GAMES PEOPLE PLAY.

ERIC BERNE.

Dr Eric Berne, like his father, graduated from the Faculty of Medicine, McGill University, Montreal, Canada, and later moved to the United States, where he had his internship in psychiatry at the New Haven Hospital and the Yale Inst.i.tute of Human Relations. He also studied at the New York and San Francisco Psychoa.n.a.lytic Inst.i.tutes. During the Second World War he served in the U.S. Army Medical Corps and was discharged with the rank of major. Afterwards he became a consultant in psychiatry and neurology to the Surgeon General of the U.S. Army. He was a practising psychiatrist in Carmel and San Francisco, California, a lecturer at the University of California Medical School, and Consultant in Group Therapy at McAuley Clinic in San Francisco. He was a corresponding member of the Indian Psychiatric Society, chairman of the board of trustees of the International Transactional a.n.a.lysis a.s.sociation and editor of the Transactional a.n.a.lysis Bulletin. He also wrote The Mind in Action, Transactional a.n.a.lysis in Psychotherapy, The Structure and Dynamics of Organizations and Groups and Principles of Group Treatment. A Layman's Guide to Psychiatry and Psychoa.n.a.lysis and s.e.x in Human Loving are also available in Penguins. While at college he wrote for the Canadian magazine Forum, the London Adelphi, and other periodicals. He died in 1970.

To my patients and students who taught me more and more and are still teaching me about games and the meaning of life.

Preface.

THIS book is primarily designed to be a sequel to my book Transactional a.n.a.lysis in Psychotherapy,1 but has been planned so that it can be read and understood independently. The theory necessary for the a.n.a.lysis and clear understanding of games has been summarized in Part I. Part II contains descriptions of the individual games. Part III contains new clinical and theoretical material which, added to the old, makes it possible to understand to some extent what it means to be game-free. Those desiring further background are referred to the earlier volume. The reader of both will note that in addition to the theoretical advances, there have been some minor changes in terminology and viewpoint based on further thinking and reading and new clinical material.

The need for this book was indicated by interested requests from students and lecture audiences for lists of games, or for further elaboration of games mentioned briefly as examples in a general exposition of the principles of transactional a.n.a.lysis. Thanks are due in general to these students and audiences, and especially to the many patients who exposed to view, spotted or named new games; and in particular to Miss Barbara Rosenfeld for her many ideas about the art and meaning of listening; and to Mr Melvin Boyce, Mr Joseph Concannon, Dr Franklin Ernst, Dr Kenneth Everts, Dr Gordon Gritter, Mrs Frances Matson, and Dr Ray Poindexter, among others, for their independent discovery or confirmation of the significance of many games.

Mr Claude Steiner, formerly Research Director of the San Francisco Social Psychiatry Seminars and presently in the Department of Psychology at the University of Michigan deserves special mention on two counts. He conducted the first experiments which confirmed many of the theoretical points at issue here, and as a result of these experiments he helped considerably in clarifying the nature of autonomy and of intimacy. Thanks are also due to Miss Viola Litt, the Secretary-Treasurer of the Seminars, and to Mrs Mary N. Williams, my personal secretary, for their continued help, and to Anne Garrett for her a.s.sistance in reading the proof.

SEMANTICS.

For conciseness, the games are described primarily from the male point of view unless they are clearly feminine. Thus the chief player is usually designated as 'he', but without prejudice, since the same situation, unless otherwise indicated, could as easily be outlined with 'she', mutatis mutandis. If the woman's role differs significantly from the man's, it is treated separately. The therapist is similarly without prejudice designated as 'he'. The vocabulary and viewpoint are primarily oriented toward the practising clinician, but members of other professions may find this book interesting or useful.

Transactional game a.n.a.lysis should be clearly distinguished from its growing sister science of mathematical game a.n.a.lysis, although a few of the terms used in the text, such as 'payoff', are now respectably mathematical. For a detailed review of the mathematical theory of games see Games & Decisions, by R. D. Luce and H. Raiffa.2 Carmel, California, May 1962 REFERENCES.

1. Berne, E., Transactional a.n.a.lysis in Psychotherapy, Evergreen, 1961.

2. Luce, R. D., and Raiffa, H., Games & Decisions, Chapman & Hall, 1957.

Introduction.

1 SOCIAL INTERCOURSE.

THE theory of social intercourse, which has been outlined at some length in Transactional a.n.a.lysis,1 may be summarized as follows.

Spitz has found2 that infants deprived of handling over a long period will tend at length to sink into an irreversible decline and are p.r.o.ne to succ.u.mb eventually to intercurrent disease. In effect, this means that what he calls emotional deprivation can have a fatal outcome. These observations give rise to the idea of stimulus-hunger, and indicate that the most favoured forms of stimuli are those provided by physical intimacy, a conclusion not hard to accept on the basis of everyday experience.

