Bloomfield and skinner: speech-community, functions of language, and scientific activity.

Abstract

The approaches of Bloomfield and Skinner on the topics of speech-community, functions of language, and language and scientific activity are characterized; a comparative analysis follows; and approximations between the two authors are pointed out. The most important approximations are the verbal community as the ultimate source of language; the main function of language of obtaining practical effects in the world; the characterization of a considerable part of scientific activity as verbal; and the scientific activity as being deeply committed with successful action. The difference found is related to the subject matter: Bloomfield describes the system of the language of the community, and Skinner does the functional analysis of the repertoire of the individual.

Keywords: Bloomfield, Skinner, verbal behavior, speech-community, functions of language, scientific activity.

Introduction

The vitality of the works of Leonard Bloomfield (1887-1949) and B. F. Skinner (1904-1990) is manifest not only in that frequent and explicit references are made to them but, more importantly, in that much of the knowledge produced in their fields--linguistics and behavior analysis, respectively--rests on the pillars of their concepts and analyses.

The influence of Bloomfield's analysis of language on Skinner's analysis of verbal behavior has been pointed out with regard to the following topics: use of the analogy model to explain, in part, the creativity of verbal behavior (Matos & Passos, 2004); the phoneme as a unit of analysis (Passos & Matos, 1998; Joseph, Love & Taylor, 2001); as well as science and scientific method, verbal episode, and meaning (Passos & Matos, submitted). These authors wrote on complementary subject matters: Bloomfield studied the language--or the practices of the verbal community, as Skinner prefers to call it--while Skinner studied the individual speaker's verbal behavior, for which the practices of the verbal community are pre-requisite (Passos & Matos, submitted).

This paper focuses on the similarities in Bloomfield and Skinner's approach to the topics of speech-community, functions of language, and language and scientific activity.

Leonard Bloomfield and his concepts of act of speech and meaning

Bloomfield was the preeminent scientist in his field in the United States, and his influence was strongly felt in Europe, as well (Fought, 1995, Koerner, 2003; Robins, 1997, pp. 237-238). He shaped structural linguistics, and his teaching is at the heart of present-day linguistics (Hockett, 1984; Matthews, 1992/1999, pp. 139, 149, 2001, pp. 142ff). Bloomfield worked in the areas of general linguistics, the description of languages, and historical and comparative linguistics. In addition, he did research in scientific methodology and created instructional methods for reading and foreign language. His book entitled Language (1933/1961), praised by so many linguists (see, e.g., Bolling, 1935/1970; Coseriu, 1986; Edgerton, 1933/1970; Howatt, 2002; Kroesch, 1933/1970; Lepschy, 1982; Sturtevan, 1934/1970), earned the following comment from Hockett in his foreword to the 1984 edition:

   [The book] is considered by many to be the most important general
   treatise on language ever written. First published just fifty-one
   years ago, it towers above all earlier works of the sort and, to
   date, above all more recent ones. Its author systematically sets
   forth, in prose poetic for its simplicity, the crucial findings of
   linguistics up to that time, with a perceptivity and realism
   derived from his own wide experience, monitored throughout by a
   reverent respect for his predecessors and an intimate knowledge of
   their achievements. (1984, p. ix)

A number of explicit references to behaviorism in his works attest to Bloomfield's embracing of it (see, e.g., Bloomfield, 1926/1970, 1927/1970, 1930/1970, 1931/1970, 1932/1970, 1933/1961, pp. 512, 515, 1935/1970, 1936/1970, 1939, 1942/1970), and behaviorism played an important role in some of his central concepts, including "act of speech" and "meaning."

Bloomfield offers his "act of speech"--a simple interaction between two persons--as a paradigm of what happens in language-mediated interactions:

   Suppose that Jack and Jill are walking down a lane. Jill is hungry.
   She sees an apple in a tree. She makes a noise with her larynx,
   tongue, and lips. Jack vaults the fence, climbs the tree, takes the
   apple, brings it to Jill, and places it in her hand. Jill eats the
   apple. ... the incident consists of three parts, in order of time:

A. Practical events preceding the act of speech.

B. Speech.

C. Practical events following the act of speech." (Bloomfield, 1933/1961, pp. 22-23)

Instead of taking the apple from the tree, which would be a practical reaction (R) to the practical stimulus (S), Jill manifested a linguistic substitute reaction (r), asking for the apple by means of speech. Jack reacted to the stimulus (s) of Jill's linguistic substitute reaction through a practical reaction (R), taking the apple from the tree and giving it to Jill (pp. 23-26). The situation is characteristic of a reaction mediated by speech:

S [right arrow] r ... s [right arrow] R (Bloomfield, 1933/1961, p. 26)

The purpose of this paradigm is to show that language is a way of reacting to environmental stimuli through the emission of sounds--speech--that will act upon another person who, in turn, will act upon the environment. The first person, the one who emitted speech, will achieve the practical effects of speech through the practical action taken by the second person.

This paradigm also serves as a foundation for a physicalist concept of the meaning of speech: the practical events which precede and follow the speech itself. They are all physical events: "A" is the stimuli acting on Jill (her hunger, the sight of the apple, the presence of Jack, etc.), and "B" is both Jack's reaction of fetching the apple for Jill and the practical consequence for Jill of getting the apple (p. 23).

Bloomfield's concept of meaning is entirely and explicitly physicalist (pp. 139ff). He considered "the entire baggage of mentalism to be empty and useless" (Bloomfield, 1935/1970, p. 309). As was the case with so many aspects of Skinner's work (Skinner, 1974, pp. 3ff), also, Bloomfield's physicalist concept of meaning, being the opposite of the prevalent mentalistic concept, has been frequently misunderstood and misrepresented as negation of meaning (Bloomfield, 1945/1987; Fought, 1999a, 1999b; Hall, 1990, pp. 47-48; Hockett, 1999). In fact, however, Bloomfield himself stressed that language is important precisely because it has a meaning: the practical events connected with speech (1933/1961, p. 27, 1943/1970). According to him, linguistics does not have a scientific method for the study of meaning. The linguist uses meaning to identify linguistic forms, which are the real aim of linguistic description (1933/1961, pp. 139ff).

The subject matter of linguistics is related more directly to the "B" section of the act of speech, that is, to the speech itself. The linguist abstracts the features which are common to the speech of several speakers, analyzes the attested forms, and describes their combinations--the system of the language. The subject matter of linguistics is the world's languages (pp. 57ff, 158).

Skinner's knowledge of Bloomfield's works

Although Skinner has not cited Bloomfield in Verbal Behavior (1957), he did so in The William James Lectures (2) he delivered at Harvard University in 1947; he used some of its content later in Verbal Behavior (1957, p. vii). According to Skinner, because Psychology at that time was unable to give a scientific explanation of speech and reaction to speech, theoreticians in several branches of knowledge had been devising tentative explanations for their own purposes. Bloomfield presented "the essential features of language by describing [sic] in which Jack asks Jill for an apple ..." (3) (Skinner, 1947, p. 5). This is the act of speech that appears in Chapter 2 entitled "The use of language" in Bloomfield's book Language (4).

