sense of "observation," weobserve neither public objects nor entities in our minds:we never observe anything at all. For example, if I havea dream, no matter how vivid, of seeing a white dog, Icannot be said to have observed a dog or, indeed, anything,since there existed nothing corresponding in any straight-forward way to the ostensibly observed object. On theother hand, if I do what ordinarily could be called "ac-tually seeing a white dog," then . . . I do not actually ob-serve anything (or even see in the usual sense), for there isnothing external to me which is white in the usual, qualita-tive sense, or etc., etc. [Maxwell, 1968b, p. 167].

Talk of the observation of objects, as is common inscience and common sense, must be taken as ashorthand formulation (in terms of naive represen-tational realism) for discourse about sense impres-sions and their structural relation to the externalobjects which are their causes.

To conclude our account of scientific knowledge,consider the role of the abstract in structural real-ism. We must note the sense in which abstractentities are the essential components of scientificknowledge.

Recall that the opposition to abstract entities inscience is that they are alleged to be too far re-moved from the deliverances of sense (or whateveris taken to be the foundation or data base of knowl-edge) to be "real." What is real to the nominalistis what is particular, concrete, visually and tactuallypresent to the investigator, etc. Against nominalismwe argued that the particular cannot be known asan instance of a particular of kind X without pre-supposing knowledge of universal, or otherwise ab-stract, entities (such as thing-kinds). But, if scien-tific knowledge is formal or structural, then it is


always abstract because the formal relationshipswhich characterize the nonmental world are notparticular, concrete, always present to the senses,etc. Structural knowledge is fundamentally ab-stract knowledge in the sense that it is not what ispresented in our experience. Theoretical entitiesare specified by reference to their (almost invari-ably mathematical) structure. "Scientific" knowl-edge of nonmental entities is limited entirely tothis structure, and we cannot attribute to theoreti-cal objects the properties that the nominalist baseshis claim upon: that is, the particular and concretedeliverances of our senses. William Kneale (1949)caught the nominalist's fallacy:

That transcendent (theoretical) hypotheses are concernedonly with structure has often been overlooked in the past,because scientists and philosophers have mistakenly al-lowed themselves to slip some imaginative elements, such asperceptible hardness, into their concepts of the objects men-tioned in the hypotheses [p. 941.

But the nominalist can be left in a quandary: hisdilemmas only lead us away from the main point.Scientific knowledge of the nonmental world is amatter of having defensible opinions about the ab-stract structure of that world. Abstract entitiesare essential to scientific knowledge for this reason.

Here it is well to pause, reflect upon the majorpoints of this section, and ask where we are withregard to the problems for which Plato postulatedforms and anamnesis. I hope to have succeeded inshowing that Aristotle and his nominalistic de-scendants have continued to ignore the paradoxesthat Plato raised. To be a nominalist today is toadmit that one cannot address any of the interest-ing questions that we have discussed.3 But de-

3 Associationism is said to unite particular elements. Butis associationism in fact nominalistic? A straightforwardargument attributed to Harold Hoffding (1891) shows thatit cannot be; that, in Bradley's (1883) cryptic formulation,"Associationism marries only universals." The Hoffdingstep and the Hoffding function destroy the nominalistic in-terpretation of associationism.

Consider what associationism is said to do. Two ele-ments, A and B (their stimulus or response characteristics,neural locus, etc., are irrelevant), because of one or moreof the "laws" of association, have been linked together,such that when one occurs, the other tends to occur also.The laws of association tell you the sufficient conditionsfor this "linkage": contiguity, similarity, frequency (andthe whole host of others that have been proposed). Butthe elements cannot be directly associated: they never recurin nature in the same manner. Due to the Heraclitean fluxthe same elements never reappear. Thus the elements Aand B may not become associated directly unless you be-lieve in one-trial learning, and unless you recognize that

molishing nominalism, no matter what its signifi-cance, is unimportant when one reflects on thequestion that naturally follows it: Can anyone sayanything significant about "the way in which themind works?" Unfortunately, the answer, we shallsee, is more no than yes.

Psycholinguistics and the Reemergenceof the Problem of CreativityPsycholinguistics is inherently Platonic in outlook.Chomsky (1966a), in tracing his heritage in lin-guistics to Descartes, overlooked his psychologicaland epistemological descent from Plato almost en-tirely. For the transformational grammarian, crea-tivity in language is the prime datum for linguistictheory. No theory that fails to provide explanatorymachinery sufficiently complex and powerful enoughto explain the novel but appropriate use of languagecan be considered adequate. Consider Chomsky's(1966b) succinct presentation of the centrality ofproductivity:

The most striking aspect of linguistic competence is whatwe may call the "creativity of language," that is, thespeaker's ability to produce new sentences, sentences thatare immediately understood by other speakers althoughthey bear no physical resemblance to sentences which are

only the "atoms" or parts of those elements (call them aand 6) can thus be associated "directly." The contempo-rary formulation of this elementism, represented in Estes'(19SO) assumption of stimulus elements, pervades themathematical psychology literature. Associationism, inorder to survive an obvious refutation of its claim to bethe mechanism of the mind, was forced to postulate (re-member, no one has ever shown us a stimulus element) anatomistic picture. But with this tacit admission the as-sociation of A and B is not of this form:

-B

But rather of this form:

There is no linkage at all between A and B, except as ashorthand notation for the linkage through a and 6. Thelink between A and a has been called the Hoffding step, andthe entire sequence from A through a-b to B is the Hoff-ding function. The "proper part" or thing-kind nature ofthe elements a and b is what is devastating to nominalism.Since thing-kind identification involves one inescapably inabstract and "universal" specification, Bradley's thesisstands: associationism marries only universals. Yet, if thedoctrine must presuppose universals in order to explainthem, then its claim to explain universals in terms of par-ticulars is false.


