Introduction:

A. ANALYSIS WITHOUT REDUCTIONISM —

An Operational Comment

In this section I will make an analysis of the positional units. But before I begin there are two kinds of information I must provide — the principles of analysis that will be followed, and some additional data from Session I.

In the analysis, elements of the whoie are isolated so they can be described and measured separately. Accordingly I will describe subunits of the positions. But there are certain principles of context analysis which govern the operations of such an analysis. If we are to see how behavioral integration is communicative we cannot destroy our view of this integration in the process of isolating and viewing the component behaviors one by one. Therefore we must avoid certain reductionistic practices that have characterized the presystems analysis of behavior.

CLASSICAL PROBLEMS IN THE ISOLATION OF VARIABLES

If one is to conduct an analysis of a transaction which is not recorded on film or videotape, he is forced to observe selectively. He must take one or a few variables, observe and note these, and ignore the others. At a single visualization there is no way to record all of the behaviors which transpire and deal systematically with its integration. So in the past, certain reductionistic procedures have been forced on the investigator. Let’s review two of these as a means of approaching the procedures we should use.

1. The a priori selection of certain behaviors: The remainder being ignored. When we were limited to noting a few of many events we had to decide in advance what behaviors we thought were especially important or interesting. The best we could do was watch for these. Thus only certain events were studied and we had no way to reconstruct the integration of events. For example: Until the last decade we had a tendency to say that communicative behavior consisted of speech. And it is still common practice to reduce the content of a transaction to a mere transcript of the sequencing of words. In this practice other elements of the position were ignored. Thus in psychological research the stress and pitch patterns, the nonlanguage sounds, and vocal qualifiers of language were neglected. The linguist ignored the postural markers, the multiple types of bodily language behaviors, and the physical task behavior. Such reductionism by selection is like seeing the organism as a system of endocrines alone.

As a consequence we could not identify the behavioral basis of the systems concept of organization (Bertalanffy 1950, 1960). We could not describe step by step or level by level the Gestalten of behaviors which were recognizable and communicative. We had to act as though the who!e consisted only of certain components. And sometimes we came to believe this. We made reductionistic generalizations. A table was ‘nothing more than’ a collection of atoms; the organism consisted of cells; speech was made up of words.

The fact, of course, is that speech consists of complex integrations of address, juncture, verbalization, pitch and stress patterns, and vocal modification. And speech is integrated hierarchically — that is, the utterance is made up of integrations of syntactic sentences and kinesic behaviors; the syntactic sentence is an integration of pitch, stress, morphemes, junctures, and head movements; the morpheme is an integration of vocalizations, stops, and head-eye movements; and so forth (Chapter 4). So we avoid the practice of reductionism without regard to intermediate levels in present attempts to study the communicative structure of behavior.

2. The reliance on abstractions about behavior. We could take cognizance of complexity without film recordings if we gave up the hope of describing the constituent elements of behavior and instead merely abstracted their properties or qualities. Then we could scan a complicated action and generalize, for instance, that it was sadistic or masochistic, aggressive or passive, attentive or autistic, communicative or incomprehensible. We could then take these abstracted, mentalistic ideas which we formed about the behavior in progress and deal with them as variables for quantification or correlation.

But the results were not very satisfactory. The next investigator would abstract a different set of qualities and, lacking a precise notation of the events which occurred, we could not compare notes. Often we ended up with a set of artificial abstractions which did not accord with the behavioral forms and segments which naturally occur in a transaction.

The classical dichotomy between verbal and nonverbal behavior is a prime example of such an artificial, conceptual dichotomy which does not fit the naturalistic integration of units in human discourse.

There are certain nonverbal acts which invariably accompany verbalization, e.g., stress and pitch changes. There are verbal sounds which are not part of language proper — sighs, sobs, and ahems, for example. And there are classes of nonspeech behavior which have no regular association with each other and, therefore, do not constitute any kind of functional entity which we can lump as ‘nonverbal.’

If we are to see how behavior is communicative we must identify the customary Gestalten or configurations which are recognizable among participants — those to which meaning is customarily ascribed.

In English language, for instance, we have a morpheme ‘dog’ and another ‘god,’ but we do not have a form, ‘og.’ We would not segment a ‘no smoking’ sign into ‘nosmo’ and ‘king.’ The half inning in baseball consists of three putouts. We cannot understand the game if we think that two or four putouts are naturally occurring subunits. Similarly the gesture, palm-on-chest, consists of raising the palm to the chest, placing it, and returning it to baseline. There is no point in subdividing the usual performance into hand-raising and hand-lowering, nor would we gain anything by including foot waggling in this gestural configuration just because it happened to co-occur.

