Dani Hathcock, 2000

Introduction

Language acquisition has been an intensely studied area of many fields in recent years, with its origin dating back to ancient times. Psychologists, philosophers, and medical doctors are a few groups among many who have taken an interest in studying the systematic development of human speech. The 1950s were a revolutionary time for the field, with many different theories emerging to answer the great mystery, yet the true cause still remains an enigma. This paper will discuss two theories of language acquisition, one from Ludwig Wittgenstein, a late 19th Century Austrian philosopher, and the psychological model of the behaviorists as was developed by B.F. Skinner in the 1950s.

Thesis

The primacy of both Wittgenstein’s and Skinner’s theories is of language acquisition as it occurs within the social context. According to both, meaning is found in such a context, and language is then the association of the terms and the objects, with a large part of basic speech development resting on training. Language then, is no different than any other behavior—according to Wittgenstein it is learned through certain games ("language-games") that are like those that children play in other realms of life, while behaviorist theory contends essentially the same thing, with external reinforcement motivating speech production. However, neither model addresses the biological aspects of language acquisition, which serves to disregard a large amount of scientific evidence that finds biology to be of at least equal importance. Hence, the following discussion will compare Wittgenstein’s theory to the behaviorist model in order to illustrate that both fail to address the biological aspects of language acquisition and therefore do not fully explain the phenomenon.

Ludwig Wittgenstein’s Meaning as Use

Ludwig Wittgenstein was born in 1889 to wealthy Austrian family in Vienna. He studied engineering in Austria, then went to Manchester in 1911 to study the design of airplane propellers and engines. He then studied mathematics and logic under Bertrand Russell until serving in the Austrian Army in World War I, where he wrote his major piece, Tractatus, in which he (originally being an analytic philosopher) thought he solved all philosophical problems. He gradually began to question his own assertions in the piece, and then later went back to school and live in great simplicity, refusing the fortune left to him by his father, and refusing to publish anything else. He did, however, dictate notes to his students who published them under the title Philosophical Investigations in 1951 (Roger Whitson’s biographical information, class notes). It is in the second piece that Wittgenstein moves away from the idea of fixed, unchanging meanings to the idea that meaning is constantly in flux due to the fact that it is found within the social context, with all meaning thus being public. It is here that language acquisition is examined, with Wittgenstein’s argument resting on the latter premise of public meaning to conclude that his idea of "language-games" are actually the driving force behind the acquisition of language. In other words, Wittgenstein basis his ideas on the framework that language is acquired by children learning different "language games" in which they learn to associate objects with meaning, and that those meanings are common to all, rather than something internalized. Hence, children learn language as they learn any other game, with the words all gaining meaning from the culture around them. This section will take a step-by-step approach to Wittgenstein’s argument, working through his basic premise and assumptions, as well as the complexities involved, while discussing any flaws occur in the text.

Wittgenstein’s Philosophical Investigations (found in Andrea Nye’s Philosophy of Language: The Big Questions), begins with a quote from Augustine’s Confessions which commences his argument concerning how human beings acquire language. According to Augustine, by watching his elders speak of an object and then turn to it, he was able to memorize and thus reference each particular object. When he was then able to pronounciate, he had a working knowledge of the language. Wittgenstein states that "in this picture of language we find the roots of the following idea: Every word has a meaning. This meaning is correlated with the word. It is the object for which the word stands" (Nye, 86). However, he doesn’t find that this quasi-empiricist view encompasses the entire realm of language acquisition--he feels that Augustine is merely thinking of nouns and does not take into consideration other types of words, which may not be accounted for in his description. Thus, Wittgenstein gives many examples to illustrate his point that this is not the whole of language. His first example is of a man going to the shopkeeper for apples. The man walks in with a slip marked ‘five red apples,’ asks for the apples, and the shopkeeper acts upon the direction of the order. While the shopkeeper may have followed orders, it may not essentially have any meaning (Nye, 86). According to Wittgenstein,

How does he know where and how he is to look up the word ‘red’ and what he is to do with the word ‘five’?"—I assume he acts as I have described… But what is the meaning of the word "five?"—No such thing is in question here, only how the word five is used (86).