An allied phenomenon is seen in grown-ups subjected to sensory deprivation. Experimentally, such deprivation may call forth a transient psychosis, or at least give rise to temporary mental disturbances; In the past, social and sensory deprivation is noted to have had similar effects in individuals condemned to long periods of solitary imprisonment. Indeed, solitary confinement is one of the punishments most dreaded even by prisoners hardened to physical brutality,3,4and is now a notorious procedure for inducing political compliance. (Conversely, the best of the known weapons against political compliance is social organization.)5 On the biological side, it is probable that emotional and sensory deprivation tends to bring about or encourage organic changes. If the reticular activating system6 of the brain stem is not sufficiently stimulated, degenerative changes in the nerve cells may follow, at least indirectly. This may be a secondary effect due to poor nutrition, but the poor nutrition itself may be a product of apathy, as in infants suffering from marasmus. Hence a biological chain may be postulated leading from emotional and sensory deprivation through apathy to degenerative changes and death. In this sense, stimulus-hunger has the same relationship to survival of the human organism as food-hunger.

Indeed, not only biologically but also psychologically and socially, stimulus-hunger in many ways parallels the hunger for food. Such terms as malnutrition, satiation, gourmet, gourmand, faddist, ascetic, culinary arts, and good cook are easily transferred from the field of nutrition to the field of sensation. Overstuffing has its parallel in overstimulation. In both spheres, under ordinary conditions where ample supplies are available and a diversified menu is possible, choices will be heavily influenced by an individual's idiosyncrasies. It is possible that some or many of these idiosyncrasies are const.i.tutionally determined, but this is irrelevant to the problems at issue here.

The social psychiatrist's concern in the matter is with what happens after the infant is separated from his mother in the normal course of growth. What has been said so far may be summarized by the 'colloquialism':7 'If you are not stroked, your spinal cord will shrivel up.' Hence, after the period of close intimacy with the mother is over, the individual for the rest of his life is confronted with a dilemma upon whose horns his destiny and survival are continually being tossed. One horn is the social, psychological and biological forces which stand in the way of continued physical intimacy in the infant style; the other is his perpetual striving for its attainment. Under most conditions he will compromise. He learns to do with more subtle, even symbolic, forms of handling, until the merest nod of recognition may serve the purpose to some extent, although his original craving for physical contact may remain unabated.

This process of compromise may be called by various terms, such as sublimation; but whatever it is called, the result is a partial transformation of the infantile stimulus-hunger into something which may be termed recognition-hunger. As the complexities of compromise increase, each person becomes more and more individual in his quest for recognition, and it is these differentia which lend variety to social intercourse and which determine the individual's destiny. A movie actor may require hundreds of strokes each week from anonymous and undifferentiated admirers to keep his spinal cord from shrivelling, while a scientist may keep physically and mentally healthy on one stroke a year from a respected master.

'Stroking' may be used as a general term for intimate physical contact; in practice it may take various forms. Some people literally stroke an infant; others hug or pat it, while some people pinch it playfully or flip it with a fingertip. These all have their a.n.a.logues in conversation, so that it seems one might predict how an individual would handle a baby by listening to him talk. By an extension of meaning, 'stroking' may be employed colloquially to denote any act implying recognition of another's presence. Hence a stroke may be used as the fundamental unit of social action. An exchange of strokes const.i.tutes a transaction, which is the unit of social intercourse.

As far as the theory of games is concerned, the principle which emerges here is that any social intercourse whatever has a biological advantage over no intercourse at all. This has been experimentally demonstrated in the case of rats through some remarkable experiments by S. Levine8 in which not only physical, mental and emotional development but also the biochemistry of the brain and even resistance to leukemia were favourably affected by handling. The significant feature of these experiments was that gentle handling and painful electric shocks were equally effective in promoting the health of the animals.

This validation of what has been said above encourages us to proceed with increased confidence to the next section.

2 THE STRUCTURING OF TIME.

Granted that handling of infants, and its symbolic equivalent in grown-ups, recognition, have a survival value. The question is. What next ? In everyday terms, what can people do after they have exchanged greetings, whether the greeting consists of a collegiate 'Hi!' or an Oriental ritual lasting several hours ? After stimulus-hunger and recognition hunger comes structure-hunger. The perennial problem of adolescents is : 'What do you say to her (him) then ? And to many people besides adolescents, nothing is more uncomfortable than a social hiatus, a period of silent, unstructured time when no one present can think of anything more interesting to say than: 'Don't you think the walls are perpendicular tonight ?' The eternal problem of the human being is how to structure his waking hours. In this existential sense, the function of all social living is to lend mutual a.s.sistance for this project.

The operational aspect of time-structuring may be called programming. It has three aspects: material, social and individual. The most common, convenient, comfortable, and utilitarian method of structuring time is by a project designed to deal with the material of external reality: what is commonly known as work. Such a project is technically called an activity; the term 'work' is unsuitable because a general theory of social psychiatry must recognize that social intercourse is also a form of work.

Material programming arises from the vicissitudes encountered in dealing with external reality; it is of interest here only insofar as activities offer a matrix for 'stroking', recognition, and other more complex forms of social intercourse. Material programming is not primarily a social problem; in essence it is based on data processing. The activity of building a boat relies on a long series of measurements and probability estimates, and any social exchange which occurs must be subordinated to these in order for the building to proceed.

