Chapter 4

THE LANGUAGE POINTS OF EXPLAINING AND LISTENING

During most of the time in Session I some participant, usually Mrs. V, was engaged in narrative or explanation. In narrative the behavioral units feature language. Syntactic structures which have meaning are articulated, i. e. , they refer to events which took place in remote contexts. Maybe forty per cent of the point units of language which occurred in Session I were informative in this sense.

The utterance consists of a sequence of point units. These are marked by address of the head and eyes and by voice projection. The face and sometimes the hands supplement the articulation. The language component consists of one or more syntactic sentences or the reciprocal behaviors of listening, questioning, and providing comprehension signals. Meanwhile, the remainder of the body may be engaged in some other relationship and some other kind of activity.

The units I am calling language points seem to correspond to what others have called ‘an utterance’ (Z. Harris 1951; Jaffe, Feldstein, and Cassota 1956).

THE COMPOSITION OF EXPLAINING

As we have noticed, Mrs. V offered an explanation in every cycle. At 23 minutes Whitaker explained the purposes and plans of the session. Malone started an explanation in Phase II but was interrupted. Marge made some brief statements of narrative, but almost all of the girl's utterances were comments about what Mrs. V said or else about her own feelings and state (see Chapter 5).

I will define narrative as a series of declarative statements about events which have taken place elsewhere• The explanation is a more comprehensive activity which involves one or more narrative sequences and some elaboration of these in response to questioning, challenges or noncomprehension. Mrs. V, for instance, would narrate in each Period 1 and defend or rationalize her account in each Period 2.

Mrs. V's Narrative

Mrs. V's narrative at the beginning of the session was typical of her later narrative behavior, and, in fact, of narration in general. We can observe many of its features at the beginning of the multimodality transcript. The lexical data of the entire session is reproduced in Appendix A.

The camera crew had technical difficulties and did not record the first minute or so of Session I. Before the film began the participants were introduced, they took chairs, and Mrs. V was invited to begin the narrative with an account of Marge's original illness. She started by telling of a morning, a year before, when Marge would not walk upstairs. At the beginning of the transcript Mrs. V quoted Marge as asking, ‘Help me upstairs.’ Then Mrs. V commented on her daughter's behavior with a tone of mock incredulity and disparagement: ‘A young girl like that.’

Then Mrs. V repeated her quotation: ‘Help me upstairs.’Mrs. V now had enacted three point units. She made two kinds of statements (and repeated one of these) — each having the configuration of head-eye positioning which demarcates it as a point unit.

Mrs. V was looking at Malone during these comments. Next she paused, saying: ‘and . . . ah . . .’ as she turned her head and eyes to Whitaker. These kinds of nonlanguage sounds are usual in a narrative. Technically they are components of paralanguage (Trager 1958; Trager and Smith 1956).

Maybe they are abortive statements that never proceed to utterance. Maybe they fill time and hold the floor until a speaker decides what to say next. In some cases they complete a measure in the meter of speaking. In any event Mrs. V used these nonlanguage utterances while she was shifting her orientation to Whitaker. A period of transition between units produces a movement that Pike (1954) called a ‘slur.’ Between very small units the transitional period may last for a large percentage of the unit duration and thus constitute a relatively conspicuous feature of the performance.

Facing Whitaker, Mrs. V articulated a next idea in her narrative: ‘The next day I thought she'd . . . get better by, ah, — she has the doctor's prescription.’ Meanwhile Marge had whispered something in her mother's ear; so Mrs. V turned to her daughter and said: ‘Did I say anything wrong, dear?’ Then Mrs. V turned back to Whitaker and went on with her narrative.

Notice the features of this sequence. A series of utterances are made, each of which deals with some idea — some phase or episode — and each is addressed to a different listener. Thus the total utterance is punctuated by a shift in the head-eye orientation.

The head, eye, face, and vocal-respiratory apparatus are involved in such a performance. The remainder of the body is held as a positional transfix oriented to the audience as a whole. The whole body is not turned from listener to listener nor was it turned to one listener as it would be in a one-to-one conversation.

