128

CHAPTER SEVEN--LEARN1NG

man. Probably the same may be true of rage although at present we are not"
so sure of this.
The Freudians twenty years fromnow, unless their hypotheses change.
when they come to analyze Albert's fear of a seal skin coat-assuming that
he comes to analysis at that age-will probably tease from him the recitalof
a dream which upon theiranalysis will show that Albertat threeyears of age R ....ziew of B, F. Skinners verbal Behavior
attempted to play with the pubic hait of the mother and was scolded vio-
lently for it. (Weare by no means denying that this might in some othercase Noam Chomsky
condition it.) If the analysthas sufficiently preparedAlbert to acceptsuch a
dream when found as an explanation of his avoiding tendencies. and
analyst has the authorityand personality to put it over, Albert may be I
convinced that the dream was a true revealerof the factors which brouaht
about the fear. linguists and philosophers concerned with language have
It is probable that many of the phobias in psychoparhology are that their studies might ultimately be embedded in a
conditioned emotional reactions either of the direct or the transferred type. provided by behaviorist psychology, and that tefractory areas of
One maypossiblyhave to believe that such persistence of earlyconditioned in which meaning is involved, will in this
responses will be found only in persons who are constitutionally inferior. fruitful exploration. Since this volume is the first large-
Ourargument is meant to be constructive. Emotional disturbances in adults incorporate the major aspects of linguistic behavior within
cannot be traced backto sex alone. Theymust be retraced along at leastthree ,h~'liotist framework, it merits and will undoubtedly receive careful at-
collateral lines-to conditioned and transferred responses set up in i8n"Skinner is noted for his conttibutions ro the study of animal behav-
and earlyyouth in all three of the fundamental human emotions. review is the product of studyof linguistic behavior ex-
twentyyears. Earlier versions of it have been fairly
are quite a few references in the psychological
major ideas. .
problem to which this book is addressed is that of giving a "func-
behavior. By functional analysis. Skinnermeans
the variables that conttol this behavior and specification of
to determine a particular verbal response. Furthermore,
~d,itl'<,llil1g"at'iat>les are to be desctibed completely in terms of such no-
"""u"us, rei"f6"cel""'~ deprivation, which havebeen given. a reason-
animal experimentation.Tn.. orher wordsc rhe goal
a way to predict and control verbal behavior by
manipulating the physical environment of the speaker.
recent advances in the labOratory studyof animal
riennit us to approach this problem with a certain optimism,since
ai,·. verh,.1 behavior its special char-
[of this experimental
\
129
 CHAPTER SEVEN--LEARN I NG CHOMSKY • A REVIEW OF B. F. SKINNER'S VERBAL BEHAVIOR" , 3'