"familiar," . . . Modern linguistics, however, is seriously atfault in its failure to come to grips with this central prob-lem [p. 111.

With this admission, Plato's paradox returns: forhow can one exhibit knowledge for which one's priorlearning history has given no preparation? Chom-sky takes exactly the same approach to produc-tivity as Plato did in explicating Meno's slave'srecollection of things "already known." Chomsky(196S) argues that

In bringing to consciousness the triple ambiguity of (S)[I had a book stolen] in this way, we present no new in-formation to the hearer and teach him nothing new abouthis language but simply arrange matters in such a way thathis linguistic intuition, previously obscured, becomes evi-dent to him [p. 22].

Plato's problem, although ignored by many of hissuccessors, has not disappeared after all. And thepoint is that associationism cannot provide a suf-ficiently powerful generative mechanism to accountfor creativity.

There is no sense of "habit" known to psychology in whichthis characterization of language use is true (just as thereis no notion of "generalization" known to psychology orphilosophy that entitles us to characterize the new sen-tences of ordinary linguistic usage as generalizations ofprevious performance). The familiarity of the referenceto normal language use as a matter of "habit" or as basedon "generalization" in some fundamental way must notblind one to the realization that these characterizations aresimply untrue if terms are used in any technical or well-defined sense, and that they can be accepted only as meta-phors—highly misleading metaphors, since they tend tolull the linguist into the entirely erroneous belief that theproblem of accounting for the creative aspect of normallanguage use is not after all a very serious one [Chomsky,1965, p. 12].

"Habit" theories cannot handle creativity becausethey would need an infinitude of associations in theorganism's central nervous system to be "activated"and displayed (probabilistically, dispositionally) innovel behavior. Even if there were a way to getall of those habits into the organism (a physicalimpossibility), they would not be "habits" at all:they could not have been learned. Associationisticlearning theories are identical element-copying the-ories, and the distinguishing feature of novelty isprecisely that it is novel: it is not a copy of any-thing at all. At no point in the organism's priorhistory has there been anything to learn from whennovelty is exhibited. Generalization in stimulus-response theory presupposes identical elements(from which generalization occurs): there simply is

no stimulus-response theory of generalization forabstract entities.4 Thus Chomsky claims that aspeaker-hearer's knowledge of his language is in-finitely greater than his prior learning history.

To account for productivity, the transformation-alist has been forced to acknowledge the explana-tory primacy of abstract entities. The only way toexplain "creative" utterances is to employ rules inthe grammar that range over inherently abstractentities that never appear in actual speech or writ-ten language. The explanatory primacy of the ab-

4 Let us make this point in one further manner. Psycho-logical theory tells us that concepts of particular and genericnature arise by the gradual predominance of the similaritiesof entities over their differences. The theory is to providea mechanism whereby similarity alone imprints itself onthe mind: abstraction, after all, merely subtracts from thetotal presentation the relevant attributes—which are definedin terms of their similarity. The similarity in diversity ismade manifest to the mind as such. Elements cannot besimilar (let alone identical) unless they are recognized assimilar.

But there are two problems immediately facing the sub-tractive theory of abstraction or "generalization." First, itmust postulate that perceptions can be ordered into seriesof similars. Second, it must account for the fact that simi-larity or dissimilarity does not appear as an element of per-ception along with the perceptible qualities of the particularentities. Each problem leads to the same conclusion: thetheory presupposes what it is to explain.

The construction of series requires the same conceptualapparatus as the structuring of sentences, that is, generationof potentially infinite sequences according to rules. In onesense, the weakness of the subtractive theory of abstractionis that it selects, from all possible logical orders and rela-tions, only the principle of similarity. It ignores the in-definite number of other syntactic structuring relations.According to what principles was similarity chosen?Clearly it was not chosen; it was presupposed. Considera series of presentations of a a, a /3, a y, etc. The connec-tion of the members of such a series by the possession of acommon property ("a") is only a special example of thepossibility of logical connection. In all series, the connectionof the members is due to a law of arrangement establishinga rule of succession: that which binds the elements togetherconceptually is not itself a new element, but is itself therule of progression. This led Cassirer (1923) to conclude:

The unity of the conceptual content can thus be "ab-stracted" out of the particular elements of its extensiononly in the sense that it is in connection with them thatwe become conscious of the specific rule, according towhich they are related; but not in the sense that we con-struct this rule out of them through either bare summa-tion or neglect of parts. What lends the theory of ab-straction support is merely the circumstance that it doesnot presuppose the contents, out of which the concept isto develop, as disconnected particularities, but that ittacitly thinks them in the form of an ordered manifoldfrom the first. The concept, however, is not deducedthereby, but presupposed; for when we ascribe to amanifold an order and connection of elements, we havealready presupposed the concept, if not in its completeform, yet in its fundamental function [p. 17].


stract is at the heart of two crucial distinctions inthe transformational account: deep and surfacestructure, and competence and performance. Ex-ploring these distinctions for a moment will showthe centrality of abstract specification and produc-tivity in psycholinguistics, and thus the affinity ofthe approach to the Platonic epistemology.