Mrs. V's narrative was short this first Period 1. But after the second minute she resumed, and longer stretches of recounting occurred. The reader is invited to follow these on pages three and four of the foldout. The same configuration obtained. She addressed an utterance to Malone, then to Whitaker, and sometimes she swept her head slowly back and forth from one of them to the other. But her basic postural orientation was not altered. She faced the observers sitting upright, ankles crossed, her hands on her lap. Occasionally she facèd Marge and responded to one of the girl’s comments. Occasionally she raised her hands and gesticulated.

Whitaker's Explaining

At about 24 minutes Whitaker raised his head and faced an area between the two women. He said: ‘We’re doctors of people's feelings — and we want to see if we can help ya — with some of these feelings you've been talking about.’ Marge commented and Whitaker addressed her for a moment, replying: ‘Crazyness, you know — that's right, emotions.’

At this point Whitaker and Marge made the tactile contact I described in earlier chapters. Then Whitaker turned to Mrs. V and continued his explanation: ‘We wanted to talk to you, and then we're going to see Marge every day — except over the weekend — and so forth.’

The complete text is reproduced in Appendix A.

As I said in Chapter 3 the bodily regions are often used differentially in speaking. Whitaker's behavior exemplified this as he spoke or listened to Mrs. V, while he engaged in hand activity with Marge.

Mrs. V did not do this very often. She sometimes gesticulated with her hands, but ordinarily she kept her body motionless and addressed the men. Thus she maintained an orientation to a point midway between the two men. Whitaker did this too, but only his lower body was used in stationary orientation.

Whitaker oriented his pelvis and pointed his feet in such a way as to include both Malone and Mrs. V. It is a regular behavior in any group for standing members to point their feet in such a way as to include all other group members (Goffman 1963). When sitting they address their torsos to each other. If, however, they turn their upper body to some one person they usually maintain an orientation to the others by lower body placement, as Whitaker did in Session I.

Figure 4-1:Multiple Simultaneous Point Performances. Above, Malone is speaking to Mrs. V and facing her, but his body is oriented to Whitaker and the two men are using the same posture and moving synchronously (see Section C). Below, Whitaker speaking to Mrs. V, moving his hands in relation to Marge and orienting his body to a point between Malone and Mrs. V.

So Whitaker sometimes maintained three simultaneous relations and concurrently performed three point units; i. e. , speaking to Mrs. V, hand-playing toward Marge, and orienting his body to Malone and Mrs. V, thereby dividing his position into three modalities (Figure 4-1).¹

BEHAVIORS THAT MARK AND ADDRESS THE LANGUAGE POINT

In performing a language point, the head and usually the eyes are held in a given position until the utterance is completed. The face is pointed to a listener and the voice is projected for his audition. As in the case of a position, then the same behavior which marks the point's duration also addresses it to a vis-à-vis. Therefore, I will speak of the marking behavior as the behavior of address.

The Head-Eye Address in Narrative

The head is ordinarily held erect, forward, and slightly elevated in the address of a narrative. In middle-class Americans, at least, the face is directed to the auditor.

In face-to-face conversations, the orientation of the Western subjects I have studied, is rarely eye to eye. Each fixates his vision just out of range for eye-to-eye gaze holding — approximately at a spot between the cheek and shoulder of the other fellow. A direct gaze holding orientation is generally seen in a seduction or a dominance battle. In fact, the dominant participant usually commands an imaginary area in front of his eyes and face. Subordinates keep their gaze out of this area unless they are to make a direct confrontation. When central vision is focused on the cheek-shoulder area of a vis-à-vis, the remainder of his upper body will be visible in peripheral vision at the usual conversational distance. When movement occurs elsewhere in the body it will trigger an orienting reflex. Focal vision is shifted briefly to observe the moving part.

There is another convention of orientation in a conversation. It is impolite to look at listeners. One is to look at the speaker of the moment.

Thus, either sustained avoidance or maintained gaze holding may be an affront with relative strangers. As a consequence we rarely get to observe the behavior of listeners and we do not ordinarily see the total bodily behavior of others in conversation.