work] have been surprisingly free of species restrictions. Recent work has specification of it. If the contribution of the organism is complex, the only
shown that the meth~ds can be extended to human behavior without seri- hope of predicting behavior even in a gross way will be through a very indi-
ous modification." rect program of research that begins by studying the detailed character of the
It is important to see clearly just what it is in Skinner's program and behavior itself and the particular capacities of the organism involved.
claims that makes them appear so bold and remarkable. It is not primarily Skinners thesis is that external factors consisting of present stimulation
the fact that he has set functional analysis as his problem, or that he limits and the history of reinforcement (in particular, the frequency, arrangement,
himself to study of observables, i.e., input-output relations. What is so sur- and withholding of reinforcing stimuli) are of overwhelmingimportance, and
prising is the particular limitations he has imposed on the way in which the that the general principles revealed in laboratory studies of these phenomena
observables ofbehavior are to be studied, and, above all, the particularly sim- provide the basis for understanding the complexities of verbal behavior. He
ple nature of the junction which, he claims, describes the causation of behav- confidently and repeatedlyvoiceshis claim to have demonstrated that the con-
ior. One would naturally expect that prediction of the behavior of a complex tribution of the speaker is quite trivial and elementary,and that precise predic-
organism (or machine) would require, in addition to information about ex- tion of verbal behavior involves only specification of the few external factors
ternal stimulation, knowledge of the internal structure of the organism, the that he has isolated experimentallywith lower organisms.
ways in which it processes input information and organizes its own behav- Careful study of this book (and of the research on which it draws) re-
ior. These characteristics of the organism are in general a complicated prod- veals; however, that these astonishing claims are far from justified. It indi-
uct of inborn structure, the genetically determined course of maturation, cates, furthermore; that the insights that have been achieved in the labora-
and past experience. Insofar as independent neurophysiological evidence is tories of the reinforcement theorist, though quite genuine, can be applied to
not available, it is obvious that inferences concerning the structure of the or- complex human behavior only in the most gross and superficial way,
ganism are based on observation of behavior and outside events. that speculative attempts to discuss linguistic behavior in these terms
Nevertheless, one's estimate of the relative importance of external fac- omit from consideration factors of fundamental importance that
tors and internal structure in the determination of behavior will have an im- doubt, amenable to scientific study, although their specific character
portant .effect on the direction of research on linguistic (or any other) be- at present be precisely formulated. Since Skinner's work is the
havior, and on the kinds of analogies from animal behavior studies that will sive attempt to accommodate human behavior involving higher
be considered relevant or suggestive. ulties within a strict behaviorist schema of the type that has attracted
Putting it differently, anyone who sets himself the problem of analyz- linguists and philosophers, as well as psychologists, a ceranec dccumenra
ing the causation of behavior will (in the absence of independent neuro- don is of independent interest. The magnitude of the
physiological evidence) concern himself with the only data available, namely to account for verbal behavior serves as a kind of measure
the record of inputs to the organism and the organism's present response, of the factors omitted from consideration, and an indication
and will try to describe the function specifying the response in terms of the really known about this remarkably complex phenomenon,
history of inputs. This is nothing more than the definition of his problem.
There are no possible grounds for argument here, if one accepts the problem
as legitimate, though Skinner has often advanced and defended this defini-
tion of a problem as if it were a thesis which other investigators reject. The
differences that arise between those who affirm and those who deny the im- Consider first Skinner's use of the notions stimu!usandresponse. In Behavior
portance of the specific "contribution of the organism" to learning and per- of Organisms (9) he commits himself to the narrow definitions for these
formance concern the particular character and complexity of this function, terms. A part of the environment and a part of behavior are called stimulus
and the kinds of observations and research necessary for arriving at a precise (eliciting, discriminated, or reinforcing) and response, respectively, only if
\
'3 2 CHAPTER SEVEN--LEARNING (HOMSKY • A REVIEW OF B. F. SKINNER'S VERBAL BEHAVIOR· 133

they are lawfully related; that is, if the dynamic laws relating them show words Eisenhower and Moscow, which I presume are proper nouns if any-
smooth and reproducible curves. Evidently, stimuli and responses, $0 de- thing is, but have never been stimulated by the corresponding objects. How
fined, have not been shown to figure very widely in ordinary human behav- can this fact be made compatible with this definition? Suppose that I use the
ior. We can, in the face of presently available evidence, continue to maintain name of a friend who is not present. Is this an instance of a proper noun un-
the lawfulness of the relation between stimulus and response only by de- the control of the friend as stimulus? Elsewhere it is asserted that a stim-
priving them of their objective character.A typical example of stimulus control controls a response in the sense that presence of the stimulus increases
for Skinner would be the response to a piece of music with the utterance the probability of the response. But it is obviously untrue that the probabil-
Mozart or to a painting with the response Dutch. These responses are as- that a speaker will produce a full name is increased when its. bearer
serted to be "under the control of extremely subtle properties" of the physi- the speaker. Furthermore, how can one's own name be a proper noun
cal object or event (108). Suppose instead of saying Dutch we had said sense?
Clashes withthewallpaper, I thought you liked abstract work, Never saw it before, A multitude of similar questions arise immediately.It appears rhar the
Tilted, Hanging too low, Beautiiu], Hideous, Rememberour campingtrip last sum- control here is merely a misleading paraphrase for the traditional denote
mer?, or whatever else might come into our minds when looking at a picture refi":rh,,.,,serti()D (115) that so far as the speaker is concerned, the relation
(in Skinnerian translation, whatever other responses exist in sufficient referenceis "simplythe probability that the speaker will emit a responseof
strength). Skinner could only say that each of these responses is under the form in the presence of a stimulus having specified properties" is
control of some other stimulus property of the physical object. If we look at surely incorrect if we take the words presence, stimulus,and probability in their
a red chair and say red,the response is under the control of the stimulus red- sense. That they are not intended to be taken literally is indicated by
ness; if we say chair, it is under the control of the collection of properties (for many examples, as when a response is said to be "controlled" by a situation or
Skinner, the object) chairness (110), and similarly for any other response. of affairs as "stimulus." Thus, the expression a needle in a haJvsttICk
This device is as simple as it is empty. Since properties are free for the asking COI1tro!1"d as a unit by a particular type of situation" (116); the words
(we have as many of them as we have nonsynonymous descriptive expres- part of speech, e.g., all adjectives, are under the control of a
sions in our language, whatever this means exactly), we can account for a _',,,1,.1. properties of stimuli (121); "the sentence The boyruns a
wide class of responses in terms of Skinnerian functional analysis by identi- control of an extremelycomplex stimulus situation"
fying the controlling stimuli. But the word stimulus has lost all objectivity in may function as a standard response under the
this usage. Stimuli are no longer part of the outside physical world; they are which might also control He is ailing" (325); when an
driven back into the organism. We identify the stimulus when we hear the in a foreign country and reports upon his return, his
response. It is clear from such examples, which abound, that the talk ofstim- "remote stimulus control" (416); the statement Thisis war
ulus controlsimply disguises a complete retreat to mentalistic psychology. We "confusing international situation" (441); the
cannot predict verbal behavior in terms of the stimuli in the speaker's envi- "subtle property of stimuli which we speak of as action-in-the-past"
ronment, since we do not know what the current stimuli are until he re- as the -s in The boyruns is under the control
sponds. Furthermore, since we cannot control the property of a physical ob- situation as its "currency" (332). No characterization
ject to which an individual will respond, except in highly artificial cases, is remotely related to experiment (or that
Skinner's claim that his system, as opposed to the traditional one, permits the faintest objectivity) can be made
the practical control of verbal behavior is quite false. in which, for example,rhe controlling stimulus
Other examples ofstimulus controlmerely add to the general mystifica- on the responding organism.
tion. Thus, a proper noun is held to be a response "under the control of a Consider now ~kinner's use of the of
specific person or thing" (as controlling stimulus, 113). I have often used the identifying units in verbal behavior has concern of
\
Bilkent Umvel'$!t)'
Ubrary
134 CHAPTER SEVEN--LEARN1NG (HOMSKY • A REVIEW OF B. F. SKINNER'S VERBAL BEHAVIOR 135