His second example consists of a two-person dialogue operating under Augustine’s system, in which the builder, A, asks for a slab/pillar/stone from assistant B, and each time B brings it, with this being the whole of the dialogue. Is Augustine correct? Is this communication, or the whole of language? Wittgenstein answers this with the statement, "Only for this narrowly circumscribed region, not for the whole of what you were claiming to describe" (86). In other words, Augustine’s description of referencing may work within the latter situation, but it does not describe the meanings of the words in the larger sense; in this way meaning is found in the social context.

Next he brings in basic language acquisition, claiming that, though it is not the whole of language, children do use primitive forms of language when first learning to speak, and that this "teaching of the language is not explanation, but training" (87). This is accounted for in his "ostensive teaching of words," which he claims, "creates an association between the word and the thing" (87). This association may be used to evoke an image in the child’s mind when he or she hears the word, however, while this may help to attain the end goal (e.g. Bring the slab), this may not be the actual purpose of such language. If the direction is acted upon, then it is not merely the ostensive teaching of words that brought this about, rather it is when used in conjunction with particular training that the act occurs. It is here that the teaching of language is training. In fact, the actual meaning of words in such cases, is how they are used in conjunction with other words—given alone they could have different or no meanings. This point is illustrated by the example "I set the brake by connecting up rod and lever," which is correct, given the "whole of the rest of the mechanism…Only in conjunction with that is it a brake-lever" (87).

Wittgenstein next attempts to expand the language and the process by which it is acquired, bringing in the idea of a "language game," which he defines as "the whole, consisting of language and the actions into which it is woven" (88). A language game is very similar to any other game that a child uses to learn his or her native language—one such language game might be of an adult naming objects and the child repeating them, though there are many others. He then introduces more words into the simple language of orders, such as numerals, ‘this,’ ‘there,’ alphabetical letters, and colors. These words are also taught ostensively, though not to the extent that the first basic orders are. Numerals, for example, are taught to name groups of things rather than to count, while the teaching of ‘this’ and ‘there’ points not only to the use of such words, but the use of the language as well. The language games are what give children common meanings of terms and allow for new terms to come into existence. It is within this that the public meaning is contained. It is not internalized-- it is something that is developed through such games.

His next main question is that of what language signifies:

The description ought to take the form of "The word…signifies…" which can be reduced to the word ‘slab’ signifies this object. Equally one can say that the signs "a", "b", "c" signify numbers when…this removes the mistaken idea that [such signs] play the part actually played in language by "block", "slab", "pillar" (88).

However, assimilating the words in such a way does not make their uses any more alike one another. In his example of a toolbox, the words themselves are as "diverse as the functions of the objects" (88). One problem with this that Wittgenstein fails to note is that for the latter form to work, the physical object must be present, otherwise a definition must be given as to what the object itself actually signifies, and that definition will have no meaning unless it is able to signify the words in the definition with some particular object. Thus, as he points out, when it is said that every word in language signifies something, nothing has been said, unless a specific distinction has been made (89). He illustrates this by pointing out that the assimilation made in the phrase "All tools serve to modify something" actually says nothing. For the words "to signify" to actually mean anything, they must be related to something that is "marked with a sign"—named or labeled (89).

His argument then proceeds to examine the grouping of different words together, and the subjectiveness of doing so. It is possible to say that there are different kinds of words, and that often it is the function of such words that are more alike than the actual terms. How they are classified will be due to "our own inclinations" (90). This again reveals the social context in which meanings are found—it is not a fixed process, but one that is constantly changing and molding to the culture around it. How words are grouped will stem from the social meanings that are put upon them; this in turn will determine their classification—it was only due to the inclinations of those using the words that their meaning was given and the groupings were formed.

His discussion then examines the dialogue in the beginning of his text between the builder and assistant in which the builder would call out "Slab!" and the assistant would bring it. When "Slab" is used rather than "Bring me a slab," Wittgenstein wonders how it is known that the latter is meant, especially when there are other possibilities, such as "Bring me two slabs." He contends that it is only because one has a knowledge of the language as well as the context and body gestures that are used-- the language games children play early in life develop the social meanings such that the working knowledge of the language allows for the shortening of phrases without the intention being lost, which a foreigner would not be able to decipher. In fact, Wittgenstein recognizes that it is actually the knowledge of these other possibilities that allows for the actual intent to be understood. However, he doesn’t find such a sentence to be "elliptical" because it lacks something that is being thought when it is said, but because "it is shortened—in comparison with a particular paradigm of our grammar" (91).