The competence-performance distinction leadsdirectly to "creativity." Our knowledge of our lan-guage is vastly greater than our learning history.Linguistic experience is limited since we have beenexposed to only a relatively small sample of ourpotential linguistic utterances. Yet we can under-stand any sentence that is grammatically wellformed in our language. Competence has to dowith our knowledge of the structure of language.Linguistic performance, on the other hand, has todo with the actual speaking of language (see Chom-sky, 1967, p. 398).

Performance is what the speaker actually doesin speaking a language, but language per se is notperformance. The language, the goal of the lin-guist's investigation, is an abstract system, that is,a competence system (of rules or principles) thata speaker has that enables him to speak and under-stand. Thus language, as the subject matter in-vestigated by linguists, is an abstract system quiteremoved from the deliverances of sense (speechsounds per se) or whatever else is taken to be the"basis" of knowledge. Need I now remind you thatthe knowledge of language disclosed by contem-porary linguistics is wholly of its structural prop-erties, as Maxwell used the term in describing struc-tural realism?

Linguistic creativity automatically rules out allextant associationistic theories of language learn-ing, for they cannot, because of the limitations ofthe principles they employ, account for the infinitenumber of sentences that every speaker-hearer ispotentially capable of understanding (with no pre-vious "learning" history). All learning-theory for-mulations of language are by definition performancetheories: they cannot talk to the problems of com-petence at all.

Ambiguity, as a specific example of the problemsof meaning, provides a convenient entrance intothe surface-deep structure distinction. A grammar,according to Chomsky, relates sounds to meanings.Sounds exist as part of the "objective" or non-mental world. But what about meanings? Theyare not part of the "physical" world in the same

way as are sounds: meanings must somehow beabstractions from the "primary" data of the senses.Chomsky developed the surface-deep structure dis-tinction to explicate the abstract nature of meaningand its mode of representation in particular utter-ances. Let us first develop the notions "abstractly,"and then relate them to ambiguity.

In order to explain certain phenomena in lan-guage, Chomsky found it necessary to draw a dis-tinction paralleling the distinction between appear-ance and reality. Consider the case of scientifictheory construction. The commonsense man livesin a world of perceptible objects, persistent things,sensible qualia, etc., that is, in what Sellars (1963)called the manifest image of enlightened commonsense. In attempting to construct a theory of real-ity, science has come up with quite a different pic-ture of the world, populating it with microphysicalentities, events in space-time, etc. Thus, scien-tific theory functions to deny the status of realityto appearances. It reconstructs appearances asmanifestations of the underlying reality disclosedby its postulations. The surface structure is theworld of appearances. Deep structure is the under-lying reality postulated by science to explain theworld of appearance. Surface-structure appearanceis a manifestation of underlying deep structurereality.

The surface structure of a language is what isheard or spoken in the actual occurrence of speech(or written, in the case of writing). It is all theinformation contained in an utterance as it is heardor as it is written. A labeled bracketing or parsingof constituents exhausts the information containedin surface structure. Performance exhibits, but isnot synonymous with, surface structure. Deepstructure is the abstract, underlying causal orderfrom which the surface is generated (within thesystem as it stood in Chomsky, 196S, the basecomponent). Deep structure is essentially abstractstructure since it is nowhere manifested in surfacestructure. And yet, the abstract deep structure de-termines the meaning of the surface-structurestrings.

This problem of the relation between the mean-ing of an utterance and the deep structure is madeabundantly clear in those cases where one andonly one surface structure can be understood indifferent ways, that is, in cases of deep-structureambiguity. In these cases, no amount of interpret-ing the surface-structure string will help in ex-


plaining the multiple meanings of the utterance.Consider the following sentence which has onlyone surface-structure representation (parsing orbracketing):

Flying planes can be dangerous. [1]

Sentence 1 has two possible interpretations, neitherof which can be extracted unambiguously fromjust its surface structure:

(a) Being a passenger or pilot can be dangerous(the plane might crash).

(b) A plane might crash into you (while you areon the ground, etc.).

The point is obvious: surface-structure analysis can-not account for the understanding of an indefinitelylarge class of sentences. Deep structure is a neces-sary postulation in order to do so. The generalstrategy for requiring the notion of deep structurein any explanatory linguistic theory is found inChomsky (1966b, pp. 37ff) . Should doubts persist,this source should convince one of the necessity ofthe postulation of deep structure in explanatorymodels in linguistics.

To summarize the import of this digression, thefunction of the surface-deep-structure distinction intransformational linguistics is to separate the worldof appearance (surface structure) from the reality(deep structure) underlying it. Deep structure, in-sofar as it is not manifested (even indirectly) insurface structure, is abstract structure. The ex-plantory entities of linguistic theory are as abstractas any forms that Plato could have countenancedor desired.

