24 Speech and Language

Speech and Language

Civilization began the first time an angry person cast a word instead of a rock.

Sigmund Freud

Observational learning has been evidenced in many species of animals including birds (Zentall, 2004) but approximations to speech appear practically unique to humans. Paul Revere famously ordered a lantern signal of “one if by land and two if by sea” during his Revolutionary War midnight ride through the streets of Massachusetts. This is not functionally different from the distinct alarm calls emitted by vervet monkeys in the presence of eagles, snakes, and leopards (Strushaker, 1967; Seyfarth and Cheney, 1980). Through observational learning, young vervets learn to respond to different screeches for “heads up”, “heads down”, and “look around!” Vervets hide under trees to the eagle warning, rear on their hind paws to the snake warning, and climb the nearest tree to the leopard warning. Recently, even more descriptive “speech” has been demonstrated in prairie dogs (Slobodchikoff, Perla, & Verdolin, 2009). These examples are the closest we see to social learning of speech in other animals. Slobodchikoff (2012) has written a fun and informative review of animal communication entitled Chasing Dr. Doolittle: Learning the Language of Animals.

Meltzoff and Moore (1977, 1983) demonstrated unambiguous examples of imitation in infant humans as young as 12- to 21-days of age, leading to the conclusion that humans normally do not need to be taught this mode of learning.

Skinner (1986) contributed an interesting but admittedly post-hoc speculative theoretical article describing possible evolutionary scenarios for the adaptive learning of imitation and speaking. An imitative prompt is more informative than an ordinary gestural prompt in that it specifies the specific characteristics of a desired response. Speech is preferable to signing as a means of communication since it is possible at long distances and other circumstances where individuals cannot see each other.

Hockett’s Features of Language

If we are to understand human behavior, we must understand how language is acquired and its impact upon subsequent adaptive learning. Before we proceed, we must consider what we mean by language. Charles Hockett (1960) listed 13 features that he considered essential to language:

1. Vocal-auditory channel – We saw in Chapter 1 that the human being’s brain, with its disproportional amount of space dedicated to the tongue, larynx, and voice box, facilitates the acquisition of speech. Sign language, involving a manual-visual channel, is mostly restricted to deaf people and those wishing to communicate with them.
2. Broadcast transmission and directional reception – Sound is sent out in all directions while being received in a single place. This provides an adaptive advantage in that people can communicate with others out of their line of sight.
3. Rapid fading (transitoriness) – Sounds are temporary. Writing and audio-recordings are techniques used to address this limitation of speech (and alas, lectures).
4. Interchangeability – One must be able to transmit and receive messages.
5. Total feedback – One must be able to monitor one’s own use of language.
6. Specialization – The organs used for language must be specially adapted to that task. Human lips, tongues and throats meet this criterion.
7. Semanticity – Specific signals can be matched with specific meanings. Different sounds exist for different words.
8. Arbitrariness – There is no necessary connection between a meaningful unit (e.g., word) and its reference.
9. Discreteness – There are distinct basic units of sound (phonemes) and meaning (morphemes).
10. Displacement – One must be able to communicate about things that are not present. One must be able to symbolically represent the past and the future.
11. Productivity – The units of sound and meaning must be able to be combined to create new sounds and meaningful units (sentences).
12. Duality of patterning – The sequence of meaningful units must matter (i.e., there must be a syntax).
13. Traditional Transmission – Specific sounds and words must be learned from other language users.

Although all of Hockett’s features are frequently cited as  essential characteristics of language , the first three elements are restricted to speech. These features do not apply to sign language, letter writing, reading, and other examples of non-vocal/auditory modes of symbolic communication.  The essential characteristics are interchangeability, semanticity, arbitrariness, discreteness, productivity, syntax, and displacement.

Language Acquisition

The principles of predictive and control learning help us understand the acquisition of language and the role it plays in subsequent human adaptation. At a few months old, infants start to babble and are able to make all the possible human sounds. Eventually, as the child is increasingly exposed to the sounds of her/his social unit, some of the sounds are “selected” and others removed from the repertoire. Routh (1969) demonstrated that infants are able to make subtle discriminations in sounds. The frequency of speaking either vowels or consonants could be increased if selectively reinforced with tickles and “coos.” It has been demonstrated that the mother’s vocal imitation of a child’s verbalizations is also an effective reinforcer (Pelaez, Virues-Ortega, and Gewirtz, 2011).