Lastly, Wittgenstein analyses 19th Century German philosopher Gottlob Frege’s idea that "every assertion contains an assumption, which is the thing that is asserted," and finds that when this is the case, it either takes on a form that is not found within the language-game, or that something has to be asserted, in which the phrase, ‘it is asserted’ becomes superfluous (92). He also notes that an assertion sign might be used instead of a question mark or other grammatical symbol; Frege’s assertion mark, he claims, marks the beginning of a sentence, but that it does not serve to tell him anything. For example, in the statement, "It’s raining," "I do not know whether I have heard the beginning or the end of the period," henceforth it does not relay all of the information that is supposed to be contained (92). He also notes that there are many kinds of sentences, as well as multiple types of symbols and words used, and that these are not fixed, but that, in keeping with the social context of meanings, "new types of languages, new language-games…come into existence, and other become obsolete and get forgotten" (92).

Wittgenstein’s argument of language acquisition rests on the premise that speaking is an activity, a form of life. Speech is acquired through games that are played as children, as associations are made between words and objects, in which the meanings of the words are found within the social context in which they lie, with all meaning therefore being public rather than private and internalized. Speech is also then a changing, modifying entity. His position is perhaps best clarified in the following metaphor:

Our language can be seen as an ancient city: a maze of little streets and squares, and of houses with additions from various periods; and this surrounded by a multitude of new boroughs with straight regular streets and uniform houses…And to imagine a language means to imagine a form of life (90).

The next main section concerns the psychological model of behaviorism, which is quite similar to Wittgenstein’s theory. Both the behaviorists and Wittgenstein view language as a skill to be acquired just as any other skill, with the meanings of the words found within the social context—meaning in both theories is something that is not fixed, but something that is changing as those who use the language are inclined to change it and assign different terms. Once the behaviorist model has been defined and laid out, the discussion will focus on the comparisons between the two, and then the critical flaw that both have failed to address.

The Behaviorist Model

The behaviorist theory is a psychological model that, in essence, defines human behavior as something that is conditioned or instilled. The model basically states that humans are products of their environments and all behaviors are positively reinforced, negatively reinforced, or punished. If a behavior is positively reinforced, or praised, then there is incentive to repeat it. If instead, it is negatively reinforced or punished, then the behavior will cease. One of the classic cases given as evidence of this theory is of Ivan Pavlov and his dogs. When Pavlov rang the bell, he would feed his dogs, and they would salivate. After repeating this a number of times, Pavlov noticed that every time he rang the bell, whether he fed the animals or not, they would salivate, proving to Pavlov that if there is a reward associated with a behavior, then eventually the behavior will continue even without the reward. The behaviorists have applied the same principle to human behavior. The Father of Behaviorism is usually considered to be John B. Watson, who has been rumored as boasting, "Give me a child, and I’ll make him anything you want him to be." This attitude has influenced many psychologists, with the more recent behaviorists still holding strong to the idea that any human being can be conditioned to behave in very specific manners. B.F. Skinner, the most recent and perhaps well-known advocate of this theory, contends that language is acquired the same way—it is no different than any other behavior a child is conditioned to do. Skinner’s work in the field of language acquisition was done in the late 1950s, a revolutionary time in the field as many theories concerned with language acquisition were born. This section on the psychological aspects of language acquisition with specific regard to behaviorism will begin with an overview of the historical trends of language acquisition research, highlight the basic stages in child speech development, and then discuss the basic assumptions of the theory.

According to Jean Berko Gleason’s The Development of Language, interest in language development dates back to ancient times, though the study of systematic development in children is a somewhat young field. (Gleason, 24). The latter half of the 19th Century and the first part of the 20th revealed German and French studies, and in the first half of the 20th many psychologists kept diaries recording speech progress. However, it was not until the 1950s that the field was seriously studied, with many revolutionary theories emerging, such as Chompsky’s idea of an innate device in the brain that naturally produces language (LAD: language acquisition device), and Skinner’s behaviorist model. The 1960s were "characterized by studies of grammar" while the "primacy of syntax in research gave way to a broadening interest that included the context in which children’s language emerges and an emphasis on the kinds of semantic relations children are trying to express in their early utterances" in the 1970s and 1980s (25). While areas of study have changed in recent decades, the behaviorist model is still widely accepted in the psychological community. Next, the basic phases of speech development are examined, which lay the foundation for all of the theories—each theory provides a description of how and why these occur.