And the surface-deep-structure distinction hasapplication far beyond the linguistic realm forwhich it was introduced originally. The entirepsychological domain, all behavior rather than justlanguage, is susceptible to analysis in terms of thedistinction. Chomsky's revolution in linguistics has,among other things, shown very clearly that noexplanatory theory that does not make the distinc-tion can provide an adequate psychology (sincelinguistics is only one branch of cognitive psy-chology). The problem of productivity in all be-havior forces psychology to acknowledge the ex-planatory primacy of the abstract, and effectivelyreinstates Plato's approach to the domain's prob-lems.5 But the recognition of a problem is only a

BIn an obvious sense, the transformational grammarianhas made a significant improvement over Plato's handling

first step. And what must be emphasized is thatin recognizing the gravity of the problems of pro-ductivity and meaning we have reinstated Plato'sepistemological problems in the Meno. We mustnow reevaluate the doctrines he proposed as theirsolutions. We have already reinstated (at leastimplicitly) the doctrine of forms in accounting forthe nature of our scientific knowledge, and now wemust reconsider the case for anamnesis.

Recollections Revisited: The BiologicalBasis of CognitionTo the empiricist and the behaviorist, the doctrineof recollections is an anathema. But what is itsstatus with the contemporary nativist, and thetheorist who rejects associationism as the necessaryand sufficient mechanism of the mind? Is thereany hope of resuscitating the doctrine in anythinglike its original form? That the answer is "Yes, itcan be reinstated" becomes obvious when we reflecton what the doctrine requires, and what it wasdesigned to accomplish. It requires somethingakin to a soul, that can be present in an individual,and yet contain a priori knowledge that transcendsany given individual's lifetime and experience.There must be a mechanism that presents reincar-nation and earlier lives influencing present compe-tence (as one essential characteristic of "souls")as other than an absurdity, granted our presentscientific perspective. To reinstate the doctrine,we need a mechanism of a priori (in the sense ofinnate) competence (knowledge in the sense ofcapacity, rather than specific contents) as the

of creativity. As Sesonske and Fleming's quotation fromPlato above indicates (see p. 17), he really had no accountof creativity or productivity other than the postulation ofan immortal soul that had already experienced everythingthat could be manifested as "creative." Plato's soul is aneternally enduring entity which can know (in the sense ofbe acquainted with) all possible instances of a concept.This is in a sense a denial of the ability or "competence"to be creative: it claims that creativity is an illusion ofremembrance. Chomsky does not deny creativity, andhis significant advance over both Plato and the associa-tionists is the theory that a syntax which allows recursionto occur is a mechanism of creativity: that is, Chomsky'sgenerative grammar is a mechanism that uses finite re-sources infinitely. He has provided an explanatory mecha-nism that makes infinite use of finite means: Chomskyexplains creativity rather than explaining it away, asPlato does. The negative task of this article is to arguethat we cannot improve upon the Platonic approach to theproblems of the higher mental processes, but it is not todeny that we have made some real improvements overPlato's specific proposals.


FIG. 1. Triangles as instances of triangularity.

product of "earlier lives." All the modern Platonistneeds to retain anamnesis is a mechanism account-ing for the "ideas," as capacities, of prior livesbeing built into a given individual—a mechanismprovided by the theory of evolution. The slavewhose creative competence Socrates helped todemonstrate had not lived before, but his nervoussystem and its capacities were determined by amechanism utilizing (albeit very indirectly) the ex-periences of racial ancestors who did live prior tothe boy. The competence underlying productivitycan be innate—and evolution can tell us how. Asan afterthought, note that the biology of Darwinthat revolutionized the discipline was a rebellionagainst the biology of Aristotle.

But surely this is not right—surely the recon-ceptualization in biology to which we must pledgeallegiance is Lamarck's rather than Darwin's. Andis it not well known that the Lamarckian doctrineof the inheritance of acquired characteristics isrepudiated thoroughly today? To this questionthe answer is clearly yes and no. Lamarckianbiology is indeed in disrepute: there is no evidencefor the inheritance of acquired physical character-istics. But the inheritance oj behavioral char-acteristics is well documented, as the discipline ofbehavior genetics discloses. And the inheritance ofcapacity to respond, that is, competence as a struc-tural concept underlying the functional notion ofdisposition to respond, is all the contemporaryPlatonist needs to maintain his thesis. He doesnot have to talk about the inheritance of learnedbehaviors at all—only of structural capacity, whichdetermines or constrains behavioral capacity.

But, if specific learned behaviors are not innate(or, in traditional terminology, there are no spe-cific "innate ideas"), what is built into the pro-ductive or creative organism? Here we have onlythe wildest speculations supported by the leastpossible evidence.

The Role oj Abstract Entities inPerception and Conceptual Learning

In examining the nominalistic tenor of psychology,we saw that it can account for neither the abstractnature of concepts nor the similarity with respectto which they are ordered. Now we must see howabstract entities are essential to perception and con-ceptual learning. We can do this by noting thatconcept formation is fundamentally a "productive"or "creative" behavior, and thus requires the ruleswhich would generate it to range over abstractentities.

Suppose an organism is to be taught the conceptof "triangularity." It is impossible for any list ofphysical attributes to exhaust the meaning (refer-ence) of "triangularity": there is an infinitude ofdistinct triangles. Thus, no identical-elementscopying theory can teach an organism the concept.The number of associations required to learn thatthe figures in Figure 1 (and indefinitely manyothers) are all triangles would be infinitely large.An organism which has learned that these figuresare instances of "triangularity" has learned therules characterizing the concept. This is just adescription, not an explanation. An explanationmust specify the rules, their order of application(if any), whether they are learned or innate, andmany similar points. All that we know at presentis that the rules range over abstract entities (i.e.,structures not found in the surface structure ofparticular triangles).