linguists, and it seems.very likely that experimental psychologists should be ties, etc. These are very remarkable abilities. We constantly read and hear
able to provide much-needed assistance in clearing up the many remaining new sequences of words, recognize them as sentences, and understand
difficulties in systematic identification. Skinner recognizes (20) the funda- them. It is easy to show that the new events that we accept and understand
mental character of the problem of identification of a unit of verbal behav- as sentences are not related to those with which we are familiar by any sim-
ior, but is satisfied with an answer so vague and subjective that it does not ple notion of formal (or semantic or statistical) similarity or identity of
really contribute to its solution. The unit of verbal behavior-the verbal op- grammatical frame. Talk of generalization in this case is entirely pointless
erant-is defined as a class of responses of identifiable form functionally re- empty.It appears that we recognize a new item as a sentence not because
lated to one or more controlling variables. No method is suggested for de- matches some familiar item in any simple way,but because it is generated
termining in a particular instance what are the controlling variables, how the grammar that each individual has somehow and in some form inter-
many such units have occurred, or where their boundaries are in the total re- And we understand a new sentence, in part, because we are some-
sponse. Nor is any attempt made to specify how much or what kind of simi- of determining the process by which this sentence is derived in
larity in form or control is required for two physical events to be considered grammar.
instances of the same operant. In short, no answers are suggested for the Suppose that we manage to construct grammars having the properties
most elementary questions that must be asked of anyone proposing a outlined above.We can then attempt to describe and study the achievement
method for description of behavior. Skinner is content with what he calls an speaker, listener, and learner. The speaker and the listener, we must
extrapolation of the concept of operant developed in the laboratory to the have already acquired the capacities characterized abstractly by the
verbal field.In the typicalSkinnerian experiment, the problem of Identifying g~ammar. The speaker's task is to select a particular compatible set of op-
the unit ofbehavior is not too crucial. It is defined, by fiat, as a recorded peck rules. If we know, from grammatical study, what choices are available
or bar-press, and systematic variations in the rate of this operant and its re- and what conditions of compatibility the choices must meet, we can
sistance to extinction are studied as a function of deprivation and schedul- proceed meaningfullvto investigate the factors that lead him to make one or
ing of reinforcement (pellets). The operant is thus defined with respect to a choice. The listener (or reader) must determine, from an exhibited
particular experimental procedure. This is perfectly reasonable and has led optional rules were chosen in the construction of the utter-
to many interesting results. It is, however, completely meaningless to speak It must be admitted that the abilityof a human being to do this farsur-
of extrapolating this concept of operant to ordinary verbal behavior. Such our present understanding. The child who learns a language has in
"extrapolation" leaves us with no way of justifying one or another decision sense constructed the grammar for himself on the basis of his obser-
about the units in the "verbal repertoire." vation of sentences and nonsentences (i.e., corrections by the verbal com-
munity). Study of the actual observed ability of a speaker to distinguish sen-
from nonsentences, detect ambiguities, etc., apparently forces us to
conclusion that this grammar is of an extremely complex and abstract
character, and that the young child has succeededin carryingout what from
The behavior of the speaker, listener, and learner of language constitutes, of formal point of view, at least, seems to be a remarkable type of theory
course, the actual data for any study oflanguage. The construction of a gram- construction. Furthermore, this task is accomplished in an astonishingly
mar which enumerates sentences in such a way that a meaningful structural time, to a large extent independently of intelligence, and in a compa-
description can be determined for each sentence does not in itself provide an way by all children.Any theory of learning must cope with these facts.
account of this actual behavior. It merely characterizes abstractly the ability It is not easy to accept the view that a child is capable of constructing
of one who has mastered the language to distinguish sentences from non- extremelycomplex mechanism for generating a set of sentences, some of
sentences, to understand new sentences (in part), to note certain ambigui- which he has heard, or that an adult can instantaneously determine whether
\
13 6 CHAPTER SEVEN--LEARNING