In context analysis the subunits of a unit are determined by systematic observation by a methodology I describe in Appendix B. We determine which actions regularly occur together and pass as recognized Gestalten in a transaction.

By the same token we do not want to describe units of behavior as combinations of events which are of various logical types. The behavioral unit which can be seen in communication consists of sounds and movements. Items of the physical ecology may be used in the performance of a unit, but these items are of a different logical type. Also we can abstract qualities and make judgments about behavior, but these judgments constitute behaviors of a different logical type. Because they are not in the behavioral unit, the unit does not contain goodness, masochism, or recognizability, and we must distinguish between what occurred in the participant’s actions and the inferences which he or we made about them. So the descriptive adjectives like social, psychological, or environmental are not elements of the behavioral unit, but belong rather to the mentalistic operations which accompany visible behavior. These, too, are important in communicational processes (see Chapter 8), but we do not want to mingle them willy-nilly with our descriptions of visible behavior. So the position may consist of speech, smiling, and smoking. It does not consist of words, friendliness, and a pipe.

ANALYSIS OF THE POSITIONS IN SESSION I

To fulfill the criteria of context analysis, then, I have to study and describe all of the behavioral elements which occurred. This is nearly possible because I can go over the film again and again until I have described all of the movements, sounds, and holds which are captured on the moving picture. We can hold to the operational principle that all participants are doing ‘something’ at all times, for even silence and immobility are meaningful. So I cannot have gaps in the chronology of the analysis.

I must examine the relations of each behavioral item to find out which sets or clusters occur together regularly. These make up the traditional, customary, or naturally occurring units which are communicative (see Appendix B). Then I have to determine the contextual location of each subunit. I have to be able to assign it to the larger unit of behavior in which it occurs. Eventually I reconstruct the schema of integration level by level — from positions to microacts and from integrations of micoacts back to positions. This I have done for Session I.

PRESENTING THE DATA IN THIS VOLUME

In order to present the analysis in this volume, then, I should take each position one by one and present the details of its composition. I should show the exact behavioral relations of the speech, gestures, postures, and other subunit components for explaining, passive protesting, contending, and other positions. But no reader would wade through such endless detail. To make this volume readable I must jump to certain abstractions and present some measure of generalization.

Here are the steps I will follow in presenting the analysis.

First, I will isolate certain subunit behaviors of the position and indicate their level. Some of these I will catalogue as completely as I can. For example, I will try to describe all of the gestures, tasks, representational and juncture behaviors which occurred, because these kinesic units are not well known and need to be characterized. But in the case of the speech units I will simply illustrate some of these in order to review their structure. The remainder will merely be reproduced in Appendix A as a traditional transcript or orthography of speech content.

Second, I will classify the various subunits of behavior according to logical type schemas which have already been advanced by other authors. Miller (1965) has distinguished systems of three logical types: physical, behavioral, and conceptual. I will begin by converting the phenomena of Session I into the logical type of behavioral systems. (Later I will move into a discussion of the invisible behaviors of cognition.) M. Harris (1964) has distinguished between actonic and communicative behaviors, and Bateson (1955) has distinguished between communicative and metacommunicative behaviors. Having studied the appearances of the subunits of Session I in larger units, I can abstract what each one appeared to do in the larger context. I will use these abstractions of function as a basis for assigning the various subunits to one of the logical type categories which Harris and Bateson have distinguished.

Then, I will present a description of each of three logical types of behavior in each of the three chapters of Section B: Chapter 3: Point Units Which Serviced and Related the Participants; Chapter 4: The Language Points of Explaining and Listening; and Chapter 5: The Metacommunicative Points of Session I.

Thirdly, I will add an addendum to Section В to review the units of Session I at the level of the point — a level just below that of the position. Then the reader who is interested in detail can refer back to Section I and get an idea of how the subunits were integrated into the positions of the session.

B. THE MULTIMODALITY TRANSCRIPT

To depict more of the details of the positions, I prepared a multimodality transcript of Session I and will produce the first five and one-half minutes of this transcript in the fold-out sheets to follow. The transcript depicts the lexical behavior of each participant through the first cycle, cycle A, and the second Period 1 of cycle B. It also depicts certain postural configurations and body movements for each participant. Unfortunately a detailed linguistic and paralinguistic analysis was not made of Session I, so I can only report the speech orthography and mention certain linguistic and paralinguistic features.

I hope the reader will follow this transcript as he reads the text of the chapters to come. See the foldout following page 80.