Continuing in The William James Lectures, in relation to meaning, Skinner mentions Bloomfield only as an exception to the dualistic concept of verbal behavior seen in almost all linguists, as well as in logicians, philosophers of science, and critics (p. 6). According to Skinner, Bloomfield "identifies the mngs [meanings] of a language with the whole of the universe and argues that a study of mng [meaning] would require the techniques of all the sciences. Mngs [meanings] in this sense are certainly observable ..." (5) (p. 7). Skinner is probably referring to the following paragraph from Chapter 9 of Language, which is entitled "Meaning":

   The situations which prompt people to utter speech, include every
   object and happening in their universe. In order to give a
   scientifically accurate definition of meaning for every form of a
   language, we should have to have a scientifically accurate
   knowledge of everything in the speakers' world. The actual extent
   of human knowledge is very small, compared to this. We can define
   the meaning of a speech-form accurately when this meaning has to do
   with some matter of which we possess scientific knowledge
   (Bloomfield, 1933/1961, p. 139).

Skinner refers to Bloomfield's behaviorism in The William James Lectures when addressing the concepts and techniques of various psychological provenances that have been used by those who were working in the field of verbal behavior. In this connection, he mentions that "Bloomfield uses a modified behaviorism under the influence of George Herbart Mead and A. P. Weiss" (Skinner, 1947, p. 16).

Chapter 9 entitled "Meaning" is referred to once more in The William James Lectures when Skinner distinguishes between metaphorical extensions in the speaker's verbal behavior and expressions with forms that are similar but bound by the verbal community. In this context, he mentions Bloomfield's (1933/1961, p. 150) teaching that idioms based on metaphors are language-specific: "... as Bloomfield has pointed out, the expression eye of a needle is lacking in both French and German, though it might arise as a metaphor in either case" (Skinner, 1947, p. 61).

   In The William James Lectures, Skinner explicitly mentions
   Bloomfield's anti-mentalism: As Bloomfield later put it in
   criticizing another linguist, "(Paul) accompanies his statements
   about language with a paraphrase in terms of mental processes which
   the speakers are supposed to have undergone. The only evidence for
   these mental processes is the linguistic process; they add nothing
   to the discussion, but only obscure it." (1947, p. 150)

This passage is found in Chapter 1 of Language (p. 17), called "The study of language," where Bloomfield presents a brief but clear and sophisticated history of his field.

Other references to Bloomfield in Skinner's subsequent works will be mentioned only briefly here; a more detailed presentation is found in Matos & Passos (2006). In Contingencies of Reinforcement (1969, p. 11), he describes Bloomfield as one who tried to find a behavioral explanation of language, which did not succeed because it was based on stimulus-response psychology.

In Skinner's autobiography (1979), there are further references to Bloomfield. The first one (p. 150) appears in a letter to Fred Keller in which Skinner mentions two books by Bloomfield--Introduction to the Study of Language (1914) and Language (1933); the shift of Bloomfield's psychological reference from Wundt to Weiss; and the act of speech. The other reference to Bloomfield in Skinner's autobiography is presented and discussed later in this paper.

Speech-Communities, Functions of Language, and Language and Scientific Activity

Bloomfield's concept of speech-community

Speech is a universal characteristic of human groups and is executed in many languages (Bloomfield, 1933/1961, p. 29). A group of people who share a language, interact by means of the same system of speech-signals, and constitute a speech-community (pp. 29, 42). The utility of language depends on people in the same group emitting speech-signals that are appropriate for a given occasion and reacting appropriately after hearing them emitted by other people (p. 29).

Homogeneity in a speech-community is only relative; in fact, "every person uses speech-forms in a unique way" (p. 75). Even though two people who belong to the same speech-community do not speak exactly alike, in general, their speech is sufficiently similar to meet daily communication needs. In the process of describing a language, the linguist works with a method of abstraction, ignoring certain differences in speech among various people in the same community (p. 45).

In a given speech-community, the degree of communication among people is differentiated--that is, each speaker does not communicate with all the other members of the community and communicates to varying degrees with those with whom he or she does communicate (p. 46-47). These differences in communication density are, to a large extent, due to the existence of sub-groups in the community within which individuals speak with each other to a much greater extent than with other people who are in the same community but are not part of that sub-group (p. 48). These sub-groups are mainly formed on the basis of geographical proximity as well as socio-economic factors that differentiate the various social classes (pp. 46-51).

Differences in speech among members of the same verbal community can be very minor or quite major. Similarity of speech among members of a sub-group is a matter of degree and cannot be established in absolute terms; for this reason, the term speech-community always has relative value (pp. 52-54). The linguist, then, when describing a language, not only uses abstraction but also chooses one or more of the varieties of speech found in some of the groups that make up the community (p. 311).

The differentials in communication density among members of a speech-community explain the observed differences of speech among speakers of that community since the "speaker's language, except for personal factors which we must here ignore, is a composite result of what he has heard other people say" (p. 46).

Sub-groups based on geographical and socio-economical differentiation are further subdivided on the basis of work (carpenters, miners, doctors, artists, scientists, etc.), recreational activities, age, and marginalization (criminals, gypsies, etc.) (Bloomfield, 1933/1961, pp. 46-51, 1939, pp. 3-4). Groups based on specialized activities develop a specialized terminology that favors accuracy of responses as, for example, in the case of scientists:

   The special vocabularies and turns of speech which are used in the
   various branches of science belong in this same general type; only,
   as scientific observation reaches beyond the interests of ordinary
   life, the vocabulary of science becomes very large. From timid
   neologisms it grows to a state where some scheme of word creation
   stands at the service of every member of the guild. (Bloomfield,
   1939, p. 42)

Skinner's concept of verbal community

A verbal community comprises speakers and listeners whose verbal behavior and responses to verbal behavior are the result of relatively homogeneous practices of reinforcement. To study verbal behavior means to study the processes and conditions by which a speaker's behavior is strengthened by the listener; likewise, to study the listener's behavior means to study the practices in the verbal community that maintain the listener's behavior:

   The special conditioning of the listener is the crux of the
   problem. Verbal behavior is shaped and sustained by a verbal
   environment--by people who respond to behavior in certain ways
   because of the practices of the group of which they are members.
   These practices and the resulting interaction of speaker and
   listener yield the phenomena which are considered here under the
   rubric of verbal behavior. (Skinner, 1957, p. 226)

Listeners have the basic function of reinforcing speakers' verbal behavior. Because listeners are part of the situation in which this behavior is reinforced, they become discriminative stimuli--that is, their presence alters the probability that verbal behavior will be emitted. In this latter function, they are called "the audience":

   An audience ... is a discriminative stimulus in the presence of
   which verbal behavior is characteristically reinforced and in the
   presence of which, therefore, it is characteristically strong.
   Discriminative stimuli become in turn reinforcing, and this is
   confirmed by the reinforcing effect of the appearance of an
   audience. (Skinner, 1957, p. 172)

The audience alters the probability of emission of verbal responses. The specific effect--whether weakening or strengthening--will depend on the history between the speaker and a particular listener because the audience can either reinforce or punish verbal behavior. The audience also controls groups of responses: "The repertoire under the control of an audience may be a language, a jargon, a cant, or some less sharply-defined functional subdivision of the behavior of the speaker" (Skinner, 1957, p. 186).

Linguistics describes these repertoires of languages, jargons, etc., as they appear in the verbal community, not in any particular speaker. The verbal community undertakes to train listeners by making them reinforce the verbal behaviors that have the properties prevalent in the verbal community (p. 461).

The repertoire of an individual results from his exposure to various subtypes of verbal communities during his life. Skinner specifies some of these: within the grand community of "languages" there are smaller sub-communities where jargon and technical vocabularies are spoken such as those of sailors or engineers; the "little language" of children; as well as subdivisions that strengthen verbal behavior that is "bookish," pedantic, literary, archaic, etc. (pp. 173-174).

