Excerpts from Must Read Books & Articles on Mental Health Topics
Books, Part X

The Pragmatics of Human Communication- Paul Watzlawick, et al.
Chapter III, Pathological Communication, pp. 72-117

Introduction

Each of the axioms just described implies, as corollaries, certain inherent pathologies that will now be elaborated. In our opinion, the pragmatic effects of these axioms can be illustrated best by relating them to disturbances that can develop in human communication. That is, given certain principles of communication, we shall examine in what ways and with what consequences these principles can be distorted. It will be seen that the behavioral consequences of such phenomena often correspond to various individual psychopathologies, so that in addition to exemplifying our theory we will be suggesting another framework in which the behavior usually seen as symptomatic of mental illness may be viewed.

The Impossibility of Not Communicating

Mention has already been made in the foregoing of the dilemma when it was pointed out that mentally ill patients sometimes behave as if they tried to deny that they are communicating and then find it necessary to deny also that their denial is itself a communication. But it is equally possible that the patient may seem to want to communicate without, however, accepting the commitment inherent in all communication. For example, a young woman bounced into a psychiatrist's office for her first interview and cheerfully announced: "My mother had to get married and now I am here." It took weeks to elucidate some of the many meanings she had condensed into this statement, meanings that were at the same time disqualified both by their cryptic format and by her display of apparent humor and zestfulness. Her gambit, as it turned out, was supposed to inform the therapist that
(1) she was the result of an illegitimate pregnancy;
(2) this fact had somehow caused her psychosis;
(3) "had to get married," referring to the shotgun nature of the mother's wedding, could either mean that Mother was not to be blamed because social pressure had forced her into the marriage, or that Mother resented the forced nature of the situation and blamed the patient's existence for it;
(4) "here" meant both the psychiatrist's office and the patient's existence on earth, and thus implied that on the one hand Mother had driven her crazy while on the other hand she had to be eternally indebted to her mother who had sinned and suffered to bring her into the world.
    This, then, is a language which leaves it up to the listener to take his choice from among many possible meanings which are not only different from but may even be incompatible with one another. Thus, it becomes possible to deny any or all aspects of a message. If pressed for an answer to what she had meant by her remark, the patient above could conceivably have said casually: "Oh, I don't know; I guess I must be crazy." If asked for an elucidation of any one aspect of it, she could have answered: "Oh no, this is not at all what I meant. . . ." But even though condensed beyond, immediate recognition, her statement is a cogent description of the paradoxical situation in which she finds herself, and the remark "I must be crazy" could be quite appropriate in view of the amount of self-deception necessary to adapt herself to this paradoxical universe. For an extensive discussion of negation of communication in schizophrenia, the reader is referred to Jay Haley (60, pp. 89-99), where there is a suggestive analogy to the clinical subgroups of schizophrenia.
    The converse situation exists in Through the Looking Glass when Alice's straightforward communication is corrupted by the Red and the White Queens' "brainwashing." They allege that Alice is trying to deny something and attribute this to her state of mind:

"I'm sure I didn't mean-" Alice was beginning, but the Red Queen interrupted her impatiently.
"That's just what I complain of!  You should have meant! What do you suppose is the use of a child without any meaning? Even a joke should have a meaning--and a child is more important than a joke, I hope. You couldn't deny that, even if you tried with both hands."
"I don't deny things with my hands," Alice objected.
"Nobody said you did," said the Red Queen. "I said you couldn't if you tried."
"She is in that state of mind," said the White Queen, "that she wants to deny something--only she doesn't know what to deny!"
"A nasty, vicious temper," the Red Queen remarked; and then there was an uncomfortable silence for a minute or two.

One can only marvel at the author's intuitive insight into the pragmatic effects of this kind of illogical communication, for after some more of this brainwashing he lets Alice faint.  The phenomenon in question, however, is not limited to fairy tales or mental illness. It has much wider implications for human interaction. Conceivably the attempt not to communicate will exist in any other context in which the commitment inherent in all communication is to be avoided. A typical situation of this kind is the meeting of two strangers, one of whom wants to make conversation and the other does not, e.g., two airplane passengers sitting next to each other.  Let passenger A be the one who does not want to talk. There are two things he cannot do: he cannot physically leave the field, and he cannot not communicate. The pragmatics of this communicational context are thus narrowed down to a very few possible reactions:

"Rejection" of Communication
Passenger A can make it clear to passenger B, more or less bluntly, that he is not interested in conversation. Since by the rules of good behavior this is reproachable, it will require courage and will create a rather strained and embarrassing silence, so that a relationship with B has not in fact been avoided.

Acceptance of Communication
Passenger A may give in and make conversation. In all probability he will hate himself and the other person for his own weakness, but this shall not concern us. What is significant is that he will soon realize the wisdom of the army rule that "in case of capture give only name, rank, and serial number," for passenger B may not be willing to stop halfway; he may be determined to find out all about A. including the latter's thoughts, feelings, and beliefs. And once A has started to respond, he will find it increasingly difficult to stop, a fact that is well known to "brainwashers."

Disqualification of Communication
A may defend himself by means of the important technique of disqualification, i.e., he may communicate in a way that invalidates his own communications or those of the other. Disqualifications cover a wide range of communicational phenomena, such as self-contradictions, inconsistencies, subject switches, tangentializations, incomplete sentences, misunderstandings, obscure style or mannerisms of speech, the literal interpretations of metaphor and the metaphorical interpretation of literal remarks, etc.   A splendid example of this type of communication is given in the opening scene of the motion picture Lolita when Quilty, threatened by the pistol-wielding Humbert, goes into a paroxysm of verbal and nonverbal gibberish while his rival tries in vain to get across his message: "Look, I am going to shoot you!" (The concept of motivation is of little use in deciding whether this is sheer panic or a clever defense.) Another example is that delightful piece of logical nonsense by Lewis Carroll, the poem read by the white Rabbit:

They told me you had been to her,
And mentioned me to him:
She gave me a good character,
But said I could not swim.
He sent them word I had not gone
(We knew it to be true):
If she should push the matter on,
What would become of you?
I gave her one, they gave him two,
You gave us three or more;
They all returned from him to you,
Though they were mine before.

And so on for three more stanzas. If we now compare this with an excerpt from an interview with a normal volunteer subject who is obviously uncomfortable in answering a question put to him by the interviewer but feels that he should answer it, we find that his communication is suggestively similar both in its form and in the paucity of its content:

Interviewer: How does it work out, Mr. R., with your parents living in the same town as you and your family?
Mr. R: Well we try, uh, very personally I mean . . . uh, I prefer that Mary [his wife] takes the lead with them, rather than my taking the lead or what. I like to see them, but I don't try too much to make it a point to be running over or have them . . . they know very definitely that . . . oh, it's been always before Mary and I ever met and it was a thing that was pretty much just an accepted fact--in our family I was an only child--and they preferred that they would never, to the best of their ability, not, ah, interfere. I don't think there is . . . in any case I think there is always a-an underlying current there in any family, I don't care whether it's our family or any family. And it is something that even Mary and I feel when we . . . both of us are rather perfectionists. And, ah, yet again, we're very . . . we are . . . we are st- rigid and . . . we expect that of the children and we feel that if you got to watch out--I mean, if ah . . . you can have interference with in-laws, we feel, we've seen others with it and we've just . . . it's been a thing that my own family tried to guard against, but ah . . . and, uh, like here--why we've . . . I wouldn't say we are standoffish to the folks.