In learning that such instances are triangles, theorganism is constructing a theory of triangularity.As such, its task is identical to that of the scientistconstructing (we want to say "more sophisticated,"but that prejudges the issue) theories of reality.Could it be that the same procedure is used in bothbehaviors? Probably the answer is yes. Theorganism's task is to determine which proper subsetof, say, visual presentations to which it is exposedconstitutes instances of triangularity, and to ex-trapolate from that given "corpus" of triangles thedefining characteristics of the concept. Put differ-ently, the organism attempts to construct a theoryof the regularities of its corpus of visual presenta-tions. In these and all other cases of conceptformation, the organism may fruitfully be con-sidered as a theory of his environment.

But the visual presentations constituting exem-plars of triangularity are an infinitely large class.


Indeed, all generic concepts extend over potentiallyinfinite ranges of instantiations. How can anorganism recognize all such instances, with no priorexposure to them, as instances of the same con-cept? How can the organism answer the "similarwith respect to what concept" question when pre-sented with a novel instance?

The only way known for organisms to "makeinfinite use of finite means"-—such as they do inrecognizing an indefinitely large class of objectsas members of the same concept—is to employ"grammars" (in the linguist's sense) of perceptionor behavior whose rules make use of abstractentities. Only grammars that allow indefinite re-cursion and employ nonterminal symbols in theirderivations of surface structures from deep struc-tures can do this. What such grammars show ishow an abstract "deep structure" meaning can bemapped into indefinitely many distinct surfacestructures. A generic concept is the same as anunderlying deep structure that can be characterizedonly by the rules of the grammar, rather than bylisting the infinite number of "attributes" that arepresent in its (surface structure) instantiations.Concept formation is thus a generative process thatmay be understood through the model of language.To realize this is to realize that abstract entitiesare necessary for the scientific explanation of all"productive" or "creative" behavior. Concepts doexpress the "essence" of things in very much theway that Plato's doctrine of forms indicated, ratherthan the common features of "attributes" thatAristotle's nominalism proposed.8

6 The problems of meaning, in particular those of refer-ence, must be conceived as involving abstract entities. Thereference of classes that are not denumerated specifically isnot in principle determinable by a list of particulars. Thisis because the list would have to be infinitely long: refer-ence is determined by rules generating the denotata, notindefinite lists of associations. In this respect, conceptformation is exactly analogous to creativity or productivityin language: no finite state grammar can explain eitherprocess. The reference of a concept cannot be equated withany notion of association, be it particularistic or generalizedand abstract. This is clear when stated in another way. Ifthe number of properties a concept has is its content (orintension), then it is clear that content increases as wedescend from higher, abstract concepts to lower, particular-istic ones. Also, increasing extension (range) of conceptscorresponds to the progressive diminution of content (inten-sion). The most general concepts are provably not associa-tionistic or elementaristic simply by noting that they haveno intension at all.

The conceptual pyramid, which we form in this way,reaches its summit in the abstract representation of

How Does an Organism Become Familiarwith Abstract Entities?

Evolution is corroborated as well as any theorycan be: perhaps our present picture of the detailsby which it occurs (especially at the genetic level)is incorrect, but quite likely Darwin successfullypainted "the big picture" (see Waddington, 1969).In order to survive, an organism must come toknow its environment. This knowledge, at thelower phylogenetic levels, cannot be conscious.Yet, speaking either philosophically or common-sensically, the problem of knowledge is bound upinextricably with the notion of consciousness. So,,in these senses, to speak of Nature (capital N) asconstructing Her organisms to contain a theory oftheir environment (knowledge) is at best a colorfulmetaphor. But we must talk this way, and there-fore it is time to break with both philosophy andcommon sense. If we drop Nature in favor of atheory of evolutionary selection, then the problemof unpacking the "metaphor" that the organism isa theory of its environment becomes much simpler:all we must do is explain how an organism couldbe a theory of anything at all.

If an organism is to survive in this world, itmust be able to operate as a biological mechanismwithin the world. The mere fact of survival im-plies that an organism has been effective in main-taining the appropriate commerce with its environ-ment. The nervous system of the organism isresponsible for this commerce. How does the ner-vous system function? How did it come to behaveas it does? The answer, literally rather thanmetaphorically, is that the nervous system func-tions as a theory of the environment of the or-ganism; it makes inferences about environmentalcontingencies. To the extent that the theory isadequate and the inferences valid, the individual,and thus the species, survives. If the theory isinadequate, both may be expected to die out.Evolution is a mechanism that allows nervous sys-tems to construct more and more adequate theoriesof the environment. Man, the thinking, rationalanimal, is an emergent phenomenon in this respect:only his central nervous system has developed tothe extent that it can create consciously, as well as

"something" under the all-inclusive being of which everypossible intellectual content falls, but which at the sametime is totally devoid of specific meaning [Cassirer, 1923,p. 6].


operate according to, theories of the environment.Insightful or "reasoning" behavior in lower animalsfalls far short of this ability. It is in this sensethat comments, such as Sellars' (1963) crypticaphorism that with language, man's actions cameto have reasons as well as causes, are to be under-stood. But returning to the central nervous sys-tem as a theory, how could it come to recognizeabstract entities as underlying the surface structureof appearances?