The linguist tries to identify the verbal practices of the community (p. 461) by collecting data from several of its speakers and listeners. On the contrary, the behavior analyst wants to find the reinforcing contingencies that control verbal operants in the individual (pp. 1ff). Although the contingencies that govern the individual are, ultimately, the product of contingencies maintained by the community, the specific combination of contingencies will be unique for each speaker:

   The relation between a property of a response and a controlling
   property of a stimulus can be demonstrated only by comparing many
   instances of the verbal behavior of a single individual. Such a
   relation need not be obvious to the speaker. It may not be
   identified with any reaction of a listener or with the reinforcing
   practices of the verbal community (Skinner, 1957, p. 120).

In addition to altering the probability of the occurrence of verbal behavior and strengthening specific parcels of it, the audience has also the function of selecting the theme of verbal behavior. For example, we talk with children about playing, with parents about children, and with teachers about parents and children.

   A third function of an audience is to select a subject matter.
   Listeners differ in the extent to which they reinforce different
   types of verbal operants and, particularly, various classes of
   intraverbal responses and tacts. Given a single speaker with a
   specific history and a specific current situation, the audience
   will determine not only whether verbal behavior occurs, or the
   subdivision of the language in which it occurs, but also what types
   of responses are made and 'what is talked about.' (Skinner, 1957,
   p. 175)

There are many similarities between Bloomfield and Skinner's concepts on the subject "speech/verbal community." For both of these authors ...

* the speech/verbal community is a prerequisite for the existence of language/verbal behavior, since they are considered learned from the community where the individual lives;

* the speech/verbal community comprises a group of people who emit certain speech-forms in relation to the same situations and respond to forms of speech in the same way. Their verbal practices are similar because their language/verbal behavior results from similar situations in which relatively stable correspondences between stimuli and responses were established;

* the homogeneity of the speech/verbal community is relative, since the relation between situations and speech-forms, as well as responses to the latter, present variations within the same speech/verbal community;

* the speech/verbal community is subdivided into sub-communities formed by people of the same occupation, of similar ages, who use the same speech dialects, slang, etc. The language/verbal behavior of these sub-communities presents certain specifics in relation to the one of the larger verbal communities.

Bloomfield emphasizes that the linguist focuses on that which speakers have in common rather than on the differences in the use of speech-forms by speakers in the same community. While worded differently, Skinner's understanding of this issue is the same: the linguist registers the verbal community's reinforcement contingencies, whereas the behavior analyst searches for what controls the individual speaker's verbal behavior, and it is of no interest to him whether the variables that control the verbal behavior of this particular speaker also control the verbal behavior of other speakers.

Bloomfield's concepts of the functions of language

As previously discussed (p. 2), through a linguistic substitute stimulus--produced by the speaker's linguistic reaction to a practical stimulus--the listener presents a practical reaction related to the practical stimulus that reaches the speaker. Thus, language enhances a stimulus's power to provoke action: a stimulus that, originally, could reach only those individuals directly exposed to it is now able to reach others, as well (Bloomfield, 1933/1961, p. 26).

The practical events to which language is related--"S" and "R"--are those that have biological importance; speech is just a means, not an end in itself. The events of speech--"s......... r"--have no value to speakers and listeners in and of themselves but only through the effects they produce:

   The normal human being is interested only in S and R; though he
   uses speech, and thrives by it, he pays no attention to it. Saying
   the word apple or hearing it said, appeases no one's hunger. It,
   along with the rest of speech, is only a way of getting one's
   fellow-men to help. (Bloomfield, 1933/1961, p. 26)

Through the coordination of individuals that language allows, each person disposes of "the strength and skill of every person in the group.... The division of labor, and, with it, the whole working of human society, is due to language" (p. 24).

It can also happen that the effect of an act of speech does not appear in the form of an immediate practical action on the part of the listener, but it does change "the predisposition of the hearer for further responses: a beautiful poem, for instance, may make the hearer more sensitive to later stimuli" (p. 41). Even when the effect of the act of speech is another speech, at some point we will "find some modification of non-verbal activity: even poetry or fiction will in the end lead to a more than verbal result" (1939, p. 16).

A list made by Bloomfield of some of the ways in which we benefit from language indicates how much he was, in fact, committed to the point of view that language has the eminently practical function of making action in the non-verbal environment more efficient. Language makes possible ...

* the transmission of communication through which we can, for example, enunciate a necessity. Thus, a petition that states, We want a bridge on this little stream, directed from citizens to the mayor, could result in engineers taking action to construct a bridge after the petition passes through processing that involves several people and speeches (1933/1961, p. 28);

* abstraction, which means trying out an activity verbally before executing it through practical actions in the non-verbal world, just as we do preliminary numerical calculations. This minimizes the occurrence of errors in practical life, because many of them can be corrected while the person is dealing with the situation in speech-forms only (p. 28);

* self-stimulation, where the "utterance of a speech-form is not only a response but serves also as a stimulus to the speaker himself. The utterance can be easily repeated and replaces conveniently an evanescent or remote stimulus" (Bloomfield, 1939, p. 17). A large part of the self-stimulation called thinking (Bloomfield, 1933/1961, p. 28) appears when, as the result of correction by others, the child who used to speak aloud starts to speak inaudibly:

   [T]he movements of speech ... are replaced by internal movements, at
   first presumably as mere reductions of the normal movements of
   speech, but capable, in the course of time, of any degree of
   substitution. This inner speech accounts for the main body of the
   vaguely bounded system of actions that in everyday parlance goes by
   the name of "thinking." (Bloomfield, 1939, p. 17)

Bloomfield uses the process of counting to illustrate the advantages of thinking (1933/1961, pp. 28-29). We have two ways of knowing whether two sets have the same number of objects. One is nonlinguistic and involves placing each of the objects in one set next to each of the objects in the other set. If, when all objects from one of the sets have been place, there is no object from the other set that is not paired, then we say that the sets have the same number. It is not always possible to do this and, even when it is possible, it requires much time and effort.

The second way of knowing whether two sets have the same number of objects is to count them, and this depends on language. Bloomfield's analysis of counting is simple and elegant:

   The numerals one, two, three, four, and so on, are simply a series
   of words which we have learned to say in a fixed order, as
   substitutes [for the pairing of objects]. Using them, we can
   "count" any set of objects by placing them into one-to-one
   correspondence (as mathematicians call it) with the number-words,
   saying one for one of the objects, two for another, three for the
   next, and so on, taking care to use each object only once, until
   the objects of the set are exhausted. ... we can decide whether any
   set of objects has the same number as this first set, by merely
   repeating the counting process with the new set. Mathematics, the
   ideal use of language, consists merely of elaborations of this
   process. (1933/1961, p. 29)

Other forms of substitute signals for environmental stimuli are constructed from the system of speech. This is the case with gestures as well as marking, drawing, and writing--reactions whose value lies in the fact that they leave a permanent mark that survives as a stimulus over a long period of time and can also be carried as a stimulus to distant people (p. 40). Writing produces traces that follow or substitute for spoken words, and this made language still more effective:

   A written record can be carried to any place and preserved for any
   length of time. We can see more things at one time than we can
   hear, and we can deal better with visible things: charts, diagrams,
   written calculations, and similar devices, enable us to deal with
   very complex matters. The speech-stimuli of distant people, and
   especially of persons in the past, are available to us through
   writing. This makes possible an accumulation of knowledge.
   (Bloomfield, 1933/1961, p. 40)

Skinner's concept of the functions of verbal behavior

In the expression functions of verbal behavior, function means a biological or social role--not a mathematical function, as when we say that behavior is a function of environmental variables.

