In our everyday vocabulary, we often associate language with words. From a behavioral perspective, this presents a number of problems. Consider the word cat. Upon seeing that word, one might say "cat," write cat, or make a hand sign for cat; or, one might speak, write, or sign a similar word, such as meow. Hearing the word cat might also cause one to say "cat" or "meow," write cat or meow, or sign cat or meow. How are we to classify all these various combinations of stimuli and responses? This is to say nothing of what the word cat itself refers to and how we might account for its meaning.
Psychologists define language in many ways (Catania, 1986). Let us assume that language is behavior and that behavior occurs in relation to the environment. There are two general approaches we could take (Catania, 1998). First, we can hold the relation between the behavior and the environment constant, and explore the critical properties of either one. This is called a structural approach. For example, a structural analysis of language might focus on the physical features of vocal behavior to determine the organization of muscular movements involved in articulation, phonology, and quality of voice. Or, the focus might be on the physical features of environment, a related topic being syntax and how words are organized in sentences.
Second, we can conduct a functional analysis by holding the critical properties of the behavior and the environment intact, and varying the relation between the two. With respect to language, we would seek to discover the environmental variables that control various instances of verbal behavior. This is consistent with Skinner's (1957) approach to language.
For Skinner (1957), behavior is verbal to the extent that it produces a reinforcing consequence mediated by the behavior of another person. (We assume that you are already familiar with the concept of reinforcement; if not, click here to learn more about it.) What do we mean by "mediated"? Suppose you want to know the time. You might look at your watch and see that it is 10:00 PM. Reinforcement for your looking behavior is direct; another person is not involved. Or, you might ask your father "What time is it?," and he replies "10:00 PM." Here, reinforcement for your asking behavior is indirect; it was obtained through another person's action.
In most communication, both persons involved are engaging in some sort of verbal behavior. For purposes of analysis, we need to take a particular perspective. The person whose verbal behavior we are attempting to analyze will be called the speaker, and the person who mediates reinforcement for that behavior will be called the listener. (The fact that a speaker can function as his or her own listener greatly expands the interpretative power of a functional analysis; however, this is an advanced topic beyond the scope of the present tutorial.)
Note that Skinner's (1957) definition of verbal behavior says nothing about the form of the response. We typically equate the words verbal and vocal, but this is not the case in Skinner's analysis. A verbal response could be speaking, but it could also be writing, gesturing, signing, drawing, nodding, or smiling. The form of the response is irrelevant; rather, its function determines its verbal nature.