Children may learn their first word as early as 9 months. Usually the first words are names of important people (“mama”, “dada”), often followed by greetings (“hi”, “bye”) and favored foods. As described in Chapter 5, classical conditioning procedures may be used to establish word meaning. For example, the sound “papa” is consistently paired with a particular person. Children are encouraged to imitate the sound in the presence of the father. It may be the source of humor (or embarrassment) when a child over-generalizes and uses the word for another male adult. With experience, children learn to attend to the relevant dimensions and apply words consistently and exclusively to the appropriate stimuli or actions (e.g., “walk”, “run”, “eat”, etc.). Similarly, words are paired with the qualities of objects (e.g., “red”, “circle”, etc.) and actions (e.g., “fast”, “loud”, etc.). Children learn to abstract out the common properties through the process of concept formation. Words are also paired with quantities of objects. In the same way that “redness” may be a quality of diverse stimuli having little else in common, “three-ness” applies to a particular number of diverse stimuli.

Much of our vocabulary applies to non-observable objects or events. It is important to teach a child to indicate when “hurt” or “sick”, or “happy” or “sad.” In these instances, an adult must infer the child’s feelings from his/her behavior and surrounding circumstances. For example, if you see a child crying after bumping her head, you might ask if it hurts. As vocabulary size increases, meaning can be established through higher-order conditioning using only words. For example, if a child is taught that a jellyfish is a “yucky creature that lives in the sea and stings”, he/she will probably become fearful when swimming in the ocean.

Since different languages have different word orders for the parts of speech, syntax (i.e., grammatical order) must be learned. At about 18 months to 2 years of age, children usually start to combine words and by 2-1/2 they are forming brief (not always grammatical) sentences. With repeated examples of their native language, children are able to abstract out schemas (i.e., an organized set of rules) for forming grammatical sentences (e.g., “the car is blue”, “the square is big”, etc.). It is much easier to learn grammatical sequences of nonsense words (e.g., The maff vlems oothly um the glox nerfs) than non-grammatical sequences (e.g., maff vlem ooth um glox nerf). This indicates the role of schema learning in the acquisition of syntax (Osgood, 1957, p.88). Children usually acquire the intricacies of grammar by about 6 years of age. In the next chapter, we will describe the process of abstraction as it applies to concept learning, schema development, and problem-solving.

Vocabulary size has been found to be an important predictor of success in school (Anderson & Freebody, 1981). Major factors influencing vocabulary size include socio-economic status (SES) and the language proficiencies of significant others, particularly the mother. In a monumental project, Hart and Risley (1995) recorded the number of words spoken at home by parents and 7-month-to 36-month-old children in 42 families over a 3-year period. They found that differences in the children’s IQ scores, language abilities, and success in school were all related to how much their parents spoke to them. They also found significant differences in the manner in which low and high SES parents spoke to their children. Low SES parents were more likely to make demands and offer reprimands while high SES parents were more likely to engage in extended conversations, discussion, and problem-solving. Whereas the number of reprimands given for inappropriate behavior was about the same for low and high SES parents, high SES parents administered much more praise.

Speech becomes an important and efficient way of communicating one’s thoughts, wishes, and feelings. This is true for the Nukak as well as for us. Given the harshness of their living conditions and the limits of their experiences, the Nukak have much in common with low SES children within our society. Declarative statements (e.g., “the stick is sharp”, “the stove is hot”; “pick up the leaves”, “don’t fight with your sister”; “I am happy”, “you are tired”, become the primary basis for conducting much of the everyday chores and interactions.

Spoken language is observed in stone-age hunter/gatherer and technologically advanced cultures. There has been controversy concerning the role of nature and nurture in human language development (Chomsky, 1959; Skinner, 1957). Skinner, writing from a functionalist/behavioral perspective, tellingly entitled his book Verbal Behavior, not “Using Language.” Watson (1930) described thinking as “covert speech” while Skinner (1953) referred to “private behavior.” According to Vygotsky (originally published in 1934), children initially “think out loud” and eventually learn to “think to themselves.” Skinner suggested that speaking and thinking were not different in kind from other forms of behavior and that respondent conditioning (predictive learning) and operant conditioning (control learning) could provide the necessary experiential explanatory principles. There was no need to propose a separate “language acquisition device” to account for human speech.

We saw in Chapter 5, how predictive learning principles could be applied to the acquisition of word meaning. Basically, Skinner argued that words could serve as overt and covert substitutes for the control learning ABCs. As antecedents, words could function as discriminative stimuli and warning stimuli. For example, “Give mommy a kiss” or “Heads up!” As consequences, words can substitute for reinforcers and punishers (e.g., “Thank you.”, “Stop that!”). A rule is a common, useful, and important type of verbal statement including each of the control learning ABCs (Hayes, 1989). That is, a rule specifies the circumstances (antecedents) under which a particular act (behavior) is rewarded or punished (consequence). For example, a parent might instruct a child, “At dinner, if you eat your vegetables you can have your dessert” or, “When you get to the curb look both ways before crossing the street or you could get hit by a car.”