Verbal behavior has two types of functions--those related to the selection of operant units in the individual's repertoire and those related to the selection of cultural practices that are advantageous for the group. In regard to the first type of function, we can point out the general advantage of verbal behavior for speakers and listeners; moreover, for each type of operant, we need to identify what strengthens and maintains the speaker's verbal behavior and what strengthens and maintains behavior on the part of the listener that reinforces the speaker's verbal behavior. In regard to the second type of function, we need to point out the advantage of verbal behavior, in general, and of each type of operant, in particular, for the group.

Skinner's more general statement concerning the advantages of verbal behavior is relevant to both types of functions. Verbal behavior enhances the capacity for action in the environment by both the speaker (who becomes able to act on it verbally, also, and not just non-verbally) and the listener (who becomes able to react to verbal stimuli, also, and not just to non-verbal stimuli):

   Verbal behavior extends both the sensory powers of the listener,
   who can now respond to the behavior of others rather than directly
   to things and events, and the power of action of the speaker, who
   can now speak rather than do. If, as a result of a division of
   labor, the wise-but-weak can control the uninformed-but-strong,
   their combined accomplishments may exceed anything possible for
   either alone. (Skinner, 1957, p. 432)

As Skinner observes, we can only speculate as to the conditions in which the first verbal environment originated (pp. 461ff), since we have no record of the earliest forms of verbal behavior nor of the conditions that produced them. The enhancement of speaker and listener capacities raises the possibility that verbal behavior was advantageous for the group when it was involved in tasks that demanded the coordinated action of several people such as, for example, the differentiated functions in tasks as hunting, fishing, construction of shelters, war, etc. (p. 432). Beyond this early function related to group coordination, over the course of time, many other functions related to each of the verbal operants must have appeared (p. 433).

The first function of verbal behavior is seen in the practical action that the listener presents in response to the speaker's verbal behavior, and a prerequisite of this action is the maintenance of relatively constant relations between the physical and social environment and the speaker's verbal behavior:

   The ultimate explanation of any kind of verbal behavior depends
   upon the action which the listener takes with respect to it.
   Effective action requires a verbal stimulus which is "intelligible"
   in the sense of loud and clear and which stands in a reasonably
   stable relation to the conditions under which it is emitted.
   (Skinner, 1957, p. 314)

The listener's action strengthens the speaker's verbal behavior through generalized reinforcement, which is used primarily for installing verbal operants in the speaker's repertoire, and through a specific practical action related to the speaker's verbal behavior:

   Verbal behavior would be pointless if the listener did nothing more
   than reinforce the speaker for emitting it. The verbal community
   maintains the behavior of the speaker with generalized
   reinforcement, but a given listener often takes specific action
   with respect to what is said. If the listener's behavior is
   reinforcing for the speaker in his current state of deprivation or
   aversive stimulation, the speaker's behavior will be affected.
   (Skinner, 1957, p.151)

The presentation of generalized reinforcement and the specific practical action are performed through verbal and non-verbal discriminated operants under control of verbal discriminative stimuli produced by the speaker's verbal behavior. An example of the verbal discriminated operant could be the listener's response of "Flock" to the speaker's verbal behavior "What is a collection of sheep called?" The non-verbal one could be exemplified by the listener's response of picking up a book from a bookcase after hearing "Could you please bring me this dictionary?" Both types of responses, verbal and nonverbal, under control of discriminative verbal stimuli illustrate the presence of behavior under the control of instructions in the listener's repertoire--an important parcel of human behavior (pp. 358ff).

The speaker's verbal behavior can also elicit listener behavior that can reinforce the speaker's verbal behavior--for example, a speaker might make a listener very angry by calling him "Swindler!," or a speaker might predispose listeners to behaving favorably by telling them a joke and making them laugh (pp. 154-158).

Even when speakers are their own listeners, the practical effects in the environment still is what maintains verbal behavior--in this case, practical effects on subsequent behavior of the speakers themselves. Speakers' verbal behavior in functioning as their own listeners constitutes a large measure of what is usually called thinking and is functionally related to the rest of the speaker's behavior (p. 433ff). Skinner offers a peculiar definition of thinking and indicates its function when he identifies "thinking with behaving which automatically affects the behaver and is reinforcing because it does so. This can be either covert or overt" (p. 438).

The speakers' own verbal behavior affects them by supplying verbal stimuli that control echoic, textual, and intraverbal operants, and these, in turn, can control other types of the speakers' verbal or nonverbal behavior (p. 439). Verbal behavior of speakers who function as their own listeners has practical consequences, as would be the case with an auto-mand--Let's go, it is time to work; with repetition of an instruction for themselves--Now that the dough is ready, it must be left to rest for twenty minutes; with a tact, as in classifications--This is an FR5 schedule of reinforcement; and in problem-solving (p. 442), which depends on the speakers themselves generating stimuli that, when added to those already presented by the situation that constitutes the problem, produce behavior that is effective in the situation (pp. 440446). The main characteristics of verbal behavior when speakers are their own listeners, including its reinforcing effects, do not differ from those found in analogous behaviors where speakers and listeners are different organisms (pp. 445-446).

Each verbal operant has a specific function. The mand and the tact are the verbal operants that most clearly function to get the listener to perform a practical action in the world. The mand, in general, specifies its reinforcer and the action that the listener must manifest (p. 36).

Skinner's statement that the tact is "the most important of verbal operants because of the unique control exerted by the previous stimulus" (p. 83) must be understood vis a vis his concept that verbal behavior functions, primarily, in benefit of the listener (p. 36). In the tact, the form of the response is under a stable relation of control by non-verbal discriminative stimuli that is very important for the functioning of the group (pp. 81ff). The tact "works for the benefit of the listener by extending his contact with the environment, and such behavior is set up in the verbal community for this reason" (Skinner, 1957, p. 85). This contact includes very specific aspects of the environment, as is the case with tacts where the response is under the control of a single property of nature, as in abstraction (pp. 107ff, 448), or with those tacts responsible for self-knowledge that are installed through the discrimination of private events (pp. 130ff).

Echoic behavior is very useful for adults in establishing, especially but not uniquely in children, a repertoire of verbal units that will serve for the acquisition of new units (p. 56). An echoic repertoire is also useful to speakers in controlling parcels of their behavior, as mentioned above.

Textual behavior is reinforced, as well, because it facilitates the acquisition of other verbal operants (p. 67). Moreover, it frees the verbal stimuli generated by the speakers' behavior from their personal presence. Once the graphical signal is produced, it controls the reader's textual behavior independent of the writer's presence. The invention of writing and the building of a textual repertoire allowed the accumulation of human knowledge:

   The "speaker" who leaves an enduring record of his behavior can
   affect "listeners" in distant places and times, and these in turn
   profit from the special points of vantage of the remote "speaker."
   The achievement of the transmission of verbal behavior is seen
   today in codes of law, books of wisdom, formularies, and religious
   writings, which amplify almost without limit the effects of the
   behavior which originally produced them, and in histories,
   biographies, diaries, and experimental reports, which give the
   reader an almost unlimited contact with the environments of other
   men. (Skinner, 1957, p. 433)

Textual behavior also allows speakers control of their own future behavior when responding to discriminative stimuli generated by their own writing; it has the additional advantage over auto-echoic behavior of leaving permanent marks (p. 69).