We must explore several possible ways (and theirimplications for psychological theory construction)to see what insights they have for the constructionof a genuine grammar of behavior. One possibilityis that organisms do not become familiar withabstract entities. This view says that the surface-structure world of appearances is the only world:that appearance is reality. Theories of organismsbased on such a view can be rejected out of hand;they must postulate an infinite complexity withinthe central nervous system to account for behaviorin infinitely varied circumstances. This is why"habit" or similar dispositional analyses fail. As-sociationistic theories, insofar as they are gram-mars defined in terms of nonabstract or terminalvocabulary elements, fail for this reason.

A second possible way for organisms to becomefamiliar with abstract entities is indirectly. Dis-tinctive feature analysis or analysis by synthesisprocedures exemplifies the indirect method. Theproblem of handwriting recognition by machine isa case in which an ill-defined concept (a writtenalphabetic character) must be recognized by thismethod. The physical invariance between differ-ent people's handwriting is much less than the in-variance between the characters of the alphabet.That is, there is more variability in the way peoplemake the same characters than there is betweencharacters in the same alphabet. There are anindefinite number of ways to make an "A," eachof which is recognizable as an "A." Hierarchicallyorganized feature detectors such as Selfridge's(1959) Pandemonium, if made complex enough,can recognize such ill-defined concepts. Thus, ab-stract entities could be recognized by the organismif something comparable to "feature detectors" arefound in the nervous system. The work of Lettvin,Maturana, McCulloch, and Pitts (1959) and Hubeland Wiesel (1959, 1962) provides prima facie sup-port for the existence of such detectors.

The third way in which an organism could be-come aware of abstract entities is directly. Thismust be qualified immediately, for direct (in thesense of noninferential or certain) knowledge is amyth. But, whereas it has been assumed by empiri-cism and associationism that what was given inexperience (presented to the organism's centralnervous system) were particulars, it is at least con-ceivable that what is presented are universal.Might it not be that we are structured such that inlooking at an object we perceive it not as a par-ticular object but as an instance of a thing-kind?If that is the case, then perception is fundamentallyof abstract entities, even though our experience iswith so-called particular or "concrete" instances.Thus, this alternative claims that insofar as we are"directly" aware of anything, it is universals ratherthan particulars. This is, of course, the view ofHayek mentioned above.

On Hayek's view, the theory which has beenbuilt into the organism's central nervous systemoperates upon truly abstract entities. This is whatKant had in mind in elaborating the a priori cate-gories of human understanding. The innateKantian categories operated with such abstract en-tities as space, time, causality, etc. But againstKant, these categories (at least some of them) arepart and parcel of all sentient creatures, not justman alone. An organism that perceives abstractentities directly would be able to tell from incom-plete physical representation (surface structures)the nature of the underlying deep-structure reality.

So we must evaluate two possible ways by whichorganisms could operate with abstract entities: di-rectly and indirectly. At first, the indirect methodseems the most likely candidate, especially for lowerorganisms. There is strong prima facie evidencefor feature detectors in the peripheral processingmechanisms of as lowly a creature as the frog.But does the frog's eye tell the frog's brain aboutabstract entities (bugs) only indirectly?

Even if the frog's eye tells its brain about "bugs"by a complex feature detector network, it does notfollow that these abstract entities (bugs are definedby rules, not by listing their potentially infinitephysical characteristics) are known by the frogonly indirectly. For to say that the frog's retina isequipped with feature detectors that indirectly alertthe higher centers of its brain to the presence ofabstract entities merely relocates the problem—itdoes not solve it. How do the feature detectors at


the retinal level in turn sense abstract entities?There is only one conceivable answer: they do sodirectly. Perception is essentially transformationalas Lenneberg (1967) indicated, and it deals (if oneunderstands that, he can say therefore it deals)with abstract entities.

What does this say for the future of psychology?Perhaps a brief example will indicate some of theramifications, at least for learning. The basic par-adigm for learning will be concept formation ratherthan conditioning (either classical or instrumental).Concept formation is the "fundamental" or basicmode of learning, not to be explained by condition-ing, but rather instanced by conditioning. Con-cept formation proceeds according to rules thatwill often (with higher organisms) range over ab-stract entities, as when, say, a rat learns "to goto a corner and turn," or "to find the goal." Thefundamental rules will be concerned with abstractentities: associations, be they between stimuli andresponses, or ideas, or whatever, are at bestepiphenomenal manifestations of underlying pro-cesses. Cornering, turning, finding a goal, andcountless other functional "responses" cannot bedefined in terms of observed behavior because thelist of physical distinctive features involved wouldbe infinitely long. These responses are abstract,defined only by rules. Thus, a grammar of eventhe rat's behavior must employ abstract entities inits derivations. Psychology as a whole now mustfollow psycholinguistics back to Plato. And thatleads us to the moral, if such there be, in thisarticle.

Have We Learned Anything about Learningin the Last 2,500 Years?Not long ago psychologists would have scoffed atanyone asking such an "absurd" question. Whocould look, for example, at the vast Skinnerianliterature on contingencies of reinforcement (Honig,1966; Skinner, 1969) or behavior modification(e.g., Krasner & Ullman, 1965), or the work onmathematical learning theory and probabilitylearning (Crothers & Suppes, 1967), etc., etc., andseriously question that we have learned a lot aboutlearning?