    It is not surprising that this kind of communication is typically resorted to by anybody who is caught in a situation in which he feels obliged to communicate but at the same time wants to avoid the commitment inherent in all communication. From the communicational point of view there is, therefore, no essential difference between the behavior of a so-called normal individual who -has fallen into the hands of an experienced interviewer and of a so-called mentally disturbed individual who finds himself in the identical dilemma: neither can leave the field, neither can not communicate but presumably for reasons of their own are afraid or unwilling to do so. In either case the outcome is likely to be gibberish, except that in the case of the mental patient the interviewer, if he be a symbol-minded depth psychologist, will tend to see it only in terms of unconscious manifestations, while for the patient these communications may be a good way of keeping his interviewer happy by means of the gentle art of saying nothing by saying something. Similarly, an analysis in terms of "cognitive impairment" or "irrationality" ignores the necessary consideration of context in the evaluation of such communications.  Let us once more point to the fact that at the clinical end of the behavioral spectrum, "crazy" communication (behavior) is not necessarily the manifestation of a sick mind, but may be the only possible reaction to an absurd or untenable communication context.

The Symptom As Communication
Finally, there is a fourth response passenger A can use to defend himself against B's loquacity: he can feign sleepiness, deafness, drunkenness, ignorance of English, or any other defect or inability that will render communication justifiably impossible. In all these cases, then, the message is the same, namely, "I would not mind talking to you, but something stronger than I, for which I cannot be blamed, prevents me." This invocation of powers or 'reasons beyond one's ' control still has a rub: A knows that he really is cheating. But the communicational "ploy" becomes perfect once a person has convinced himself that he is at the mercy of forces beyond his control and thereby has freed himself of both censure by significant others as well as the pangs of his own conscience. This, however, is just a more complicated way of saying that he has a (psychoneurotic, psychosomatic, or psychotic) symptom. Margaret Mead, in describing the difference between American and Russian personalities, remarked that an American might use the excuse of having a headache to get out of going to a party but the Russian would actually have the headache. In psychiatry, Fromm-Reichmann, in a little-known paper, pointed out the use of catatonic symptoms as communication, and in 1954 Jackson indicated the utility of the patient's use of hysterical symptoms in communicating with his family. For extensive studies of the symptom as communication the reader is referred to Szasz and Artiss.
    This communicational definition of a symptom may seem to contain a moot assumption, namely that one can convince oneself in this way. Instead of the rather unconvincing argument that everyday clinical experience fully supports this assumption, we should like to mention McGinnies' experiments on "perceptual defense". A subject is placed in front of a tachistoscope, a device by which words can be made visible for very brief periods of time in a small window. The subject's threshold is determined for a few trial words and he is then instructed to report to the experimenter whatever he sees or thinks he sees on each subsequent exposure. The list of test words is composed of both neutral and "critical," emotionally-toned words, e.g., rape, filth, whore. A comparison between the subject's performance with the neutral and with the critical words shows significantly higher thresholds of recognition for the latter, that is, he "sees" fewer of these words. But this means that in order to produce more failures with the socially tabooed words, the subject must first identify them as such and then somehow convince himself that he was unable to read them. Thus he spares himself the embarrassment of having to read them out loud to the experimenter. (In this regard, we should mention that, in general, psychological testing must consider the communicational context of these tests. There can hardly be any doubt, for instance, that it must make quite a difference to the subject and his performance whether he has to communicate with a shriveled old professor, a robot, or a beautiful blonde. In fact Rosenthal's recent careful investigations into experimenter bias have confirmed that complex and highly effective though as yet unspecifiable communication transpires even in rigidly controlled experiments.)
    Let us recapitulate. Communication theory conceives of a symptom as a nonverbal message: It is not I who does not (or does) want to do this, it is something outside my control, e.g., my nerves, my illness, my anxiety, my bad eyes, alcohol, my upbringing, the Communists, or my wife.

The Level Structure of Communication
(Content and Relationship)

A couple in conjoint marriage therapy related the following incident. The husband, while alone at home, received a long-distance call from a friend who said he would be in the area for a few days. The husband immediately invited the friend to stay at their home, knowing that his wife would also welcome this friend and that, therefore, she would have done the same thing. When his wife came home, however, a bitter marital quarrel arose over the husband's invitation to the friend. As the problem was explored in the therapy session, both the husband and wife agreed that to invite the friend was the most appropriate and natural thing to do. They were perplexed to find that on the one hand they agreed and yet ..somehow disagreed on what seemed to be the same issue.
    In actual fact there were two issues involved in the dispute. One involved the appropriate course of action in a practical matter, that is, the invitation, and could be communicated digitally; the other concerned the relationship between the communicants--the question of who had the right to take initiative without consulting the other--and could not be so easily resolved digitally, for it presupposed the ability of the husband and wife to talk about their relationship. In their attempt to resolve their disagreement this couple committed a very common mistake in their communication: they disagreed on the metacommunicational (relationship) level, but tried to resolve the disagreement on the content level, where it did not exist, which led them into pseudo-disagreements. Another husband, also seen in conjoint therapy, managed to discover by himself and to state in his own words the difference between the content and the relationship levels. He and his wife had experienced many violent symmetrical escalations, usually based on the question of who was right regarding some trivial content matter. One day she was able to prove to him conclusively that he was factually wrong, and he replied, "Well, you may be right, but you are wrong because you are arguing with me." Any psychotherapist is familiar with these confusions between the content and relationship aspects of an issue, especially in marital communication, and with the enormous difficulty of diminishing the confusion. While to the therapist the monotonous redundancy of pseudo-disagreements between husbands and wives becomes evident fairly quickly, the protagonists usually see every one of them in isolation and as totally new, simply because the practical, objective issues involved may be drawn from a wide range of activities, from TV programs to corn flakes to sex. This situation has been masterfully described by Koestler: "Family relations pertain to a plane where the ordinary rules of judgment and conduct do not apply. They are a labyrinth of tensions, quarrels and reconciliations, whose logic is self-contradictory, whose ethics stem from a cozy jungle, and whose values and criteria are distorted like the curved space of a self-contained universe. It is a universe saturated with memories--but memories from which no lessons are drawn; saturated with a past which provides no guidance to the future. For in this universe, after each crisis and reconciliation, time always starts afresh and history is always in the year zero."
    The phenomenon of disagreement provides a good frame of reference for the study of disturbances of communication due to confusion between content and relationship. Disagreement can arise on the content or the relationship level, and the two forms are contingent upon each other. For instance, disagreement over the truth value of the statement "Uranium has 92 electrons" can apparently be settled only by recourse to objective evidence, e.g., a textbook of chemistry, for this evidence not only proves that the uranium atom does indeed have 92 electrons, but that one of the contestants was right and the other wrong. Of these two results, the first resolves the disagreement on the content level, and the other creates a relationship problem. Now, quite obviously, to resolve this new problem the two individuals cannot continue to talk about atoms; they must begin to talk about themselves and their relationship. To do this they must achieve a definition of their relationship as symmetrical or complementary: for example the one who was wrong may admire the other for his superior knowledge, or resent his superiority and resolve to be one-up on him at the next possible occasion in order to re-establish equality. Of course, if he could not wait until that next occasion, he could use the "to hell with logic" approach and try to be one-up by claiming that the figure 92 must be a misprint, or that he has a scientist friend who has just shown that the number of electrons is really quite meaningless, etc. A fine example of this technique is supplied by Russian and Chinese party ideologists with their hair-splitting interpretations of what Marx "really" meant in order to show what bad Marxists the others are. In such struggles words may eventually lose their last vestige of content meaning and become exclusively the tools of one-upmanship, as stated with admirable clarity by Humpty Dumpty:

"I don't know what you mean by 'glory,"' Alice said.
Humpty Dumpty smiled contemptuously. "Of course you don't--till I tell you. I meant 'there's a nice knock-down argument for you!"'
"But 'glory' doesn't mean 'a nice knock-down argument,"' Alice objected.
"When I use a word," Humpty Dumpty said, in a rather scornful tone, "it means just what I choose it to mean-neither more nor less."
"The question is," said Alice, "whether you can make words mean so many different things."
"The question is," said Humpty Dumpty, "which is to be master-that's all." (Last italics ours)

This, then, is merely another way of saying that in the face of their disagreement the two individuals have to define their relationship as either complementary or symmetrical.

Definition of Self and Other
Now suppose that the same statement about uranium is made by one physicist to another. A very different kind of interaction may arise from this, for most probably the other's response will be anger, hurt, or sarcasm-"I know you think I am a complete idiot, but I did go to school for a few years . . ." or the like. What is different in this interaction is the fact that here there is no disagreement on the content level. The truth value of the statement is not contested; in fact, the statement actually conveys no information since what it asserts on the content level is known to both partners anyway. It is this fact--the agreement on the content level--that clearly refers the disagreement to the relationship level, in other words, to the metacommunicational realm. There, however, disagreement amounts to something that is pragmatically far more important than disagreement on the content level. As we have seen, on the relationship level people do not communicate about facts outside their relationship, but offer each other definitions of that relationship and, by implication, of themselves.  These definitions have their own hierarchy of complexity. Thus, to take an arbitrary starting point, person P may offer the other, O, a definition of self. P may do this in one or another of many possible ways, but whatever and however he may communicate on the content level, the prototype of his metacommunication will be "This is how I see myself." It is in the nature of human communication that there are now three possible responses by O to P's self-definition, and all three of them are of great importance for the pragmatics of human communication.

Confirmation
O can accept (confirm) P's definition of self. As far as we can see, this confirmation of P's view of himself by O is probably the greatest single factor ensuring mental development and stability that has so far emerged from our study of communication. Surprising as it may seem, without this self-confirming effect human communication would hardly have evolved beyond the very limited boundaries of the interchanges indispensable for protection and survival; there would be no reason for communication for the mere sake of communication. Yet everyday experience leaves no doubt that a large portion of our communications are devoted precisely to this purpose. The vast gamut of emotions that individuals feel for each other--from love to hate--would probably hardly exist, and we would live in a world devoid of anything except the most utilitarian endeavors, a world devoid of beauty, poetry, play, and humor. It seems that, quite apart from the mere exchange of information, man has to communicate with others for the sake of his own awareness of self, and experimental verification of this intuitive assumption is increasingly being supplied by research on sensory deprivation, showing that man is unable to maintain his emotional stability for prolonged periods in communication with himself only. We feel that what the existentialists refer to as the encounter belongs here, as well as any other form of increased awareness of self that comes about as the result of working out a relationship with another individual. "In human society," writes Martin Buber, "at all its levels, persons confirm one another in a practical way, to some extent or other, in their personal qualities and capacities, and a society may be termed human in the measure to which its members confirm one another. . . . The basis of man's life with man is twofold, and it is one--the wish of every man to be confirmed as what he is, even as what he can become, by men; and the innate capacity of man to confirm his fellowman in this way. That this capacity lies so immeasurably fallow constitutes the real weakness and questionableness of the human race: actual humanity exists only where this capacity unfolds."

Rejection
The second possible response of O in the face of P's definition of himself is to reject it. Rejection, however, no matter how painful, presupposes at least limited recognition of what is being rejected and, therefore, does not necessarily negate the reality of P's view of himself. In fact, certain forms of rejection may even be constructive, as for instance a psychiatrist's refusal to accept a patient's definition of self in the transference situation in which the patient may typically try to impose his "relationship game" on the therapist. The reader is here referred to two authors who within their own conceptual frameworks have written extensively on this subject, Eric Berne and Jay Haley.

Disconfirmation
The third possibility is probably the most important one, from both the pragmatic and the psychopathological viewpoints. It is the phenomenon of disconfirmation, which, as we will see, is quite different from that of outright rejection of the other's definition of self. We are drawing here partly on the material presented by R.D.Laing of the Tavistock Institute of Human Relations in London, in addition to our own findings in the field of schizophrenic communication. Laing quotes William James, who once wrote: "No more fiendish punishment could be devised, even were such a thing physically possible, than that one should be turned loose in society and remain absolutely unnoticed by all the members thereof". There can be little doubt that such a situation would lead to "loss of self," which is but a translation of the term "alienation." Disconfirmation, as we find it in pathological communication, is no longer concerned with the truth or falsity--if there be such criteria--of P's definition of himself, but rather negates the reality of P as the source of such a definition. In other words, while rejection amounts to the message "You are wrong," disconfirmation says in effect "You do not exist." Or, to put it in more rigorous terms, if confirmation and rejection of the other's self were equated, in formal logic, to the concepts of truth and falsity, respectively, then disconfirmation would correspond to the concept of undecidability, which, as is known, is of a different logical order.
    Sometimes--admittedly rarely--literal undecidability may play an outstanding part in a relationship, as can be seen from the following transcript from a conjoint therapy session. The couple concerned had sought help because their sometimes violent quarrels left them deeply worried about their mutual failure as spouses. They had been married for twenty-one years. The husband was an eminently successful business man. At the beginning of this interchange, the wife had just remarked that in all these years she had never known where she stood with him.

Psychiatrist:  So what you are saying is that you don't get the clues from your husband that you need to know if you are performing well.
Wife:  No.
Psychiatrist: Does Dan criticize you when you deserve criticism--I  mean, positive or negative?
Husband: Rarely do I criticize her . . .
Wife: (overlapping): Rarely does he criticize.
Psychiatrist: Well, how--how do you know ...
Wife: (interrupting): He compliments you. (Short laugh.) You see, that is the befuddling thing ... Suppose I cook something and I burn it--well, he says it's really "very, very nice." Then, if I make something that is extra nice--well, it is "very, very nice." I told him I don't know whether something is nice--I don't know whether he is criticizing me or complimenting me. Because he thinks that by complimenting me he can compliment me into doing better, and when I deserve a compliment, he--he is always complimenting me--that's right, . . .so that I lose the value of the compliment.
Psychiatrist: So you really don't know where you stand with someone who always compliments . . .
Wife: (interrupting): No, I don't know whether he is criticizing me or really sincerely complimenting me.