Intraverbal behavior allows the acquisition of knowledge that is constituted basically of connections between verbal stimuli; these connections parallel relations between stimuli and properties of stimuli in the physical and cultural environment such as, for example, historical facts, definitions, tables of elements, etc. (pp. 362, 421). These connections between verbal stimuli could be brought up to date later in tacts and mands (pp. 421ff).

The combination of intraverbal operants in the speaker's repertoire generates new verbal behavior, as in the case of counting. Even when the stimuli are quite abundant or geographically distant, counting is made possible by means of the intraverbal connection of the sequence of numbers. Skinner explains how speakers are able to emit the response four upon counting how many people are in ten rooms:

   A familiar example of constructing a verbal response is counting.
   When a speaker says four in response to four men seated about a
   table, his response may be as directly controlled by a property of
   the situation as men or sitting. But if he says four after checking
   a dozen rooms, some of which contain men, his response is not a
   simple tact. It is the result of a special use of the intraverbal
   linkage, one, two, three, four, where (because he has "learned to
   count") he has emitted one response in the sequence whenever he has
   seen a man and now reports the last response so made. Such a
   response as one hundred is always constructed-either in this way,
   if it is controlled by counting one hundred objects, or by other
   operations. Mathematics is largely concerned with verbal behavior
   constructed by counting or by derivative processes. (Skinner, 1957,
   p. 423)

Skinner is pointing to two possibilities of control of the verbal response four. The first is a property of the non-verbal stimulus "4 men sitting," in which case the response is a tact. The other includes a chain of verbal stimuli, which makes the response an intraverbal operant: a sequence of verbal response--the intraverbal chain one, two, three ...--parallels a sequence of non-verbal stimuli (each man sitting).

Autoclitic behavior is also directed mainly toward the practical consequences of verbal behavior--generally by preventing the listener from taking improper practical actions, as the listener might do under control of verbal stimuli that result from the speaker's behavior if the listener does not have access to the kind of variables that control parcels of the speaker's verbal behavior. Thus, for example, when the listener hears The President promised that the budget for public education will increase, the way he is influenced by the information regarding more financial support for public education will depend on the degree of confidence the listener has in the President who was mentioned in the autoclitic emitted by the speaker.

   The ultimate explanation of autoclitic behavior lies in the effect
   it has upon the listener--including the speaker himself. In general
   the reactions of the listener at issue are those which can be
   wrong--that is, which may be ineffective in dealing with the
   environment responsible for the speaker's behavior. ...

   Autoclitic behavior is concerned with practical action or with
   responses on the part of the listener which depend upon a
   correspondence between verbal behavior and a stimulating state of
   affairs. (Skinner, 1957, p. 344)

The concept of the functions of language/verbal behavior held by these two authors is similar in several important respects, of which the following deserve to be highlighted:

* The ultimate value of language/verbal behavior does not reside in the behavior itself but in the effects that it makes possible in the practical world. Language/verbal behavior is, therefore, just a way to achieve effects in the non-verbal environment;

* Language/verbal behavior amplifies a practical stimulus's power of action (i.e., speakers react efficiently to practical stimuli through a linguistic reaction, and listeners react to practical stimuli that affected them only indirectly through the speakers' linguistic reaction to the practical stimuli);

* Speech has an effect on listeners, also, by changing their predisposition toward speakers rather than by directly provoking an immediate practical action;

* Through language people share their differentiated strengths and abilities and coordinate their actions, making the division of labor and the functioning of society possible;

* Language/verbal behavior constitutes a very large part of thinking, and thinking allows speakers to control other parcels of their own behavior;

* Counting is basically a process--or a derivation of one--by which the speaker emits one speech-form/verbal response that is part of a conventional series of speech-forms/verbal responses in the presence of stimuli and in a one-to-one correspondence;

* Mathematics is a form of language/verbal behavior derived mainly from counting;

* Language/verbal behavior is the origin of products such as maps, diagrams, written calculations, tables of elements, etc.;

* Writing extends verbal stimuli so that they can act upon readers situated historically and geographically distant from the writer.

On functions of language/verbal behavior, a few issues are found in Skinner but not in Bloomfield because they pertain to the analysis of the individual's behavior, which properly belongs to the realm of the behavior analyst, not the linguist. Some of these issues are: specific functions of each type of verbal operant with respect to speakers and listeners; specification of the function of generalized reinforcement, as well as reinforcements that are the practical effects of verbal behavior; and behavioral processes whereby the verbal community is able to install verbal operants and reinforcing behavior on the part of listeners. The behavioral model used by Skinner differentiates types of effect of verbal behavior on the listener as well as the discriminated verbal and non-verbal operants that make behavior under control of instructions possible. By indicating how each type of verbal operant contributes to achieving practical effects in the environment, and by specifying the variables that control each of these types, Skinner offers us instruments with which to build speaker and listener repertoires.

Bloomfield's concept of language in scientific activity

Bloomfield studied how language was used in the growth of scientific knowledge and how it is related to the non-linguistic part of scientific activity. In 1935, he published an article entitled Linguistic Aspects of Science in the journal Philosophy of Science, and in 1939, he published a monograph with the same title as part of the International Encyclopedia of Unified Science, edited by Rudolf Carnap, Charles W. Morris, and Otto Neurath, among others. According to Tomalin (2004), the "encyclopedia was a forum for assessing the methodology of scientific research, and many of the contributors were associated with the type of logical empiricism broadly espoused by members of the Vienna Circle" (pp. 127-128). In these works, Bloomfield elucidated the linguistic foundations of science and analyzed the forms of language used in logical and scientific activities.

Bloomfield (1933/1961, pp. 3-4) lamented the fact that people outside the field of linguistics have no knowledge of it. Even scientists and philosophers are often unaware of the extent to which their knowledge is founded on language: "The endless confusion of what is written about the foundations of science or of mathematics is due very largely to the authors' lack of linguistic information" (Bloomfield, 1935/1970, p. 308).

Bloomfield was a restless opponent of mentalism (see, e.g., 1930/1970; 1933/1961, pp. vii-viii, 17, 38; 1935/1970; 1939; 1944/1970), and he thought that people assume the existence of mental entities when actually it is speech-forms that are being used but are simply not recognized as such. He noticed that, in spite of the fact that mentalistic approaches did not appear in the procedures of physics, biology, and linguistics, mentalistic terms were still used in theoretical explanations in these fields (1939). The substantial portion of human behavior that was still unexplained by science would be amenable to scientific treatment if speech-forms were substituted for the mentalistic terms erroneously used in their place (1930/1970, 1936/1970). In Bloomfield's perspective,

   science shall deal only with events that are accessible in their
   time and place to any and all observers (strict behaviorism) or
   only with events that are placed in co-ordinates of time and space
   (mechanism), or ... shall employ only such initial statements and
   predictions as lead to definite handling operations
   (operationalism), or only terms such as are derivable by rigid
   definition from a set of everyday terms concerning physical
   happenings (physicalism). (1939, p. 13).

For Bloomfield, the above requirements did not exclude any part of the world from the scope of science because, although of different origins, these requirements have the same meaning relative to scientific methods, and "this delimitation does not restrict the subject matter of science but rather characterizes its method" (1939, p. 13).

In fulfillment of its above-mentioned primary function--to allow successful action--language frequently appears in the form of a report, which describes an expected succession of stimuli. Reports "imply that if you make such and such movements you will incur such and such a stimulus. We respond to them as we do to other speeches, in various ways, and often with error and confusion" (Bloomfield, 1935/1970, p. 310). The report does not convey all elements of the situation--these present themselves in our everyday life as a continuum of stimuli. Language is abstract because it consists of a set of discrete forms, while the practical situations which language treats are continuous (1939).