(and if S0, how) a particular item is generated by this mechanism, which has
many of the properties of an abstract deductive theory. Yet this appears to be
a fair description of the performance of the speaker, listener, and learner. If
this is correct, we can predict that a direct attempt to account for the actual
behavior of speaker. listener, and learner, not based on a prior understand-
ing of the structure of grammars,willachieveverylimited success.The gram- From Wild Minds: What Animals Really Think
mar must be regarded as a component in the behavior of the speaker and lis-
tener which can only be inferred, as Lashley has put it, from the resulting Marc D. Hauser
physical acts. The fact that all normal children acquire essentially compara-
ble grammars of great complexity with remarkable rapidity suggests that hu- Afemale [apanese macaque drops a heap of wheat arid sand into the ocean,
man beings are somehow specially designed to do this, with data-handling the wheat off the surface once the sand has settled to the bot-
or "hypothesis-formulating" ability of unknown character and complexity. tom ..Although this technique is now a tradition in.the population of monkeys
The study of linguistic structure may ultimately lead to some significant in- [apanese island of Koshima, it was invented by a highly creative
sights into this matter. At the moment the question cannot be seriously members ofthe population. Naive blue tits,
posed, but in principle it may be possible to study the problem of determin- birds remove foil from a milk bottle and then drink the rich
ing what the built-in structure of an information-processing (hypothesis- top, will then operate the foil in the same way. Human infants;
forming) system must be to enable it to arrive at the grammar of a language only one hour old, stick out their tongues after an adult has performed the
from the available data in the available time. At any rate, just as the attempt same display. An unmated female guppy will copy the mating preferences of
to eliminate the contribution of the speaker leads to a "mentalistic" descrip- another female.if she watches the model's selection of males. These observa-
tive system that succeeds only in blurring important traditional distinctions, all well documented, suggest that in group living animals, an individ-
a refusal to study the contribution of the child to language learning permits ual's actions are highly influenced by social interactions. But howdotheseso-
only a superficial account of language acquisition, with a vast and unana- interactions help in solving the problems of extracting food, choosing a
lyzed contribution attributed to a step called generalization which in fact finding safe refuge from predators?
includes just about everything of interest in this process. If the study of Assume you observed a Japanese macaque pick upa heap of Vv'heat~rl~
language is limited in these ways, it seems inevitable that major aspects of sand, walk over to the water, drop the mixture in andthe~,asthesaI1ddro~s
verbal behavior will remain a mystery. down, skim the wheat off the surface and eat it. How did this rnonkey,()rat1Y
of the others in the population, acquire the wheat-Vv'as~in~tec~niqtle?8~e
possibility is that a naive individual walks over to t~eVv'ater\Vitl1ou.tanygr(mp
members in sight. Some wheat is floating in the\Vater.S~es~rnsitoffand
eats it. All of a sudden she is struck by insightian~4e~tl.cesthea~sVv'ertothis
foraging problem. Like a contestant playingfeopardy, she has been givenrhe
solution and must work out the question. 'I'hea~s\V~~is,uThewheatfloats;
the sand sinks." The question is, "What happe~s.V/~e~ youbring Viheatand
sand over to the water and drop the mixture iIll" I~tl1is scenario; deduction,
not social learning, drives skill a(;quisitionaf1dt~usk!l0\Vledge.Here's a sec-
ond, similarly asocial method of discoveryr the animal walks over to the water

, for a dtink and happens to have some sand arid wheat on her hands. As she