What makes this example so interesting is that although both spouses are evidently fully aware of the pattern they are caught in, this awareness does not help them in the least to do something about it.
    To quote Laing: "The characteristic family pattern that has emerged from the study of disturbed families does not so much involve a child who is subject to outright neglect or even to obvious trauma, but a child whose authenticity has been subjected to subtle, but persistent, mutilation, often quite unwittingly. The ultimate of this is . . . when no matter how [a person] feels or how he acts, no matter what meaning he gives his situation, his feelings are denuded of validity, his acts are stripped of their motives, intentions and consequences, the situation is robbed of its meaning for him, so that he is totally mystified and alienated."
    And now a specific example that has been published in greater detail elsewhere. It is taken from a conjoint psychotherapy session with an entire family composed of the parents, their twenty-five year-old son Dave (who was first officially diagnosed schizophrenic while in military service at age twenty and had afterward lived at home until about a year before this interview, when he had been hospitalized), and their eighteen-year-old son, Charles. When the discussion focused on how the patient's weekend visits strained the family, the psychiatrist pointed out that it seemed as if Dave were being asked to bear the intolerable burden of the whole family's solicitude. Dave thus became the sole indicator of how well or poorly things went over the weekend. Surprisingly, the patient immediately took up this point:

Dave:  Well, I feel that sometimes my parents, and Charles also, are very sensitive to how I might feel, maybe overly sensitive about how I feel, cause I don't--I don't feel I raise the roof when I go home, or . . .
Mother:  Mhm. Dave, you haven't been like that either since you had your car, it's just--but before you did.
Dave:  Well, I know I did . . .
Mother:  (overlapping): Yeah, but even--yeah, lately, the last twice since you had your car.
Dave:  Yeah, OK, anyhow, ah (sigh), that's-ah, I wish 1 didn't have to be that way, I guess, 'it'd be nice if I could enjoy myself or somethin' (sighs; pause)
Psychiatrist: You know, you change your story in mid-stream when your mother is nice to you. Which . . . is understandable, but in your position you just can't afford to do it.
Dave: (overlapping): Mhm.
Psychiatrist: It makes you kookier. Then you don't even know what you're thinking.
Mother: What did he change?
Psychiatrist: Well, I can't read his mind so I don't know what he was going to say precisely--I have a general idea, I think, just from experience . . .
Dave: (interrupting): Well, it's just, just the story that I'm the sick one in the family and so this gives everyone else a-a chance to be a good Joe and pick up Dave's spirits whether Dave's spirits are necessarily down or not. That's what it amounts to sometimes, I feel. In other words, I can't be anything but myself, and if people don't like me the way they am--ah, the way I am--then I appreciate when they--tell me or something, is what it amounts to.

    The patient's slip of the tongue illuminates his dilemma: he says "I can't be anything but myself," but the question remains, is myself "I" or "they"? Simply to call this an evidence of "weak ego boundaries" or the like ignores the interactional fact of disconfirmation just presented, not only in Dave's report on his weekend visits but by the mother's immediate disconfirmation in the present example (statements 1-5) of the validity of Dave's impression. In the light of both present and reported disconfirmation of his self, the patient's slip emerges in a new aspect.

Levels of Interpersonal Perception
At last we are ready to return to the hierarchy of messages that is found when analyzing communications on the relationship level. We have seen that P's definition of self ("This is how I see myself . . .") can be met with one of three possible responses by O: confirmation, rejection, or disconfirmation. Now, these three responses have one common denominator, i.e., through any one of them O communicates "This is how I am seeing you."
    There is, then, in the discourse on the metacommunicational level, a message from P to O: "This is how I see myself." It is followed by a message from O to P: "This is how I am seeing you." To this message P will respond by a message asserting, among other things, "This is how I see you seeing me," and O in turn by the message "This is how I see you seeing me seeing you." As already suggested, this regress is theoretically infinite, while for practical purposes it must be assumed that one cannot deal with messages of a higher order of abstraction than the last one mentioned. Now, it should be noted that any one of these messages can be subjected by the recipient to the same confirmation, rejection, or disconfirmation described above, and that the same holds, of course, for O's definition of self and the ensuing simultaneous metacommunicational discourse with P. This leads to communicational contexts whose complexity easily staggers the imagination and yet which have very specific pragmatic consequences.

Imperviousness
Not very much is as yet known about these consequences, but very promising research in this area is being carried out by Laing, Phillipson and Lee, who have given us permission to quote here some of their results from an unpublished paper. Disconfirmation of self by the other is mainly the result of a peculiar unawareness of interpersonal perceptions, called imperviousness and defined by Lee as follows:

What we are concerned with is the aspect of awareness and unawareness. For smooth, adequate interaction to occur, each party must register the other's point of view. Since interpersonal perception goes on on many levels, so, too, can imperviousness go on on many levels. For there exists for each level of perception a comparable and analogous level of possible imperception or imperviousness. Where a lack of accurate awareness, or imperviousness, exists, the parties in a dyad relate about pseudo-issues. . . . They attain an assumed harmony which does not exist, or argue over assumed disagreements that similarly do not exist. It is this which I find to be the characteristic situation within a distrubed family: they are constantly building harmonious relationships on the shifting sands of pseudo-agreements or else have violent arguments on the basis of pseudo-disagreements.

Lee then goes on to show that imperviousness can exist on the first level of the hierarchy, that is, to P's message "This is how I see myself" O responds, "This is how I see you," in a way which is not congruent with P's self-definition. P may then conclude that O does not understand (or appreciate, or love) him while O, on the other hand, may assume that P feels understood (or appreciated, or loved) by him (O). In this case, O does not disagree with P, but ignores or misinterprets P's message, and thus is consistent with our definition of disconfirmation. A second-level imperviousness can be said to exist when P does not register that his message has not gotten through to O; that is, P does not convey accurately "This is how I see you seeing [in this case, misunderstanding] me." At this level, then, imperviousness to imperviousness occurs.
    From their study of disturbed families, Lee describes an important conclusion about the pragmatics of this kind of communication:

The typical pattern is that the parental imperviousness exists on level No. 1, while the child's imperviousness exists on level NO. 2. That is, typically the parent fails to register his child's view, while the child does not register that his view has not been (and perhaps cannot be) registered.
    Most often the parent appears to remain impervious to the child's view because he feels it is uncomplimentary to him, or because it does not fit his value system. That is, the parent insists that the child does believe what he (the parent) feels the child "should" believe. The child, in turn, fails to recognize this. He believes that his message has gotten through and has been understood, and acts accordingly. In such a situation he is bound to be confused by the subsequent' interaction. He feels as if he continuously runs into an invisible solid glass wall. This results in his experiencing a continuous sense of mystification which leads to dismay and eventually to despair. Ultimately he feels that life just does not make any sense.
    Such a child, during the course of therapy, finally realized this state of affairs, and stated his dilemma this way: "Whenever I disagree with my mother she seems to say to herself, 'oh, I know what you are saying out loud, but I know that isn't what you really think inside,' and then she proceeds to forget what I have just said."