The hearer's successful action depends on the accuracy of the report which, in turn, depends on uniformity of meanings. As we have seen, in general, the meanings of speech-forms are not so uniform, and accuracy will require use of those speech-forms to which people react in uniform ways. The best example of such speech-forms would be the number-words (1935/1970).

Through writing, "permanent visible marks are produced conventionally as responses to speech and serve as stimuli for the production of speech" (1939, p. 9). However, in scientific activity, writing developed specialized forms that do not correspond to speech. This aspect of scientific activity can be carried out through written forms alone. It is true that this specialization is based on language conventions, also, because the written symbols are conventionally established as substitutes for linguistic forms (1939).

Because it allows stimuli to be viewed simultaneously in a single space while speech sounds appear in succession in time, and because the stimuli it produces are permanent, writing leads to special achievements. The advantages of writing in scientific activity are obvious, for the structure of scientific utterances "is so complex that a visual record, for simultaneous survey and back-reference, is indispensable ..." (1935/1970, p. 308).

Science magnifies characteristics of activities of daily life that guarantee their effectiveness, and it is based primarily (1939) on systematic observation and a specialized use of language: "The linguist naturally divides scientific activity into two phases: the scientist performs "handling" actions (observation, collecting of specimens, experiment) and utters speech (report, classification, hypothesis, prediction) ..." (6) (1935/1970, p. 307).

Scientists are trained to emit a very specialized vocabulary. Their reports are concise; they would be much longer if stated in everyday language (1935/1970). The most important objective is to guarantee accurate, uniform, and predictable responses from the hearers (1935/1970, 1939). Scientists' private events are not included in their reports: they are part of the subject matter of certain sciences, but they are excluded from scientific procedures. Science is objective, it is a public activity in both its handling and linguistic phases (1939).

The accuracy and uniformity of response makes science very powerful, and the test of its success is the ability to predict and to create (1935/1970). According to Bloomfield, in mathematics we find uniformity and accuracy guaranteed by speech-forms and combinations of speech-forms at the best. Mathematics is a specialized use of language and finds its highest application in science (1935/1970, 1939).

As Hockett (1970) noted, Bloomfield's concept of the linguistic nature of mathematics was probably formed very early in his career, since it had already been mentioned by Weiss in 1925. Tomalin (2004) showed that the debate on the foundation of mathematics involving the leading mathematicians and logicians of the early 20th century was an important part of the background of both Bloomfield's 1935 paper and his 1939 monograph. This debate was stirred by some paradoxes of set theory and gave rise to alternative solutions offered by mathematicians and logician-philosophers. Among the protagonists in this debate, as listed by Tomalin, were Russell and Whitehead, as well as logicians of the Vienna Circle such as Carnap and Neurath. According to Tomalin, Bloomfield was aware of a significant part of the key literature on this debate, and he intended his analysis of the linguistic foundation of mathematics to contribute to resolving it.

According to Bloomfield, the specialization of language in science poses requirements in relation to specific parts of the language structure. In his 1939 monograph, he characterizes the more general features of the phonemic, grammatical (morphological and syntactical), and lexical structure of language, and relates them to the practical events that constitute meaning. He stresses that ambiguity and inaccuracy must be reduced or eliminated not only in everyday language but, even more importantly, in scientific activity to increase the possibility of uniform responses. This is accomplished through the use of technical words as well as by syntactic and stylistic restrictions in informal scientific discourse; in formal scientific discourse, as in mathematics and logic, generally this is achieved through written notation, a sharply limited vocabulary and syntax, with transitions from sentence to sentence also sharply defined by rules (1939).

Skinner's concept of verbal behavior and scientific activity

The practices of logical and scientific communities optimize the principle function of verbal behavior--namely, the attainment of successful action: "In the history of logic and science we can trace the development of a verbal community especially concerned with verbal behavior which contributes to successful action" (Skinner, 1957, p. 418).

Scientists' behavior is partly verbal, including constructed verbal behavior, and partly non-verbal. For example, when we use instruments (such a microscope) to magnify and clarify stimuli, part of our response to these stimuli is verbal and part is non-verbal. Both parts, verbal and non-verbal, are committed to practical actions: "Empirical science is ... a set of practices which are productive of useful behavior" (p. 428).

Generally speaking, the consequence that ultimately maintains verbal behavior is the attainment of practical effects in the world. The emphasis on this as a distinct objective of the scientific verbal community means that, in the case of scientific behavior, this is really the exclusive effect with which this type of verbal behavior is concerned. It is through this effect that Skinner distinguishes the scientist's verbal behavior from that of writers, poets, and laypersons (p. 429).

Successful action does not necessarily mean applied technology. Scientific testing entails not only verbal confirmation but, eventually, effective action--that is, prediction and control of nature:

   [T]he behavior of both logician and scientist leads at last to
   effective nonverbal action, and it is here that we must find the
   ultimate reinforcing contingencies which maintain the logical and
   scientific verbal community. (Skinner, 1957, p. 429).

The scientific community establishes consequences in order to prevent effects unrelated to practical actions of verbal behavior from being reinforcing. Effects such as we find in fiction--exaggeration, understatement, and lying--can be reinforced by other verbal communities but are not reinforced and are even punished by the scientific community (p. 420). One noteworthy characteristic of the scientific verbal community is seen in its effort to suppress the speaker's verbal behavior under control of deprivation and aversive stimuli:

   Verbal behavior in which the reinforcement is thoroughly
   generalized, and the control of which therefore rests almost
   exclusively with the environment, is developed by the methods of
   science. The reinforcing practices of the scientific community
   thoroughly suppress the special interests of the speaker. This is
   not necessarily a sign of superior ethics in scientists; it is
   merely an evolved practice which has proved to be particularly
   valuable. It is responsible for much of the power of the scientific
   method. (Skinner, 1957, pp. 83-84).

Scientific and logical community sharpens the stimulus control of non-verbal and verbal stimuli over verbal behavior. In order to make practical action more effective, it is important to guarantee that verbal response is under the control of relevant properties of non-verbal stimuli. Devices such as classificatory schemes, technical vocabulary, graphs, models, tables, etc., are used with the aim of manipulating relevant and irrelevant properties of stimuli in order to strengthen behavior under control of the relevant properties (p. 419).

An assertion includes autoclitics indicative of the nature of stimulus control, such as I observe (which describes speaker behavior); It is probable (which qualifies the rest of the speech); and any, all (describing the range of application of the response). Verbal behavior under this precise stimulus control will serve as effective discriminative stimulus and instruction for the listener (p. 420).

The logical and scientific community also establishes contingencies whose aim is to guarantee accuracy of the control of verbal responses made to verbal stimuli, as in echoic, textual, and intraverbal behavior. These contingencies are designed to assure that the verbal response to verbal stimuli is indicative of the relation between the verbal response to verbal stimuli and the non-verbal circumstance where these last verbal stimuli were generated by a previous speaker. Scientists and logicians are required to specify the sources of control of their verbal behavior whenever this control could be ambiguous for the listener (pp. 420-422).

Symbols (generally replacing lay vocabulary) and special vocabulary serve to eliminate unwanted echoic, textual, and intraverbal responses. Definitions, memorized facts, and tables of constants are intraverbal behavior whose acquisition is part of scientists' special training (pp. 421-422). Rules of logical and mathematical thinking, Laws of Thought, and forms of syllogism are also intraverbal sequences. Autoclitics indicate the nature of the control of the scientist's verbal behavior--for example, whether by authorities, constructed through other verbal stimuli, etc. (pp. 420-422).