Gleason has determined that there are essentially five stages of basic child speech development (46-47). Stage I consists of reflexive crying and vegetative sounds, which occurs between zero and eight weeks. The infant has very little voice capability at this point, since the vocal cords are positioned relatively high and the tongue fills up most of the oral cavity. In Stage II, there is cooing and laughter, as well as some brief consonant sounds, such as ‘c’ and ‘k’. Crying is also less frequent between this period of eight to twenty weeks. The cooing in this phase is generally thought to be in response to external stimuli, which the behaviorists will claim is true for all phases. Between sixteen and thirty weeks the infant is in Stage III in which he or she engages in vocal play: "[uttering] single syllables with prolonged vowel- or consonant-like sounds…which is the transitional period between the earlier comfort-related cooing and later true babbling" (46). In Stage IV, between twenty-five and fifty weeks, the infant engages in reduplicated babbling, in which she has much more control over her articulatory structures and repeats the same vowel-consonant combinations a number of times. Interestingly enough, research does not have evidence that such babbling is meant to be a form of communication; to the contrary is often occurs more when the infant is alone. Behaviorists justify with the idea of self-reinforcement (or, inner reinforcement): after a child learns to speak, she recognizes what she says with familiarity, is pleased by it, and thus repeats it. Lastly, Stage V occurs, characterized by nonreduplicated babbling and expressive jargon, between nine and eighteen months. In this stage, "the baby’s babbling now becomes more complex: the consonant-vowel-syllables change within an utterance of babbled sounds…[and] the total repertoire of consonant sound increases" (47). While the study of language acquisition does reach into the areas after infancy, looking at the school years of a child and the acquisition of a more complex vocabulary as well as intentional voice fluctuation, the main concern of this discussion will be behaviorist theory as it relates to early language acquisition as was examined in the latter five stages, in which very basic speech is attained, with less attention given to the later years. The discussion will next move into the basic assumptions of the behaviorist model, and then compare it to Wittgenstein’s theory earlier discussed.

The basic assumption of the behaviorist model is that all behavior is conditioned. More specifically, when examining language acquisition, there are eight basic assumptions, as given by Diane Nelson Bryen in her book Inquiries into Child Language:

The organism is psychologically empty, and the environment is all important. Language is merely another learned, observable behavior. No distinction exists between speech and language. No real distinction exists between the comprehension and production of language. Meaning is the association of the word with its referent. No distinction exists between thought and language. To know is largely to be able to talk (Skinner, 1957). Learning language is largely contingent upon the processes of imitation, reinforcement, stimulus discrimination and generalization, and shaping. Mastery of one’s language is demonstrated through observable responses (i.e., speech production) (Bryen, 151). Each assumption will now be explained and discussed, which will be the examination of the behaviorist theory in the language acquisition context.

Assumption one contends that the organism is psychologically empty. This essentially means that human beings are not hard wired to behave in any specific manner—they enter the world as a blank slate, not predisposed toward any behavior. The environment in which one is raised determines what behavior one will adapt. It is not genetics or biology that determines action; rather, it is the environmental (external) stimuli. In other words, human beings are merely reacting to their surroundings—the surroundings are imposed upon them rather than them actively imposing upon their surroundings. This assumption carries over into all of behaviorist theory.