The accuracy of an orientation is also worthy of comment. In an audience of, say, forty or fifty people a speaker can ‘point’ his head and eye convergence with sufficient accuracy to single out one auditor. This object of attention usually is aware that he is being addressed. Similarly we can project our voices accurately enough to evoke a response from some one person in a group — sometimes even if this person is behind us. With reasonably high resolution, motion pictures, and a little practice, an observer can tell which participants are addressing each other at any time.

The interested reader should become acquainted with the careful experimental studies of Kendon (1965) on gaze in interaction.

The head is not always held rigidly during the point performance. If the speaker is addressing several listeners whom he regards as a unit, he may sweep the head laterally, back and forth from one listener to the next. But the movement is regular and oscillatory and the head is held in the same vertical plane. Or the speaker may nod at one listener and thus oscillate the head in a vertical plane. Sometimes the speaker advances his head progressively higher and higher or farther and farther forward with each point as he develops a theme. In any case the regularities of head posture during a point are recognizable. ²

Supplementary Transfix Behavior in the Point

Head and eye position are not the only features of the point transfix. The utterance is made in a voice projection appropriate to the distance between the speaker and his listeners. He usually holds this projection until the end of the point.

Some quality of language, some paralanguage form usually accompanies the point utterance. Such a paralinguistic feature may be customary for addressing that kind of listener, e. g. , baby talk, formal tones, contempt, or the like, and each utterance in that point may be marked with that voice quality.

If the hands are used in a point utterance, they are often positioned for the duration of the point.

Marge sometimes placed a tactile hold on her mother while making a point. Whitaker would take his pipe out of his mouth and hold it in the air until he finished a point.

Sometimes the hands are simply held in the air until a point is completed.

Figure 4-2:Supplementary Transfixes of the Point. Above, Marge kept her arm upright toward her face throughout a segment of lamenting behavior. Below, Whitaker would suspend his hand play with point utterances of his explanation.

On occasions a speaker or listener will hold a foot in the air or point a toe until the completion of a point unit. Sometimes point units are marked by cocking the head to one side during these utterances.

Junctures of the Language Point

When a speaker has completed his point unit or given up the prospect of doing so, he will shift his head position. Ordinarily he will lower the head and eyes for a moment and he may bring down his hands or fold them on his lap. If his head has been cocked, he may bring it erect. In short he discontinues the transfix behavior which he held during the point performance. He may, then, remount the address to make another point, shift the address to another listener, or himself become a listener.

As we will see, a characteristic linguistic juncture also occurs at the end of the last sentence of a point utterance. A supplementary juncture may also occur (Figure 4-3).

Figure 4-3:Supplementary Juncture Behavior. At the end of any utterance the speaker will turn his eyes or face away from the addressee or lower them. Whitaker often turned his head far away from Marge when he finished an utterance.

THE CONTENT OF THE POINT UTTERANCE

Markers of the Syntactic Sentence

Technically speaking, the point utterance consists of one or more syntactic sentences and associated kinesic elements. Normally these constituent subunits produce a recognizable and meaningful Gestalten of language, which are often called the ‘content.’

Each syntactic sentence of the point is a behavioral unit which is also marked.

In a sequence of syntactic sentences, each will be separated by a brief pause in articulation, and pitch will be held constant at this juncture. So will the head, the eyelids, and brows (Birdwhistell 1963). Such a ‘single-bar’ juncture indicates that the utterance is incomplete and the speaker will continue (Z. Harris 1951; Gleason 1955).

The last syntactic sentence of the point, however, will have a special, terminal juncture: a change in pitch and a stop in vocalization. This change is usually a fall in pitch in the case of a declarative. If the point ends with a question, however, the pitch will rise and the questioner may hold his address until he receives an answer (or gives up the expectation).

This terminal linguistic juncture is accompanied by a terminal kinesic or body-motion signal (Birdwhistell 1963). The head and eyelids (or sometimes the hand) are slightly lowered with the pitch fall of a declarative and raised with the pitch rise of a question. These syntactic junctures occur at the completion of speech even if the head-eye orientation is maintained beyond this time (Figure 4-4).