There are also contingencies that establish the conditions for constructing new verbal behavior which will most likely be effective, as we saw in counting (p. 10). Verbal responses, spoken or written, can be manipulated and can be treated as objects. This is usually done by logicians and scientists through a list of written words and rules which specify the ways for constructing new responses. The substitution of terms, such as entering values in an equation, transposing, etc., are examples of manipulation of verbal responses. Semantics and deductive logic often deal with usage rules for substitution of terms (pp. 422-425).

The responses constructed through scientific activity require confirmation as an intraverbal or a tact, and this is done by generating additional stimuli. For example, manipulation of established verbal responses can generate a statement about the position and size of a presumed planet; subsequent observation, probably with adequate instruments, can create the opportunity for observation of the planet as a tact. The confirmation can also be a type of intraverbal operant, as is the case with the theory of evolution--confirmation through the accumulation of verbal responses from the fields of geology, genetics, paleontology, etc. (pp. 426-428).

According to Skinner, logical and scientific methodology had been analyzing logical and scientific verbal behavior and the practices of the community that maintains it; these analyses and their products are also verbal behavior, and one "of the ultimate accomplishments of a science of verbal behavior may be an empirical logic, or a descriptive and analytical scientific epistemology, the terms and practices of which will be adapted to human behavior as a subject matter" (1957, p. 431).

Bloomfield and Skinner's concept of logical and scientific activity are very similar. They both emphasize that:

* The principle function of logical and scientific activity is a pronounced enhancement of the more general function of language--guaranteeing successful action;

* Scientific activity involves verbal and non-verbal behavior;

* Scientific testing is based on the extent to which the knowledge produced gave rise to effective action--that is, to the possibility of prediction and creation/control of the object;

* Events relevant only to logicians and scientists as individuals are banned from logic and science procedures, but these events are the subject matter of sciences;

* Classification, technical vocabularies, tables of constants, and symbols are language-based devices that aim to enhance accuracy and precision of logical and scientific activities;

* Writing, by virtue of leaving permanent marks simultaneously presented in space, allows some scientific practices that would be impossible through spoken language;

* Their own discipline is very important in the study of logical and scientific activities.

The difference we found on this topic between the works of these authors arises from the difference between their respective subject matters. Bloomfield unveils how language is ingrained in logical and scientific activity, describes the general forms of language as well as the specific forms that appear in logical and scientific activity, and how these forms contribute to achieving the specific objectives of these activities.

Skinner characterizes the verbal operants that appear in logical and scientific activities, also, because they contribute to achieving the primary objective of these activities. In addition, he describes the behavioral processes, such as control of stimuli, generalized reinforcement, etc., by which the logical and scientific community is able to build the individual's logical and scientific repertoires.

Conclusion

The similarities in Bloomfield and Skinner's approach to the topics addressed in this paper are so evident and related to such key aspects of these topics that their approaches can be considered clearly--and happily, since they are valuable approaches--compatible. They are also--again, happily--complementary, for Bloomfield describes the language structures involved in practices of the verbal community, which installs the speaker and listener repertoires through behavioral processes described by Skinner.

We must consider the possibility that Skinner was directly influenced by Bloomfield with regard to the topics addressed. This influence can be more comfortably asserted with regard to the topics of verbal community and functions of verbal behavior, since Bloomfield's teachings on these topics appear in Language, which Skinner referred to a few times.

In relation to the last topic, verbal behavior and scientific activity, the similarities in their approaches and the singularity of Bloomfield's teaching about the linguistic roots of logical and scientific activities also invite us to suggest a direct influence. Skinner's autobiography offers some information on this issue.

It is well known that Skinner began to write his book on verbal behavior in 1934 (1979, p. 160, pp. 456-457), and he mentions writing it several times in his autobiography (see, e.g., pp. 150, 281, 294, 324, 342). In one passage, he refers to Bloomfield in connection with the topic of verbal behavior and scientific activity. He mentions the general approach and subject matter of the book, some topics that would be included, and the relative ease of studying everyday and literary behavior. However,

   [l]ogic and the verbal behavior of scientists would be more
   difficult. Russell, Bridgman, Carnap, Reichenbach--I had discussed
   all of them with Cuthbert Daniel, Ivor Richards, Quine, and Feigl,
   but their position had not really coalesced. Bloomfield had been
   aware of the relevance of logical positivism. In 1930 he had
   written that linguistics had not yet reached the stage at which
   science can "win through the understanding and control of human
   conduct," but in 1936 he noted that "the logicians of the Vienna
   Circle have independently reached the conclusion of physicialism:
   any scientifically meaningful statement reports a movement in space
   and time. This confirms the conclusion of A.P. Weiss and other
   American workers: The universe of science is a physical universe.
   This conclusion implies that statements about ideas are to be
   translated into statements about speech-forms. (Skinner, 1979, pp.
   281-282).

Skinner is quoting Bloomfield's "Linguistics as a Science" (1930/1970) and "Language or Ideas?" (1936/1970), published in the journals Studies in Philology and Language, respectively.

To my knowledge, Skinner did not refer to the 1935 article or the 1939 monograph where Bloomfield studied specifically the language of logicians, mathematicians, and scientists. In 1932, however, Skinner wrote a plan for part of his life that included work on related issues such as behavioristic methodology, operational definitions of psychological concepts, theories of scientific knowledge, a theory of meaning, etc. (Skinner, 1979, p. 115). At that time, he started to work on A Sketch for an Epistemology, and a year or two later he subscribed to (p. 115) Philosophy of Science, which began to be published in 1934, and where Bloomfield's article Linguistic Aspects of Science appeared in 1935. It is very possible that Skinner had read it.

The International Encyclopedia of Unified Science where Bloomfield published his monograph Linguistic Aspects of Science (1939) was edited by logical positivists; Skinner was well acquainted with and interested in their work, as we can see by his comment in the last quotation above. Skinner could have had knowledge of this monograph.

We are very much in need of clarification of the connections between Skinner's behaviorism and Bloomfield's linguistics. This would improve considerably our understanding of Verbal Behavior. Willard Day's wise words are entirely up to date:

   [T]he thing to do with the book is largely to simply try to
   understand it well, as an entity in itself. As with understanding
   any other behavior, this means to me an effort to bring one's
   behavior (in this case as reader) increasingly under the control of
   the same sorts of variables which functioned in the control of the
   actual emission of the behavior under investigation. (1980, p. 165)

References

Bloomfield, L. (1970). A set of postulates for the science of language. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 128-138). Bloomington: Indiana University Press. (Reprinted from Language, 2, 153-164, 1926)

Bloomfield, L. (1970). On recent work in general linguistics. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 173-190). Bloomington: Indiana University Press. (Reprinted from Modern Philology, 25, 211-230, 1927)

Bloomfield, L. (1970). Linguistics as a science. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 227-230). Bloomington: Indiana University Press. (Reprinted from Studies in Philology, 27, 553-557, 1930)

Bloomfield, L. (1970). Albert Paul Weiss. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 237-239). Bloomington: Indiana University Press. (Reprinted from Language, 7, 219-221, 1931)

Bloomfield, L. (1970). Lautgesetz und Analogie, by Eduard Hermann. [Review]. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 240-251). Bloomington: Indiana University Press. (Reprinted from Language, 8, 220-232, 1932)

Bloomfield, L. (1961). Language. New York: Holt, Rinehart & Winston. (Original work published 1933)

Bloomfield, L. (1970). Linguistic aspects of science. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 307-321). Bloomington: Indiana University Press. (Reprinted from Philosophy of Science, 2, 499-517, 1935)

Bloomfield, L. (1970). Language or ideas? In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 322-328). Bloomington: Indiana University Press. (Reprinted from Language, 12, 89-95, 1936).