A rich variety of clinical illustrations of imperviousness at the relationship level as just described can be found in Laing and Esterson:

The Punctuation of the Sequence of Events
"He laughed because he thought that they could not hit him--he did not imagine that they were practicing how to miss him"- Bertoldt Brecht

A few examples of the potential complications inherent in this phenomenon have already been presented in the preceding chapter. They show that unresolved discrepancies in the punctuation of communicational sequences can lead directly into interactional impasses in which eventually the mutual charges of madness or badness are proffered.  Discrepancies in the punctuation of sequences of events occur, of course, in all those cases in which at least one of the communicants does not possess the same amount of information as the other but does not know this. A simple example of such a sequence would be the following: P writes a letter to O proposing a joint venture and inviting O's participation. O replies in the affirmative, but the letter is lost in the mail. After a while P concludes that O is ignoring his invitation and resolves to disregard him in turn. O, on the other hand, feels offended that his answer is ignored and also decides not to contact P any more. From this point their silent feud may last forever, unless they decide to investigate what happened to their communications, that is, unless they begin to metacommunicate. Only then will they find out that P did not know O had replied, while O did not know that his reply had never reached P. As can be seen, in this example a fortuitous outside event interfered with the congruency of punctuation.
    One of the authors experienced this phenomenon of discrepant punctuation when he once applied for an assistantship with a psychiatric research institute. At the appointed hour he reported to the director's office for his interview and the following conversation took place with the receptionist:

Visitor:  Good afternoon, I have an appointment with Dr. H. My name is Watzlawick [VAHTsla-vick].
Receptionist:  I did not say it was.
Visitor:  (taken aback and somewhat annoyed): But I am telling you it is.
Receptionist:  (bewildered): Why then--did you say it wasn't?
Visitor:  But I said it was!

At this point the visitor was "certain" that he was being made the object of some incomprehensible but disrespectful joke, while, as it turned out, the receptionist had by then decided that the visitor must be a new psychotic patient of Dr. H's. Eventually it became clear that instead of "My name is Watzlawick" the receptionist had understood "My name is not Slavic," which, indeed, she had never said it was. It is interesting to see how even in this brief interchange in a rather impersonal context the discrepant punctuation, here due to a verbal misunderstanding, immediately led to mutual assumptions of badness and madness.
    Generally speaking, it is gratuitous to assume not only that the other has the same amount of information as oneself but that the other must draw the same conclusions from this information. Communication experts have estimated that a person receives ten thousand sensory impressions (exteroceptive and proprioceptive) per second. Obviously, then, a drastic selection process is necessary to prevent the higher brain centers from being swamped by irrelevant information. But the decision about what is essential and what is irrelevant apparently varies from individual to individual and seems to be determined by criteria which are largely outside individual awareness. In all probability, reality is what we make it or, in Hamlet's words, ". . . there is nothing either good or bad, but thinking makes it so." We can only speculate that at the root of these punctuation conflicts there lies the firmly established and usually unquestioned conviction that there is only one reality, the world as I see it, and that any view that differs from mine must be due to the other's irrationality or ill will. So much for our speculations. What we can observe in virtually all these cases of pathological communication is that they are vicious circles that cannot be broken unless and until communication itself becomes the subject of communication, in other words, until the communicants are able to metacommunicate. But to do this they have to step outside the circle, and this necessity to step outside a given contingency in order to resolve it will be a recurrent theme in later parts of this book.

Cause And Effect
We typically observe in these cases of discrepant punctuation a conflict about what is cause and what is effect, when in actual fact neither of these concepts is applicable because of the circularity of the ongoing interaction. To return once more to Joad's example, we can see that nation A arms because it feels threatened by nation B (that is, A sees its own behavior as the effect of B's), while nation B calls A's armaments the cause of its own "defensive" measures. Richardson points to essentially the same problem as he describes the arms race that began to escalate about 1912:

The war-like preparations of the Entente and of the Alliance were both increasing. The usual explanation was then, and perhaps still is, that the motives of the two sides were quite different, for we were only doing what was right, proper and necessary for our own defense, whilst they were disturbing the peace by indulging in wild schemes and extravagant ambitions. There are several distinct contrasts in that omnibus statement. Firstly that their conduct was morally bad, ours morally good. About so national a dispute it would be difficult to say anything that the world as a whole would accept. But there is some other alleged contrast as to which there is some hope of general agreement. It was asserted in the years 1912-14 that their motives' were fixed and independent of our behaviour whereas our motives were a response to their behaviour and were varied accordingly.

From the pragmatic viewpoint there is little if any difference between the interactions of nations or of individuals once discrepant punctuation has led to different views of reality, including the nature of the relationship, and thus into international or interpersonal conflict. The following example shows the same pattern at work on the interpersonal level:

Husband:  (to therapist): From long experience I know that if I want peace at home I must not interfere with the way she wants things done.
Wife:  That is not true--I wish you showed a little more initiative and did decide at least something every once in a while, because . .
Husband:   (interrupting): You'd never let me do this!
Wife:  I'd gladly let you--only if I do, nothing ever happens, and then I have to do everything at the last moment.
Husband:  (to therapist): Can you see? Things can't be taken care of if and when they come up--they have to be planned and organized a week ahead.
Wife:  (angrily): Give me one example in the last few years when you did do something.
Husband:  I guess I can't--because it is better for everybody, including the children, if I let you have your own way. I found this out very early in our marriage.
Wife:  You have never behaved differently, right from the start you didn't--you have always left everything up to me!
Husband: For heaven's sake, now listen to this (pause, then to therapist)--I guess what she is talking about now is that I would always ask her what she wanted--like "where would you like to go tonight?" or "what would you like to do over the weekend?" and instead of seeing that I wanted to be nice to her, she would get mad at me . . .
Wife:  (to therapist): Yeah, what he still doesn't understand is that if you get this "anything-you-want-dear-is-all-right-with-me" stuff month after month, you begin to feel that nothing you want matters to him. . . .

The same mechanism is contained in an example reported by Laing and Esterson, involving a mother and her emotionally disturbed daughter. Shortly before her hospitalization the daughter had made a very ineffectual physical attack on her mother.

Daughter: Well, why did I attack you? Perhaps I was looking for something, something lacked--affection, maybe it was greed for affection.
Mother:  You wouldn't have any of that. You always think it's soppy.
Daughter: Well, when did you offer it to me?
Mother:  Well, for instance if I was to want to kiss you, you'd say, "Don't be soppy."
Daughter:  But I've never known you let me kiss you.