But there are other frameworks than behavior-ism, and from their point of view, it is not obviousthat we have learned so much or, indeed, anythingof importance at all. One thing that cognitive psy-

chology has learned (again) from the revolution inlinguistics is that stimulus-response associationisticbehaviorism cannot be correct. And regardless ofthe tenability of behaviorism, it never told us aboutlearning at all, only about behavior. We mayknow quite a bit about learned behavior (definedvia the transfer of training paradigms) but verylittle about the nature and mechanisms of learn-ing per se. The fact that Plato's paradoxes stillhave their paradoxical consequences is sufficient tomake the point. And the point is driven home bythe startling (and somewhat disturbing) fact thatPlato's "solutions" are still the best we have. Be-haviorism, with the doctrines of associationismand nominalism, dissolved our paradoxes. I havesketched some of the developments in contemporary(and not so contemporary) philosophy and psy-chology that have forced their reintroduction aslegitimate paradoxes. But I have most definitelynot solved or resolved either paradox, nor hasPolanyi7 or anyone else. The value of Platonicpsycholinguistics and the transformational ap-proach to cognitive psychology lies in the fact

7 A relatively novel account of the problems posed by theMeno has been advanced by Michael Polanyi in connectionwith his conception of "tacit knowing" (see especiallyPolanyi, 1966). Polanyi takes as Plato's problem, for whichthe doctrine of recollections is advanced as the solution,that of resolving the paradox that stems from pointing outthe absurdity of believing that one begins either scientificor philosophical inquiry by seeing a problem, Polanyi feelsthis is paradoxical, and advances his conception of tacitknowledge to resolve the paradox.

I believe that Polanyi has lost sight of the primary prob-lem that Plato proposed the doctrine of recollections tosolve. With the advantage of historical hindsight providedby the Chomskyian revolution, I think that Plato had theseeming anomaly or "paradox" of the productivity of be-havior very clearly in mind in the Meno. His message isthus primarily that we know and do more than our priorexperiences have given us practice with, and only inci-dentally that we can do things which we cannot expressverbally. That is, his message is that our knowledge andability (competence) goes far beyond our experience andpractice (performance). To say that we know tacitly, asPolanyi defines the concept, cannot explain how this canbe so. For Polanyi "tacitly" accepts the doctrine thatlearning is transfer of training and requires experience:hence, tacit knowledge cannot account for transcendence ofprior learning and experience. Even admitting that hisaccount explains how we can know more than we are ableto tell, it cannot account for knowing in the absence of alearning history. It is this latter problem which is againcoming to be seen as crucial, and it is the transformationalgrammarian's "solution" to it that I feel is perhaps themost significant result of psychological inquiry since Plato'stime. Yet, the point remains: Polanyi's account, thoughinformative and tantalizing, has not solved any of theparadoxes we have discussed.


that it can and does recognize the gravity ofPlato's paradoxes, not in any solution they haveproposed. Indeed, Plato himself had no real solu-tion: his insight appears to have been in the cor-rect location of the problems, not in their solution.He had two vast promissory notes that served torelocate the problems into different areas. Thisarticle has argued that his insight was better thanthat of his successors, from Aristotle down to theneobehaviorists. In 20-odd centuries we have man-aged to learn nothing at all "new" about the natureof knowledge and learning. And that does notaugur well for the future of psychology. Perhapswe are doomed to have a (tolerably efficient)technology of behavior modification, but no scienceof knowledge and learning at all. One might con-template these caustic remarks by C. D. Broad(1933) in this regard:

Poor dear Psychology, of course, has never got far beyondthe stage of mediaeval physics, except in its statistical de-velopments, where the labours of the mathematicians haveenabled it to spin out the correlation of trivialities intoendless refinements. For the rest it is only too obviousthat, up to the present, a great deal of Psychology con-sists mainly of muddle, twaddle, and quacksalving, tryingto impose itself as science by the elaborateness of its tech-nical terminology and the confidence of its assertions[p. 476].

Or, to develop the same point from a differentperspective, one might consider the parallels in thecareers of behaviorism in psychology and the logical"-isms" in philosophy. Logical positivism and itssuccessors, despite great technical virtuosity andsuccess, stifled philosophical progress because theyignored the history of philosophy. Indeed, thepractitioners convinced themselves that the con-ceptual history of science was irrelevant to thephilosophy of science. Behaviorism and its suc-cessors, despite equally great technical virtuosityand a fair amount of success in the modificationof behavior, have stifled psychology's progress be-cause they have ignored the history of psychology'sproblems. But, in psychology no less than inphilosophy, to ignore history is to be doomed torepeat it. And this is exactly what the field hasdone with regard to the human higher mentalprocesses.

REFERENCES

BERLYNE, D. E. Structure and direction in thinking,New York: Wiley, 1965.

BRADLEY, F. H. The principles of logic. London: Methuen,1883.

BROAD, C. D. The "nature" of a continuant. Originally in,Examination of McTaggart's philosophy. Vol. 1. Cam-bridge: Cambridge University Press, 1933. Reprinted inH. Feigl & W. Sellars (Eds.), Readings in philosophicalanalysis. New York: Appleton-Century-Crofts, 1949.