Assumption two finds that language is another learned behavior. Note that this implies that the behaviorists find all behaviors to be learned and observable. Behaviorists have been quite concerned with the lack of empirical validity in psychology, and therefore set out to quantify and measure behavior just as a chemist measures chemical reactions: essentially, cause and effect. Henceforth, behaviorists view all behavior as measurable—it is the effect of the environment. Language, then, is no different. It too, is learned through a series of differential reinforcements in which the act is either praised, punished, or negative influences stop when the behavior is performed. If an infant makes a sound, and the parents give positive reinforcement, such as vocal praise, hugs, or affection, then the child has incentive to repeat the sound. When a child utters a word, the same process occurs. One example from Gleason is how a child learns to say ‘good-bye.’ An adult holding the child waves the child’s hand and says (usually in a high-pitched voice), "Say bye-bye. Say bye-bye" (181). When the child finally says the words, she is showered with praise and affection, and thus continues the speech to gain the affection again. After this process has occurred a number of times, the child doesn’t need the reinforcement for the behavior to continue—the behavior has been associated with the praise enough that it will continue for some time without it. Accordingly, even pain can be relayed in such a way. When a child skins her knee and someone tells her that it is painful, she associates the term given with the pain in her leg and discriminates it from all other feelings. Situations such as the latter explain how all vocabulary develops (stimulus, reaction, discrimination of the feeling or behavior, repeated action). Also included in this is that language, like any other behavior, is subject to extinction. If the behavior is not positively reinforced, then eventually, it will cease. This logic explains how a child moves away from ‘baby talk.’ Eventually, it is no longer praised, and a child gains much more acceptance by speaking like an adult.

Assumption three consists of there being no difference between speech and language. In other words, every time someone speaks, that is language. There is no difference between the basic syntax and communication. Language, in no way, transcends speech. Language is merely the assigning of terms to objects and feelings, reinforced over time. Speech is an active participation in this—they are one in the same.

Assumption four contends that comprehension and the production of language are essentially the same as well. For example, if one says, "One red apple," then one does not have a comprehension of the words that is separate from saying them. The fact that the person said those words in conjunction with specific references is the comprehension—it is the language. There are no external, metaphysical definitions of one, red, or apple. The comprehension of those terms is found in the fact that a child is able to say them in accordance with the objects they have been assigned to. As Bryen states, "The comprehension of language is explained by Skinner quite simply as understanding what someone is saying when we ourselves say it for the same reason" (151). In other words, all meaning is common and public, all words are said simply because they are terms assigned to specific objects and feelings, thus each person means the same thing with the same word, and no distinction is then drawn between the comprehension and the production of language.

Assumption five, that meaning is the association between the word and the referent, or the object or feeling, is quite consistent with the latter description. In fact, assumption four rests on this premise, since this assumption is the framework that allows four to be possible. Meaning is merely assimilating a term with a given object, which allows for assumption four to be possible, as well as assumption six, which contends that there is no distinction between thought and language. Just as comprehension of language is the production of language, such is that thought is language. The thought is not a distinct and separate entity, since it too rests on the association described in five. Just as there is no metaphysical definition of ‘apple,’ which allows for the comprehension and speech not to differ, having no such definition also does not allow for thought and speech to differ—the thought is contained within the language, within the term itself. There is no separate thought of the actual meaning of the term, because that meaning is contained within the association, which is then the thought itself. Therefore, the knowing is within the talking, rather than talking being a product of speech.

Assumption seven is the basic premise of all behaviorist theory, which is somewhat a restatement of assumptions one and two. This assumption describes language acquisition as any other behavior-- one that is learned through imitation and reinforcement, a reaction to external stimuli and the discrimination of it from other objects and feelings. The last assumption, eight, then follows from this, with the mastery of language being demonstrated through observable responses. Otherwise stated, one can observe that language is mastered by observing the language being used in the way that it was originally reinforced.

These eight assumptions essentially contain the whole of the behaviorist model. However, while there is support for this model, there are many questions left unanswered and strong criticism about the points that it fails to cover. The following section will compare and critique the two theories, and then examine the critical flaw that neither address.

The Comparison of the Two Theories and Justification of the Thesis

Wittgenstein’s and the behaviorist’s theory are quite similar. The most obvious similarity is in the fact that they both place meaning within the social context, as was discussed earlier in each theory. The reason that meaning is found within the social context is that both claim that reinforcement (Wittgenstein uses language-games as reinforcement in comparison to the behaviorists idea of definitive reinforcement) is the primary agent in motivating the learning of speech. In other words, language is more training rather than explanation, just as in Wittgenstein’s "ostensive teaching of words." Neither one finds meaning to be individual and internalized, instead, both find that it lies in the context in which the words are used, which depends entirely on how one is using them. This point does not create a contradiction in Wittgenstein’s argument, but it does for the behaviorists. If the meaning of words is entirely dependent upon how they are used, and how they are used is dependent upon conditioning as all behaviors are, then how did those first words obtain their meaning in the very beginning? Who was first conditioned, and how, to assign objects to terms? Even if it was completely random, how then, did those words become spoken? The latter point is a flaw in Wittgenstein as well, which will later be discussed in more detail.