This leads to the important concept of the self-fulfilling prophecy which, from the interactional viewpoint, is perhaps the most interesting phenomenon in the area of punctuation. A self-fulfilling prophecy may be regarded as the communicational equivalent of "begging the question." It is behavior that brings about in others the reaction to which the behavior would be an appropriate reaction. For instance, a person who acts on the premise that "nobody likes me" will behave in a distrustful, defensive, or aggressive manner to which others are likely to react unsympathetically, thus bearing out his original premise. For the purposes of the pragmatics of human communication, it is again quite irrelevant to ask why a person should have such a premise, how it came about, and how unconscious he may be of it. Pragmatically we can observe that this individual's interpersonal behavior shows this kind of redundancy, and that it has a complementary effect on others, forcing them into certain specific attitudes. What is typical about the sequence and makes it a problem of punctuation is that the individual concerned conceives of himself only as reacting to, but not as provoking, those attitudes.

Errors in the "Translation" Between Analogic and Digital Material

In trying to describe these errors, an anecdote from Daniele Vard's novel The Gate of Happy Sparrows comes to mind. The hero, a European living in Peking during the twenties, receives lessons in Mandarin script from a Chinese professor and is asked to translate a sentence composed of three characters that he correctly deciphers as the signs for "rotundity," "sitting," and "water." In his attempt to combine these concepts into an affirmative statement (into digital language, as we would say) he decides on "Somebody is taking a sitz bath," much to the disdain of the distinguished professor, for the sentence is a particularly poetic reference to a sunset at sea.
    Like Chinese writing, analogic message material, as already mentioned, lacks many of the elements that comprise the morphology and syntax of digital language. Thus, in translating analogic into digital messages, these elements have to be supplied and inserted by the translator, just as in dream interpretation digital structure has to be introduced more or less intuitively into the kaleidoscopic imagery of the dream.
    Analogic message material, as we have seen, is highly antithetical; it lends itself to very different and often quite incompatible digital interpretations. Thus not only is it difficult for the sender to verbalize his own analogic communications, but if interpersonal controversy arises over the meaning of a particular piece of analogic communication, either partner is likely to introduce, in the process of translation into the digital mode, the kind of digitalization in keeping with his view of the nature of the relationship. The bringing of a gift, for instance, is undoubtedly a piece of analogic communication. However, depending on the recipient's view of his relationship with the giver, he can see it as a token of affection, a bribe, or a restitution. Many a husband is dismayed to find himself suspected of an as yet unconfessed guilt if he breaks the rules of their marriage "game" by spontaneously presenting his wife with a bunch of flowers.
    What is the digital meaning of growing pale, trembling, sweating, and stammering when displayed by a person under interrogation? It may be the ultimate proof of his guilt, or it may merely be the behavior of an innocent person going through the nightmarish experience of being suspected of a crime and realizing that his fear may be interpreted as guilt. Psychotherapy is undoubtedly concerned with the correct and the corrective digitalization of the analogic; in fact, the success or failure of any interpretation, will depend both ori the therapist's ability to translate from the one mode to the other and on the patient's readiness to exchange his own digitalization for more appropriate and less distressing ones.  Even where the translation appears to be adequate, digital communication on the relationship level may remain curiously unconvincing.  This fact is caricatured in the following "Peanuts" cartoon:

    In an unpublished report, Bateson hypothesizes that another of the basic mistakes made when translating between the two modes of communication is the assumption that an analogic message is by nature assertive or denotative, just as digital messages are. There is, however, good reason to believe that this is not so. He writes:

When one octopus or one nation puts on a threatening gesture, the other might conclude "he is strong" or "he will fight," but this was not in the original message. Indeed, the message itself is non-indicative and may be better regarded as analogous to a proposal or a question in the digital world.

    In this connection it should be remembered that all analogic messages are invocations of relationship, and that they are therefore proposals regarding the future rules of the relationship, to use another of Bateson's definitions. By my behavior, Bateson suggests, I can mention or propose love, hate, combat, etc., but it is up to you to attribute positive or negative future truth value to my proposals. This, needless to say, is the source of countless relationship conflicts.
    Digital language, as explained in the preceding chapter, has a logical syntax and is therefore eminently suited for communication on the content level. But in translating analogic into digital material, logical truth functions must be introduced, which are absent in the analogic mode. This absence becomes most conspicuous in the case of negation, where it amounts to the lack of the digital "not." In other words, while it is simple to convey the analogic message "I shall attack you," it is extremely difficult to signal "I will not attack you," just as it is difficult if not impossible to introduce negatives into analogue computers.
    In Koestler's novel Arrival and Departure, the hero, a young man who escaped from his Nazi-occupied homeland and whose face has been disfigured by torture, is in love with a beautiful girl. He has no hope that she will reciprocate his feelings, and all he wants is to be with her and stroke her hair. She resists these innocent advances, thereby arousing both his desperation and his passion, until he overpowers her.

She lay turned to the wall, her head in a strangely twisted position, like a doll's head with a broken neck. And now at last he could caress her hair, gently, soothingly, as he had always meant to. Then he realized that she was crying, her shoulders shaking in dry, soundless sobs. He went on fondling her hair and shoulders and muttered:
"You see, you wouldn't listen to me."
She suddenly lay rigid, interrupting her sobbing:
"What did you say?"
"I said all I wanted was that you shouldn't go away and that you should allow me to stroke your hair and to give you iced drinks . . . Really, that was all I meant."
Her shoulders shook in a slightly hysterical laughter. "By God, you are the biggest fool I have ever seen."
"Are you angry with me? Don't. I didn't mean to."
She drew her knees up, shrinking away from him, curling up against the wall. "Leave me alone. Please go away and leave me quiet for a while." She cried again, more quietly this time. He slid down from the couch, squatting on the carpet as before, but he got hold of one of her hands which lay limply on the cushion. It was a lifeless, humid hand, hot with fever.
"You know," he said, encouraged because she didn't withdraw her hand, "when I was a child we had a black kitten which I always wanted to play with, but she was too frightened and always ran away. One day with all sorts of cunning I got her into the nursery, but she hid under the cupboard and wouldn't come out. So I dragged the cupboard away from the wall, and got more and more angry because she wouldn't let me fondle her, and then she hid under the table and I upset the table and broke two pictures on the wall and turned the whole room upside down and chased the kitten with a chair all around the room. Then my mother came in and asked me what I was doing, and I told her I only wanted to fondle that stupid kitten, and I got a terrible thrashing. But I had told the truth . .   ."