Figure 4-4:Some Postural Kinesic Markers of American Syntactic Sentences [Courtesy of R. L. Birdwhistell. Drawings reproduced from the author’s article in Psychiatry (Scheflen 1964). ]

These linguistic and kinesic junctures indicate the completion of the point utterance. But they also provide additional signals about the expected response. Thus a rise in pitch (terminating a question in grammar) demands a verbal reply, while a pitch fall (with a declarative) does not.

The Lexical Structure of the Syntactic Sentence

Sometimes the point consists of a single syntactic sentence as in the case of Mrs. V's opening narrative which is described above. This is often the case as well with questions and commands (see Chapter 5) and the syntactic sentence may contain but a word or two. Thus a participant may hold a head-eye address and occasionally interject a single syntactic sentence, or he may form a point address or even a total position for a single utterance, saying, ‘No,’ or, ‘I agree.’

Syntactic Order in the Sentence

The tradition of sequencing (which we can abstract as syntax) requires that the first words of the syntactic sentences belong to a class known as the subject and the second to a class known as the predicate. The predicate consists of a verb and usually an object. So the syntactic sentence generally presents a proposition, e. g. , a noun (subject) is depicted as taking an action (verb) upon another noun (object) (Whorf 1956; Chomsky 1957).

Morphemic Subunits

Notice that the syntactic sentence is itself composed of subunits of even smaller behavior — units of pitch and stress change and units of vocalization. The lexication of the syntactic sentence consists of phrases and words (which I will not here differentiate). Some words are morphemes and others are made up of morphemes.

The morpheme is an utterance which is shorter than the word. In grammar the morphene is classed as a stem-form, a prefix, or a suffix. These are combined to form words.

A great many combinations of phonemes can occur in the morphemic sequence, but there are rules which constrain the types of combination. There are no ‘xz,’ ‘cz,’ or ‘qa’ sequences in English, for example, but sequences of consonants followed by vowels are common. By the same token, sequences of prefix-stem-suffix are established by cultural tradition, so we have forms such as ‘child,’ where the morpheme is a word in itself, and combinations such as ‘children’ and ‘childish.’ But we do not have the forms, ‘prochild’ or ‘childious.’

We can abstract rules from these observations. A good many kinds of sequential order are allowed by the customs of language, but the number is limited. The orders and combinations are determined by syntactic rules which have evolved in common culture (Spair 1921; Chomsky 1957, 1966; Sebeok, Hayes, and Bateson 1964. This lawfulness applies at all levels of integration from the morpheme to the whole transaction and probably to even high levels of integration. )

The lexical behavior of Session I consisted of the customary combinations of morphemes, each of which were more or less standard in their morphology. About 7,000 morphemes were uttered in Session I and maybe 5,000 words. Virtually all of these were standard English forms, which the researchers and all of the participants, as far as we could tell, understood. There were a few unusual forms. Marge and Mrs. V called day dreams or hallucinations, ‘mind-pictures.’ And two Italian forms appeared — ‘fungoo ala madre’ and ‘Maria.’ Thus with few exceptions all participants used the same basic system of lexical forms — they all spoke English.

Even units as small as a word appear to have markers. Condon and Ogston (1967), using very fast filming techniques, have shown a microeyeblink at the beginnings and ends of certain morphemes or words. Phrases also are often marked by head nods.

Birdwhistell (1966) has shown that a lateral microsweep of the head is made over compound words which would be hyphenated in written English. These begin exactly at the start of the first word and end with completion of the last. Presumably the head sweep indicates that the two lexical forms are to be seen as a unit.

Phonemes

Each morpheme is composed, in turn, of a sequence of elemental sounds whose forms are also patterned so that they are recognizable. Their sequencing is structured to produce given morphemes.

A speaker manipulates the vibrating column of respiratory air by placement of his larynx, glottis, tongue, teeth, and lips (McQuown 1964). Of an estimated 100,000 possible sounds or phonemes that can be made in this way, each culture uses some fifteen or more, which are the distinctive elementary units of its language. In English most linguists distinguish forty-three such basic elements of language which are called phonemes.