Bloomfield, L. (1939). Linguistic Aspects of Science. International Encyclopedia of Unified Science, Vol. 1 (4). Chicago: University of Chicago Press.

Bloomfield, L. (1970). Philosophical Aspects of Language. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 396-399). Bloomington: Indiana University Press. (Reprinted from Studies in the history of culture: The disciplines of humanities, 173-177. Menasha, Wis: George Banta, 1942)

Bloomfield, L. (1970). Meaning. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 400-405). Bloomington: Indiana University Press. (Reprinted from Monatshefte fur Deutschen Unterricht, 35, 101-106, 1943).

Bloomfield, L. (1970). Secondary and tertiary responses to language. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 413-425). Bloomington: Indiana University Press. (Reprinted from Language, 20, 45-55, 1944).

Bloomfield, L. (1945). Letter to J. M. Cowan (1/15/45). In Cowan, J M. (1987). The whimsical Bloomfield (pp. 28-29). In R. A. Hall Jr. (Ed.), Leonard Bloomfield: Essays on his life and work (pp. 23-37). Amsterdam: John Benjamins.

Bolling, George M. (1970). Language, by L. Bloomfield. [Review]. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 277-278). Bloomington: Indiana University Press. (Reprinted from Language, 11, 251-252, 1935).

Coseriu, E. (1986). Lecciones de linguistica general. Madrid: Gredos.

Edgerton, F. (1970). Language, by L. Bloomfield. [Review]. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 258-260). Bloomington: Indiana University Press. (Reprinted from Journal of the American Oriental Society, 53, 295-297, 1933).

Fought, J. (1995). American Structuralism. In E. F. K. Koerner & R. E. Asher (Eds.), Concise History of the Language Sciences: From the Sumerians to the cognitivists (pp. 295-306). New York: Pergamon..

Fought, J. (1999a). Introduction. In J. Fought (Ed.), Leonard Bloomfield: Critical assessments of leading linguists: Vol. 1. Biographical sketches (pp. 1-21). London: Routledge.

Fought, John. (1999b). Leonard Bloomfield's linguistic legacy. Historiographia Linguistica, 26(3), 313-332.

Hall Jr., R. A. (1990). A life for language: A biographical memoir of Leonard Bloomfield. Amsterdam: John Benjamins.

Hockett, C. F. (1970). Introduction to The Language of Science [Fragments], by Leonard Bloomfield. In C. F. Hockett (Ed.), A Leonard Bloomfield Anthology (pp. 333-334). Bloomington: Indiana University Press.

Hockett, C. F. (1984). Foreword. In L. Bloomfield (1984), Language (pp. ix-xiv). Chicago: The University of Chicago Press. (Original work of L. Bloomfield published in 1933).

Hockett, C. F. (1999). Leonard Bloomfield: After fifty years. Historiographia Linguistica, 26(3), 295-311.

Howatt, T. (2002). Introduction. In K. Malmkjr (Ed.), The linguistics encyclopedia (2nd ed., pp. xxv-xli). New York: Routledge.

Joseph, J. E., Love, N., & Taylor, T. J. (2001). Landmarks in linguistic thought: Vol. 2. The western tradition in the twentieth century. London: Routledge.

Koerner, E. F. K. (2003). Remarks on the origins of morphophonemics in American structuralist linguistics. Language & Communication, 23, pp.1-43.

Kroesch, S. (1970). Language, by L. Bloomfield. [Review]. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 260-264). Bloomington: Indiana University Press. (Reprinted from Journal of English and Germanic Philology, 32, 594-597, 1933).

Lepschy, G. C. (1982). A survey of structural linguistics. London: Andre Deutsch.

Matos, M. A., & Passos, M. L. R. F. (2004, May). Emergent verbal behavior and analogy: Skinnerian and Linguistics' approaches. Paper presented at the 30th Annual Convention of the Association for Behavior Analysis.

Matos, M. A., & Passos, M. L. R. F. (2006). Linguistic sources of Skinner's Verbal Behavior. The Behavior Analyst, 29, 89-107.

Matthews, P. H. (1999). Bloomfield's morphology and its successors. In J. Fought (Ed.), Leonard Bloomfield: Critical assessments of leading linguists. Vol. 3: Phonology, morphology and languages (pp. 102-152). London: Routledge. (Reprinted from Transactions of the Philological Society, 90(2), 121-186, 1992).

Matthews, P. H. (2001). A short history of structural linguistics. Cambridge, UK: Cambridge University Press.

Passos, M. L. R. F., & Matos, M. A. (1998, November). The influence of Bloomfield's "Language" (1933) on B. F. Skinner's "Verbal Behavior" (1957). Paper presented at the fourth International Congress on Behaviorism and the Sciences of Behavior, Sevilha, Spain.

Passos, M. L. R. F., & Matos, M. A. (submitted). The influence of Bloomfield's linguistics on Skinner's "Verbal Behavior" (1957).

Robins, R. H. (1997). A short history of linguistics. (4th ed.). London: Longman.

Skinner, B. F. (1947). The William James Lectures. P. Cherpas, S. J. Fath, M. Picker, & M. L. Sundberg (Eds.). Western Michigan University: The Association of the Experimental Analysis of Behavior.

Skinner, B. F. (1957). Verbal Behavior. New York: Appleton-Century-Crofts.

Skinner, B. F. (1969). Contingencies of reinforcement: A theoretical analysis. Englewood Cliffs, NJ: Prentice-Hall.

Skinner, B. F. (1974). About behaviorism. New York: Knopf.

Skinner, B. F. (1979). The shaping of a behaviorist. New York: Knopf.

Sturtevant, E. H. (1970). Language, by L. Bloomfield. [Review]. In C. F. Hockett (Ed.), A Leonard Bloomfield anthology (pp. 265-266). Bloomington: Indiana University Press. (Reprinted from The Classical Weekly, 27, 159-160, 3/26/1934).

Tomalin, Marcus. (2004). Leonard Bloomfield: Linguistics and mathematics. Historiographia Linguistica, 31(1), 105-136.

Author Contact Information

Maria de Lourdes R. da F. Passos, Av. Pref.

Bento Goncalves Pereira, 583, Paraiba do Sul, Rio de Janeiro, Brazil, 25850-000

e-mail: [e-mail]

(2) I am very grateful to Raquel W. da Cunha for giving me this valuable information, and sending me a copy of The William James Lectures.

(3) In the front page of The William James Lectures there is a note informing that they are copies of the original transcripts containing errors that were not corrected.

(4) There is a minor mistake in the reference made by Skinner. In fact, in Bloomfield's act of speech, Jill asks Jack for an apple.

(5) The brackets added contain the words that I assume Skinner was abbreviating here.

(6) By handling actions, Bloomfield (1939) means all the observable activities which are not speech, that is, not just manipulation but also mimicry, locomotion, gesture, etc.

Maria de Lourdes R. da F. Passos (UFRJ) (1)

(1) I thank Lourenco Barba for the careful reading and valuable suggestions to the final draft of the manuscript.