Here the desperation of being rejected and unable to prove that he does not mean to harm leads to violence.
    Now if, as Bateson did, one watches animal behavior for such contingencies, one finds that the only solution to this problem of signaling negation lies first in demonstrating or proposing the action to be denied, and then in not carrying it to its conclusion. This interesting and only apparently "irrational" behavior can be observed not only in animal interaction but on the human level as well.
    We have observed a very interesting communication pattern for the establishment of trust relationships between humans and bottlenosed porpoises. While this may be a ritual developed "privately" by only two of the animals, it still provides an excellent example for the analogic communication of "not." The animals had obviously concluded that the hand is one of the most important and vulnerable parts of the human body. Each would seek to establish contact with a stranger by taking the human's hand into his mouth and gently squeezing it between his jaws, which have sharp teeth and are powerful enough to bite the hand off cleanly. If the human would submit to this, the dolphin seemed to accept it as a message of complete trust. His next move was to reciprocate by placing the forward ventral portion of his body (his most vulnerable part, roughly equivalent in location to the human throat) upon the human's hand, leg, or foot, thereby signaling his trust in the friendly intentions of the human. This procedure is, however, obviously fraught with possible misinterpretation at every step.
    On a poetic level, an essentially similar form of relationship, here between man and the transcendental, is expressed in the opening lines of Rilke's first Duino Elegy, where beauty is experienced as the negation of inherent, ever possible destruction:

Who, if I cried would hear me among the angelic orders? And even if one of them suddenly pressed me against his heart, I should fade in the strength of his stronger existence. For Beauty's nothing but the beginning of Terror we are still just able to bear, and we adore it so because it serenely disdains to destroy us. (italics ours)

    As the dolphin example suggests, ritual may be the intermediary process between analogic and digital communication, simulating the message material but in a repetitive and stylized manner that hangs between analogue and symbol. Thus we can observe that animals such as cats routinely establish a complementary but nonviolent relationship through the following ritual. The "one-down" animal (usually the younger or the one outside his own territory) throws himself on his back, exposing his jugular vein, which is taken in the jaw of the other cat with impunity. This method of establishing an "I shall not attack you" relationship seems to be understood by both; what is even more interesting, this coding has been seen to be successful in interspecies (e.g., cats and dogs) communication as well. Analogic materials are often formalized in the rituals of human societies, and as such material is canonized it approaches symbolic or digital communication, revealing a curious overlap.
    On a pathological plane, the same mechanism seems to be operative in sexual masochism. It would appear that the message "I shall not destroy you" is only convincing (and only allays, at least temporarily, the masochist's deep fear of terrible punishment) by means of the analogic denial inherent in the ritual of humiliation and punishment that he knows will eventually but certainly stop short of the imagined dread.
    Those familiar with symbolic logic may by now appreciate that it is probably not necessary to prove the absence of all logical truth functions in analogic material but only a critical few. The logical truth function alternation (nonexclusive or), construed to mean "either one or both," can be seen to be similarly absent from analogic language. While it is easy to convey the meaning "one or the other or both will do" in digital language, it is not immediately obvious how this logical relation could be inserted into analogic material; indeed, it probably cannot. Symbolic logicians have pointed out that to represent all the major truth functions (negation, conjunction, alternation, implication, and equivalence), two--negation and alternation (or, similarly, negation and conjunction)--are sufficient and, of the five, necessary to represent the remaining three. According to this reasoning, although we know almost nothing specific about the pragmatic importance of the absence of the other truth functions in analogic material, we can conclude that since these are but variations of "not" and "or" they will not escape similar difficulties of translation.
    Bateson and Jackson have hypothesized the importance of analogic versus digital coding in hysterical symptom formation. According to them, a converse process from those we have been discussing takes place, a retranslation, as it were, from already digitalized message material back to the analogic mode:

A converse--but much more complex--problem arises in regard to hysteria. No doubt this word covers a wide range of formal patterns, but it would appear that at least some cases involve errors of translation from the digital to the analogic. Stripping the digital material of its logical type markers leads to erroneous symptom formation. The verbal "headache" which was invented as a conventional excuse for not performing some task may become subjectively real and be endowed with real magnitudes in the pain dimension.

    If we bear in mind that the first consequence of a breakdown in communication is usually a partial loss of the ability to metacommunicate digitally about the contingencies of the relationship, this "return to the analogic" appears as a plausible compromise solution.  Again, there is little difference between the behavior of individuals and nations. When serious tension arises between two countries the customary step is to break off diplomatic relations, and consequently to resort to analogic communications like mobilizations, troop concentrations, and other analogic messages of this kind. What is so absurd about this procedure is that digital communication (diplomatic procedure) is broken off at the exact moment when it is more desperately needed than ever before. The "hot line" between Washington and Moscow may be prophylactic in this regard, even though its official rationale is only that of speeding up communications in times of crisis.
    The symbolic nature of conversion symptoms and, generally, their affinity with dream symbolism has been realized since the days of Libault, Bernheim, and Charcot. And what is a symbol if not the representation, in real magnitudes, of something that is essentially an abstract function, an aspect of a relationship. Throughout his work, C. G. Jung shows that the symbol appears where what we would call "digitalization" is not yet possible. But it seems to us that symbolization also takes place where digitalization is no longer possible and that this typically happens when a relationship threatens to grow into socially or morally tabooed areas such as incest.

Potential Pathologies of Symmetrical and Complementary Interaction

To avoid a frequent misunderstanding, it cannot be emphasized too strongly that symmetry and complementarity in communication are not in and by themselves "good" or "bad...... normal" or "abnormal," etc. The two concepts simply refer to two basic categories into which all communicational interchanges can be divided. Both have important functions, and from what is known about healthy relationships we may conclude that both must be present, although in mutual alternation or operation in different areas. As we will try to show, this means that each pattern can stabilize the other whenever a runaway occurs in one of them, and also that it is not only possible but necessary for two partners to relate symmetrically in some areas and complementarily in others.

Symmetrical Escalation
Like any other pattern of communication, these two have their potential pathologies, which will first be described and then illustrated with clinical material. We have already suggested that in a symmetrical relationship there is an ever-present danger of competitiveness. As can be observed both in individuals and in nations, equality seems to be most reassuring if one manages to be just a little "more equal" than others, to use Orwell's famous phrase. This tendency accounts for the typical escalating quality of symmetrical interaction once its stability is lost and a so-called runaway occurs, e.g., quarrels and fights between individuals or wars between nations. In marital conflicts, for instance, it is easy to observe how the spouses go through an escalating pattern of frustration until they eventually stop from sheer physical or emotional exhaustion and maintain an uneasy truce until they have recovered enough for the next round. Pathology in symmetrical interaction is thus characterized by more or less open warfare, or schism, in Lidz's sense.
    In a healthy symmetrical relationship the partners are able to accept each other in their respective "suchness," which leads to mutual respect and trust in the other's respect and amounts to realistic, reciprocal confirmation of their selves. If and when a symmetrical relationship breaks down, we habitually observe the rejection rather than disconfirmation of the other's self.