This sketchy account does not do justice to what is known about language, but we have to cover thirty minutes of behavior, so we cannot focus on the small behaviors of a single modality. The reader is, therefore, urged to consult the comprehensive works of such structural linguists as Sapir (1921, 1956), Bloomfield (1933), Z. Harris (1951), Hockett (1958), Chomsky (1966), and Gleason (1955).

Notice also the number of levels of behavioral integration we must deal with in a transaction — at least eight are identifiable from the phoneme to the cycle. Each has recognizable structure and marking indicators.

Kinesic Microacts with Morphemes and Phrases

Kinesic indicators occur with these subunits of speech as well as with the junctures of the syntactic sentence.

There are also, at all levels, microacts which Birdswhistell (1963, 1969) has called ‘referencing signals.’ Pronouns and prepositions are often ambiguous in their reference. Participants using these classes of morphemes generally point with their faces, eyes, or hands to the referents. In using ‘we’ a head sweep is often employed to demarcate a subgroup (Birdwhistell 1966). In using phrases like ‘up and down’ or ‘over there՝ a sweep of the head is employed to indicate some special area (Birdwhistell 1966)’. Some physical feature like an item of decor or a picture that symbolically represents relevant context may be referred to by pointing with the head.

A speaker may use his hands to draw a figure in the air and thus try to picture something he is describing. Mrs. V often did this. Birdwhistell calls such behaviors ‘demonstratives’ (Birdwhistell 1969).

Ordinarily particular facial configuration occurs with each point unit and often a gesture of the hands. I will describe these behaviors in the next chapter.

Paralinguistic Forms

I have already mentioned paralanguage sounds. In addition an utterance at any level will have a pattern of vocal qualifiers. Usually these are specific for a given syntactic sentence and point, and thereby both mark the unit and qualify it (see Chapter 7).

POINT UNITS IN LISTENING AND QUESTIONING

Listeners ordinarily address the speaker and perform point units of head and eye address that are reciprocal to and synchronous with those of the speaker. They may sometimes lexicate in these point units, questioning or commenting, for example. But in either event they use the same address system and ordinarily they will change point orientation with the speaker. In rapport and other qualities of coordinate relationship the participants will nod, sweep their heads, and move their hands in microsynchrony at speeds of about one forty-eighth of a second (Condon and Ogston 1966; Condon 1968), and sometimes this synchrony of movement is gross and readily visible without special filming techniques (Scheflen 1966; Kendon 1968). Thus, as Condon says, the participants dance together.

Psychotherapists tend to use a minimum of activity in listening. They keep ‘dead-pan’ expressions and comprehension signals, for example. Whitaker and Malone did this, but not as markedly as some therapists do (see Chapter 11). In their positions of listening and questioning they performed the following three types of customary point units.

Attending. They cocked their heads slightly and addressed Mrs. V. They suppressed virtually all body movement and facial expression in the dead-pan face that psychotherapists deliberately use to minimize influence on the speaker. They would occasionally signal attentiveness: leaning the head forward, cocking it, turning the ear, and using an overwide position of the eyelids.

Questioning. Whitaker repeatedly asked questions in Session I. He would extend his neck so that his head was high, jutted forward, and slightly cocked. He would direct his face, gaze, and convergence to the addressee — Marge or Mrs. V — take his pipe out of his mouth and hold it forward. He would then articulate the question. After speaking, he would raise his head still farther, jut his jaw slightly, and raise his eyebrows until an answer was begun. Then he would lower and retract his head, and return his pipe to his mouth (Figure 4-5).

Figure 4-5:Head Protrusion by Whitaker on Asking a Question

Commenting. All the auditors of Mrs. V's narrative offered comments about what she said. These point units will concern us in the next chapter.

COMMENT: LANGUAGE UNITS THAT ARE ‘SUFFICIENTLY’ MEANINGFUL

The Sufficient Depiction of an Idea

In conversational activities where narration or description is used, it is unusual for a participant to speak a single syntactic sentence and then relinquish the floor. Usually a succession of such segments are necessary to portray an experience. A single idea might leave so much ambiguity as to give the utterance little specific meaning. In other words, it might have so many possible meanings that it would have little meaning.