A second way in which they are alike is in humans being psychologically empty. Neither theory accounts for any predisposition toward speech, and Wittgenstein only vaguely hints at biology accounting for any type of language when he notes that body gestures may lead to better understanding between communicators. This second similarity also explains their shared idea of a social meaning, since the human organism can only find meaning through contextual understanding—it is tied in with the association of the object and its term. It then follows that the signification of language is found in the latter assimilation. Both theories need a specific distinction to be made, something named or labeled, in order for language to signify and relay information.

The third main similarity is that both theories also find meanings and terms to be changing, as new ones come into the language and older ones then become obsolete. Wittgenstein uses the metaphor of the city in which new houses and streets are built upon the pre-existing foundation, while behaviorist theory recognizes that if terms are not continually reinforced, then they will become extinct. Thus, terms are constantly in flux, and there are no metaphysical definitions which allow for any transcendent meaning or a set list of words whose definitions are unchanging and set.

The largest similarity between the two is also their greatest flaw as was discussed in the thesis—both theories fail to address the biological aspect of language development. One of the biggest problems with this is what was earlier examined—even if words acquire meaning through social context, how are those words actually spoken? Even beyond the biological organs necessary for speech production, what cognitive process allows for speech to occur? Obviously some neurological or physiological process is necessary for a hearer to be able to repeat the words given to her by the communicator, but neither theory takes this into account. In fact, in Augustine’s description, he glosses over this fact with the phrase, "When I was able to pronounciate, I had a working knowledge of the language" (Nye, 85). However, being able to actually vocalize the words is obviously a necessary ingredient for speech, and it is quite unexplained by the assumption that it simply occurs naturally. In fact, there are major pieces of evidence given in scientific study that are contrary to the ideas produced by Wittgenstein and behaviorist theory. One is that language acquisition occurs relatively the same way cross-culturally. In almost every part of the world regardless of culture, childhood phases of speech are extraordinarily similar. One must conclude from that that if the behaviorists are correct, then by coincidence (since human beings are not predisposed to behave in any certain manner) all children are conditioned in the same way. While different words may be reinforced in different cultures, the basic idea still holds that children are conditioned to speak. Even more interesting is that children who have been severely neglected have speech capacities and generally obtain language in the same time frame as children who have had quite a bit of parental influence and reinforcement. While the latter hold true for both theories, an even more specific problem occurs with behaviorism. While behaviorists measure behavior as cause and effect (stimulus, response), these findings only generally hold true in a laboratory setting—generalizing a whole population based upon laboratory findings within very set conditions lacks an incredible amount of empirical validity. Another problem with both is that the learning of language, in actuality, takes place in a short amount of time. However, the association that Wittgenstein argues for and the primacy of reinforcement that the behaviorists seek to validate both take quite a bit of time and practice, especially with infants. Therefore, given the above evidence, both Wittgenstein’s theory and the behaviorist model fail to address the biological aspects of language acquisition, which is thus a critical flaw in both and leave the above questions an unexplained mystery.

Concluding Remarks

While the two theories do not account for a large portion of language acquisition, their research is not to be entirely discounted. Studies that primarily focus on the biological aspects do not take into account the social, which is also a large and important part. One theory that does focus on the biological is Ruth Millikan’s theory of Biosemantics, which essentially adopts a "quasi-normative biological function as the basis for a theory of meaning that demystifies meaning and preserves realism" (Nye, 70). In this case, language has evolved just as every other animal behavior, and is simply a part of the evolutionary continuum. Certain psychological theories, such as Chompsy’s idea of the neurological language acquisition device that innately produces language, also focus on the biological basis of language. However, no theory has definitively proved how it is that human beings acquire language, and the field remains open to much research as philosophers and scientists work to unravel the mystery.

Related Links:
Back to the Interdisciplinary Research Conference homepage