Chomsky, a psycholinguist, submitted a scathing critique of Skinner’s book, emphasizing how human genetics appears to include a “language acquisition device.” The Chapter 1 picture of the human homunculus, with its disproportional brain space dedicated to the body parts involved in speech, certainly suggests that the human being’s structure facilitates language acquisition. The homunculus also implies there is adaptive value to spoken language; otherwise these structures would not have evolved. Proposing a “language acquisition device”, similar to proposing an instinct to account for speech, is a circular pseudo-explanation. The language acquisition device is inferred from the observation of speech, it does not explain speech. Remember, a psychological explanation must specify specific hereditary and/or environmental causes. Chomsky does neither, whereas Skinner is quite specific about the types of experience that will foster different types of verbal behavior. It is not as though Skinner denies the role of human structure in the acquisition of speech or its importance as indicated in the following quote. “The human species took a crucial step forward when its vocal musculature came under operant control in the production of speech sounds. Indeed, it is possible that all the distinctive achievements of the species can be traced to that one genetic change” (Skinner, 1986). Neuroscientists and behavioral neuroscientists are actively engaged in research examining how our “all-purpose acquisition device” (i.e., brain) is involved in the learning of speech, reading, quantitative skills, problem-solving, etc.

Human beings may have started out under restricted geographic and climatic conditions in Africa, but we have spread all over the globe (Diamond, 2005). We developed different words and languages tailored to our environmental and social circumstances. There is much to be learned from the school of hard knocks, but it is limited to our direct experience and can be difficult or dangerous. Our verbal lives enormously expand learning opportunities beyond our immediate environment to anything that can be imagined. Indirect learning (i.e., observation or language) often speeds up adaptive learning and eliminates danger. It is not surprising that human parents universally dedicate a great deal of effort to teaching their children to speak. It makes life easier, safer, and better for them as well as their children.

MacCorquodale (1969) wrote a retrospective appreciation of Skinner’s book along with a comprehensive and well-reasoned response (1970) to Chomsky’s critique. Essentially, MacCorquodale described Chomsky as a structuralist and Skinner as a functionalist. That is, Chomsky attempted to describe how the structure of the mind enables language. Skinner was concerned with how language enables individuals to adapt to their environmental conditions. Paraphrasing Mark Twain, an article marking the 50th anniversary of its publication concluded that “Reports of the death of Verbal Behavior and behaviorism have been greatly exaggerated” (Schlinger, 2008).

Reading and Writing

It is language in written form that has enabled the rapid and widespread dissemination of knowledge within and between cultures. It is also the medium for recording our evolving advances in knowledge and technology. Early forms of Bronze Age writing were based on symbols or pictures etched in clay. Later Bronze Age writing started to include phonemic symbols that were precursors to the Iron Age Phoenician alphabet consisting of 22 characters representing consonants (but no vowels). The Phoenician alphabet was adopted by the Greeks and evolved into the modern Roman alphabet. The phonetic alphabet permitted written representation of any pronounceable word in a language.

The Arabic numbering system was originally invented in India before being transmitted to Europe in the Middle Ages. It permits written representation of any quantity, real or imagined, and is fundamental to mathematics and the scientific method, which rely on quantification and measurement. The alphabet and Arabic numbers permit words to become “permanent” in comparison to their transitory auditory form. This written permanence made it possible to communicate with more people over greater distances and eventually to build libraries. The first great library was established at Alexandria, Egypt in approximately 300 years B.C. Scrolls of parchment and papyrus were stored on the walled shelves of a huge concrete building (Figure 6.5).   Gutenberg’s invention of the printing press in 1439 enabled mass publication of written material throughout Western Europe (Figure 6.6). Today, e-books are available on electronic readers that can be held in the palm of your hands (Figure 6.7)! It should not be surprising that college student differences in knowledge correlate with their amount of exposure to print (Stanovich and Cunningham, 1993).

Figure 6.5  The library at Alexandria.

Figure 6.6  Gutenberg’s printing press.

Figure 6.7  The library now.

Attributions

Figure 6.5 “The library at Alexandria” by Wikimedia is licensed under CC BY-SA 4.0

Figure 6.6 “Guttenburg’s printing press” by עדירל is licensed under CC BY-SA 3.0

Figure 6.7 “Amazon Kindle” by Jleon is licensed under CC BY-SA 3.0