CASSIRER, E. Substance and function and Einstein's theoryof relativity. Chicago: Open Court Press, 1923. Re-printed by Dover Press, 1953.

CHOMSKY, N. Aspects of the theory of syntax. Cambridge:M.I.T. Press, 1965.

CHOMSKY, N. Cartesian linguistics. New York: Harper &Row, 1966. (a)

CHOMSKY, N. Topics in the theory of generative grammar.The Hague: Mouton, 1966. (b)

CHOMSKY, N. The formal nature of language. In E. H.Lenneberg (Ed.), Biological foundations of language.New York: Wiley, 1967.

CROTHERS, E., & SUPPES, P. Experiments in second-language learning. New York: Academic Press, 1967.

DOLLARD, J., & MILLER, N. E. Personality and psycho-therapy. New York: McGraw-Hill, 1950.

ESTES, W. K. Toward a statistical learning theory. Psy-chological Review, 1950, 57, 94-107.

GOODMAN, N. The structure of appearance. Cambridge:Harvard University Press, 1951.

HANSON, N. R. Patterns of discovery. Cambridge: Cam-bridge University Press, 1958.

HAYEK, F. A. The sensory order. London: Routledge,1952. Reprinted by University of Chicago Press, 1963.

HAYEK, F. A. The primacy of the abstract. In A.Koestler & J. R. Smythies (Eds.), Beyond reductionism.New York: Macmillan, 1969.

HOFFDING, H. Outlines of psychology. New York: Mac-millan, 1891.

HONIG, W. Operant behavior: Areas of research and appli-cation. New York: Appleton-Century-Crofts, 1966.

HUBEL, D. H., & WIESEL, T. N. Receptive fields of singleneurons in the cat's striate cortex. Journal of Physiology,1959, 148, 574-591.

HUBEL, D. H., & WIESEL, T. N. Receptive fields, binocularinteraction, and functional architecture in the cat's visualcortex. Journal of Physiology, 1962, 160, 106-154.

HULL, C. L. Knowledge and purpose as habit mechanisms.Psychological Review, 1930, 37, 511-525.

JENKINS, J. J., & PALERMO, D. S. Mediation processes andthe acquisition of linguistic structure. In U. Bellugi &R. Brown (Eds.), The acquisition of language. Mono-graph of the Society for Research in Child Development,1964, 29, 141-169.

KNEALE, W. Probability and induction. Oxford: ClarendonPress, 1949.

KORNER, S. Experience and theory. London: Routledge& Kegan Paul, 1966.

KRASNER, L., & ULLMAN, L. P. Research in behavior modi-fication. New York: Holt, Rinehart & Winston, 1965.

KUHN, T. S. The structure of scientific revolutions. (2nded.) Chicago: University of Chicago Press, 1970.

LENNEBERG, E. H. Biological foundations of language.New York: Wiley, 1967.


LETTVIN, J. Y., MATURANA, H. R., MCCULLOCH, W. S., &PITTS, W. H. What the frog's eye tells the frog's brain.Proceedings of the Institute of Radio Engineering, 1959,47, 1940-1951.

MAXWELL, G. Scientific methodology and the causal theoryof perception. In I. Lakatos & A. Musgrave (Eds.),Problems in the philosophy of science. Amsterdam:North-Holland, 1968. (a)

MAXWELL, G. Reply. Scientific methodology and thecausal theory of perception. In I. Lakatos & A. Musgrave(Eds.), Problems in the philosophy of science. Amster-dam: North-Holland, 1968. (b)

NEISSER, U. Cognitive psychology. New York: Appleton-Century-Crofts, 1967.

OSGOOD, C. E. Method and theory in experimental psy-chology. New York: Oxford University Press, 1953.

OSGOOD, C. E. Motivational dynamics of language be-havior. Nebraska Symposium on Motivation, 1957, 5,348-424.

OSGOOD, C. E. On understanding and creating sentences.American Psychologist, 1963, 18, 735-751.

PAP, A. Semantics and necessary truth. New Haven:Yale University Press, 1958.

POLANYI, M. The tacit dimension. New Haven: YaleUniversity Press, 1966.

POPPER, C. R. Conjectures and refutations. New York:Basic Books, 1963.

QUINE, W. V. 0. Word and object. Cambridge: M.I.T.Press, 1960.

SE'LFRIDGE, 0. G. Pandemonium: A paradigm for learn-ing. In, The mechanization of thought processes. Lon-don: Her Majesty's Stationary Office, 1959.

SELLARS, W. Science, perception, and reality. New York:Routledge & Kegan Paul, 1963.

SESONSKE, A., & FLEMING, N. Plato's Meno: Text andcriticism. Belmont, Calif.: Wadsworth, 1965.

SKINNER, B. F. Contingencies of reinforcement: A theo-retical analysis. New York: Appleton-Century-Crofts,1969.

TURNER, M. B. Philosophy and the science of behavior.New York: Appleton-Century-Crofts, 1967.

WADDINGTON, C. H. The theory of evolution today. InA. Koestler & J. R. Smythies (Eds.), Beyond reduc-tionism. New York: Macmillan, 1969,

WARREN, H. C. A history of the association psychology.New York: Scribner's, 1921.


psycholinguistics and plato's paradoxes of the menopsycholinguistics and plato's paradoxes...

Documents