Rigid Complementarity
In complementary relationships, there can be the same healthy, positive confirmation of each other. The pathologies of complementary relationships, on the other hand, are quite different, and tend to amount to disconfirmations rather than rejections of the other's self. They are, therefore, more important from a psychopathological point of view than the more or less open fights in symmetrical relations.
    A typical problem arises in a complementary relationship when P demands that O confirm a definition of P's self that is at variance with the way O sees P. This places O in a very peculiar dilemma: he must change his own definition of self into one that complements and thus supports P's, for it is in the nature of complementary relationships that a definition of self can only be maintained by the partner's playing the specific complementary role. After all, there can be no mother without a child. But the patterns of a mother-child relationship change with time. The same pattern that is biologically and emotionally vital during an early phase of the infant's life becomes a severe handicap for his further development, if adequate change is not allowed to take place in the relationship. Thus, depending on the context, the same pattern may be highly self-confirming at one time and disconfirming at a later (or premature) stage in the natural history of a relationship. Because of their greater psychiatric flamboyance, the pathologies of complementary relationships have been given more attention in the literature than their symmetrical counterparts. Psychoanalysis refers to them as sadomasochistic and views them as the more or less fortuitous liaison of two individuals whose respective deviant character formations dovetail with each other. Among more recent and more interaction-oriented studies are Lidz's concept of marital skew, Scheflen's paper on the "gruesome twosome", and the concept of "collusion" in Laing's sense. In these relationships we observe a growing sense of frustration and despair in one or both partners. Complaints of increasingly frightening feelings of self-estrangement and depersonalization, of abulia as well as compulsive acting-out are very frequently voiced by individuals who outside their homes (or otherwise in the absence of their partners) are perfectly capable of functioning satisfactorily, and who, when interviewed individually, may appear very well adjusted. This picture often changes dramatically when they are seen together with their "complements." The pathology of their relationship then becomes patent. Perhaps the most remarkable study of the pathology of complementary relationships is the famous paper, "La folie a deux," written by two French psychiatrists nearly a hundred years ago. How small a claim we have to the originality of our approach is, for instance, documented by the following passages from this paper. The authors first describe the patient and then continue:

The above description belongs to the insane person, the agent who provokes the situation in "delire a deux." His associate is a much more complicated person to define and yet careful research will teach one to recognize the laws which are obeyed by this second party in communicated insanity. . . . Once the tacit contract that ties both lunatics is almost settled, the problem is not only to examine the influence of the insane on the supposedly sane man, but also the opposite, the influence of the rational on the deluded one, and to show how through mutual compromises the differences are eliminated. (italics ours)

    As already mentioned briefly at the beginning of this section, symmetrical and complementary relationship patterns can stabilize each other, and changes from the one to the other pattern and back again are important homeostatic mechanisms. This entails a therapeutic implication, namely that, at least in theory, therapeutic change can be brought about very directly by the introduction of symmetry into complementarity or vice versa during treatment. We say "at least in theory" advisedly, for it is only too well known how difficult it is in practice to induce any sort of change in rigidly defined systems whose participants, it seems, would "rather bear those ills we have than fly to others that we know not of."
    To explain the foregoing, here are three exerpts from the so-called Structured Family Interviews. All three of them are in reply to the interviewer's standard question to the spouses: "How, of all the millions of people in the world, did the two of you get together?" It should be made quite clear that the actual historic information contained in such an account is of only secondary importance, although it may be relatively accurate and may itself portray a symmetrical or complementary interaction which took place at that time. But it is not this historic information, often distorted by selective recall and wishful thinking, that is of interest here. Thus in considering the first couple one is struck by the symmetry of their interaction while responding to the interviewer's question. The story of their meeting as told by them, is only the raw material, so to speak, which they manipulate in accordance with the rules of their game of "one-upmanship." For them, and for us, it is not important what happened, but rather who has the right to say what to, and about, the other. In other words, what is of the essence is not the content but the relationship aspect of their communication.

I) The first is an example of a typical symmetrical interchange.

    This couple sought help because they feared that their constant bickering might hurt their children. As could almost be predicted from the above excerpt, they also mentioned difficulties in their sexual relation, where, of course, their inability to relate complementarily made itself particularly felt.

2) The couple in the next example participated in a research project involving randomly selected families. It was generally felt by the investigators that they were emotionally quite distant and that the wife showed a good deal of depression. Their interaction is typically complementary, with husband in the "one-up" and wife in the "one-down" position. But, as already explained in the previous chapter, these terms must not be taken as indicators of relative strength or weakness. Quite obviously, this woman's amnesia and helplessness make it not only possible for him to play the role of the strong, realistic male, but they are also the very factors against which his strength and his realism are quite powerless. Thus we are again confronted by the interpersonal impact of any emotional symptom in the wider sense.
    The excerpt starts a little after the interviewer has asked the standard question about their meeting and after the husband explained that she had come to work in an office next to his.

Husband: And--see, when'd you start there?
Wife: W--I haven't any i-
Husband (interrupting):-seems to me it was about, I came in October, the year before . . . and you probably started about . . . February uh, January or February--probably February or March 'cause your birthday was in December, that same year.
Wife: Mm, I don't even remember . . .
Husband (interrupting): So I happened to send her some flowers, you see, when our first date out. And that never--we'd never gone anywhere had we?
Wife: (with short laugh): No, I was very surprised.
Husband And we just went from there. It was about a year later I guess we got married. Little over a year.
Interviewer: What did you . . .
Husband (interrupting): Although Jane left the company very shortly after that. Mm, I don't think you worked there over a couple of months, did you?
Wife: You know, I'm sorry, I don't remember a thing about (slight laugh) how long it was or when I went--
Husband (interrupting): Yeah, a couple of months, and then you went back into teaching. (Wife: Mhm, mhm) 'Cause we--she found I guess that this war work was not contributing as much to the war effort as she thought it was, when she went out there.
Interviewer: So you--you went to a school?
Wife: Yes, I'd been working in it, before (Interviewer: Mhm) I went to work there.
Interviewer: And you continued the contact without interruption.  (Husband: Oh yeah) What, uh, beside the fact that your wife is obviously attractive, what else do you think you have in common?
Husband: Absolutely nothing. (Laughing) We never have--had 'r we--(sharp breath). (Pause)

3) The third example is taken from the interview of a clinically normal couple who volunteered for the same type of interview. Here it can be seen how they manage to maintain a warm and mutually supportive relationship by a flexible alternation of symmetrical with complementary interchanges. Thus, even though some of the details of their account could conceivably be felt to be depreciatory of each other, these do not seem to threaten the stability of their relationship and the mutual confirmation of their roles.

   An entirely different communicational contingency arises in the area of symmetrical and complementarv interaction if a message defines the relationship as symmetrical and complementary at the same time. This is probably the most frequent and important way by which paradox can enter into human communication, and the pragmatic effects of this form of communicational inconsistency will, therefore, be taken up separately in Chapter 6.
    There are two points to be emphasized in the analysis of the preceding examples. First, content fades in importance as communicational patterns emerge. A group of second- and third-year psychiatric residents rated the couple in the third example as much .,sicker" than other, clinically disturbed, couples. Upon inquiry, it was obvious that the basis of their judgment had been the relative social unacceptability of the meeting and the open "sparring" about details. In other words their erroneous judgment was based on content rather than on the interaction of their account.
    More important, it should have become obvious that our analysis was of successive statements. No given statement in isolation can be symmetrical, complementary one-up, or whatever. It is the response of the partner that is of course necessary for the "classification" of a given message. That is, it is not in the nature of any of the statements as individual entities, but in the relation between two or more responses that the functions of communication are defined.