One might conceive the task in lexical activities as the adequate representation of a referent or context not immediately perceivable. The necessary behavior for such representation must be coded in the sense that a kinesic or lexical unit must be employed which has an agreed-upon relation to the object or event of reference. Some sketch of the referent must be drawn with icons or symbols or both. When the referent is symbolized by nouns and verbs, these may have to be supplemented with qualifying adjectives, phrases, demonstratives, and gestures.

Some lexical morphs refer to more highly specific referents known to almost everyone. Phrases like ‘the Statue of Liberty,’ ‘the U.S. Senate,’ or ‘the Crucifix’ are examples, but ordinarily words can have multiple referents. Words like ‘set’ or ‘put,’ for example, have dozens of lexographic definitions. Kinesic behavior may be even more ambiguous. Gestures or facial expressions of disapproval may give no indication of just what is being disapproved.

Bateson (1969) has said that the meaning of a behavior is inversely proportional to its ambiguity. We then achieve a greater meaning by acquiring more information and thus reducing this ambiguity. We progressively eliminate alternative possibilities. So in practice the meaning of an act will be specified by the addition of more and more behavior which supplies information about the reference and referent. The behavior necessary to communicate about abstract and personal matters is likely to be complex and highly integrated.

In Session I complex positions were assembled to carry the narrative, and multiple cyclic repetitions occurred to elucidate the family's history. Similarly in the nonlexical channels elaborate quasi-courting and tactile routines were used — apparently to develop a relationship. While it was clear that Marge was ironically representing sexual behavior, it was not clear what the implications were. To determine whether she was seeking intercourse or attention or merely teasing we would have to know much more about her, about Italian-American courtship behavior, and so on (English et al. 1965).

Assembling Higher and Higher Integrations of Behavior

Consider, then, an operational postulate. The particular sequences or units that a participant enacts in a transaction are those that are usually sufficient in his experience to get the job done.³

In a conversational transaction the speaker must progressively build an unambiguous utterance. If a syntactic sentence is insufficient, a point and perhaps a position or a sequence of positions will be used. We thus can think of the participant's task as the assemblage of units in a hierarchy of levels.

It is as though the participant were a builder. He knows how to form with his body a number of building blocks of sound and movement. He puts these together according to some cognitive map into recognizable and traditional Gestalten of greater and greater complexity.

He can assemble customary forms of movement and sound in several directions. He can order more and more symbolic units until he forms a communicative Gestalt. He can use different regions of his body to qualify the statement, to gain attention to it, and to minimize interference. And he can orient each of these actions and solicit mutual participation to build the relational structures of communication.

Cognitive Images of Behavioral Form

People must maintain some cognitive image of the behavior they perform. They have learned certain patterns of behavior and they can recall and re-enact these when they find themselves in an appropriate context.

If a participant is to produce a sufficiently meaningful unit of language behavior, he must have in mind an image of the remote event or contexts he is to depict; also he must be a cognitive blueprint of the order of speech behavior which he will have to follow to represent the image adequately for others. And he must hold an image of the contexts in which this particular behavior is appropriate.

The cognitive blueprint, then, must be more than a simple map of the conventional performance. It must be representation of complicated hierarchies of behavioral integration.

These ideas have already been developed by Miller, Galanter, and Pribram (1960). These men have hypothesized that cognitive images of performances and situations are maintained, which are hierarchically structured and they may be enacted as. plans and recalibrated with feedback information in the situation (see Chapter 8). Pribram has postulated that such images may be maintained in circuits which involved Oligodendroglia, since these proliferate with learning (Pribram 1966). Thus Tolman’s idea of cognitive maps (1948, 1951) has been considerably developed. One can think about such representations in a temporal perspective. Any larger unit or context in a hierarchy extends backward in time from immediate behavior and it appears to extend forward in time when it is replicated in memory. Thus one can visualize the next events in a pattern. He can imagine and anticipate usual consequences. He can have expectancies (Tolman 1932; Lewin 1951; Rotter 1954); that is, he can imagine goals and behaviors necessary to achieve these.