Social programming results in traditional ritualistic or semi-ritualistic interchanges. The chief criterion for it is local acceptability, popularly called 'good manners'. Parents in all parts of the world teach their children manners, which means that they know the proper greeting, eating, emunctory, courting and mourning rituals, and also how to carry on topical conversations with appropriate strictures and reinforcements. The strictures and reinforcements const.i.tute tact or diplomacy, some of which is universal and some local. Belching at meals or asking after another man's wife are each encouraged or forbidden by local ancestral tradition, and indeed there is a high degree of inverse correlation between these particular transactions. Usually in localities where people belch at meals, it is unwise to ask after the womenfolk; and in localities where people are asking after the womenfolk, it is unwise to belch at meals. Usually formal rituals precede semi-ritualistic topical conversations, and the latter may be distinguished by calling them pastimes.

As people become better acquainted, more and more individual programming creeps in, so that 'incidents' begin to occur. These incidents superficially appear to be advent.i.tious, and may be so described by the parties concerned, but careful scrutiny reveals that they tend to follow definite patterns which are amenable to sorting and cla.s.sification, and that the sequence is circ.u.mscribed by unspoken rules and regulations. These regulations remain latent as long as the amities or hostilities proceed according to Hoyle, but they become manifest if an illegal move is made, giving rise to a symbolic, verbal or legal cry of 'Foul!' Such sequences, which in contrast to pastimes are based more on individual than on social programming, may be called games. Family life and married life, as well as life in organizations of various kinds, may year after year be based on variations of the same game.

To say that the bulk of social activity consists of playing games does not necessarily mean that it is mostly 'fun' or that the parties are not seriously engaged in the relationship. On the one hand, 'playing' football and other athletic 'games' may not be fun at all, and the players may be intensely grim; and such games share with gambling and other forms of 'play' the potentiality for being very serious indeed, sometimes fatal. On the other hand, some authors for instance Huizinga,9 include under 'play' such serious things as cannibal feasts. Hence calling such tragic behaviour as suicide, alcohol and drug addiction, criminality or schizophrenia 'playing games' is not irresponsible, facetious or barbaric. The essential characteristic of human play is not that the emotions are spurious, but that they are regulated. This is revealed when sanctions are imposed on an illegitimate emotional display. Play may be grimly serious, or even fatally serious, but the social sanctions are serious only if the rules are broken.

Pastimes and games are subst.i.tutes for the real living of real intimacy. Because of this they may be regarded as preliminary engagements rather than as unions, which is why they are characterized as poignant forms of play. Intimacy begins when individual (usually instinctual) programming becomes more intense, and both social patterning and ulterior restrictions and motives begin to give way. It is the only completely satisfying answer to stimulus-hunger, recognition-hunger and structure-hunger. Its prototype is the act of loving impregnation.

Structure-hunger has the same survival value as stimulus-hunger. Stimulus-hunger and recognition-hunger express the need to avoid sensory and emotional starvation, both of which lead to biological deterioration. Structure-hunger expresses the need to avoid boredom, and Kierkegaard10 has pointed out the evils which result from unstructured time. If it persists for any length of time, boredom becomes synonymous with emotional starvation and can have the same consequences.

The solitary individual can structure time in two ways: activity and fantasy. An individual can remain solitary even in the presence of others, as every schoolteacher knows. When one is a member of a social aggregation of two or more people, there are several options for structuring time. In order of complexity, these are: (1) Rituals; (2) Pastimes; (3) Games; (4) Intimacy; and (5) Activity, which may form a matrix for any of the others. The goal of each member of the aggregation is to obtain as many satisfactions as possible from his transactions with other members. The more accessible he is, the more satisfactions he can obtain. Most of the programming of his social operations is automatic. Since some of the 'satisfactions' obtained under this programming, such as self-destructive ones, are difficult to recognize in the usual sense of the word 'satisfactions', it would be better to subst.i.tute some more non-committal term, such as 'gains' or 'advantages'.

The advantages of social contact revolve around somatic and psychic equilibrium. They are related to the following factors: (1) the relief of tension; (2) the avoidance of noxious situations; (3) the procurement of stroking; and (4) the maintenance of an established equilibrium. All these items have been investigated and discussed in great detail by physiologists, psychologists, and psychoa.n.a.lysts. Translated into terms of social psychiatry, they may be stated as (1) the primary internal advantages; (2) the primary external advantages; (3) the secondary advantages; and (4) the existential advantages. The first three parallel the 'gains from illness' described by Freud: the internal paranosic gain, the external paranosic gain, and the epinosic gain, respectively.11 Experience has shown that it is more useful and enlightening to investigate social transactions from the point of view of the advantages gained than to treat them as defensive operations. In the first place, the best defence is to engage in no transactions at all; in the second place, the concept of 'defences' covers only part of the first two cla.s.ses of advantages, and the rest of them, together with the third and fourth cla.s.ses, are lost to this point of view.

The most gratifying forms of social contact, whether or not they are embedded in a matrix of activity, are games and intimacy. Prolonged intimacy is rare, and even then it is primarily a private matter; significant social intercourse most commonly takes the form of games, and that is the subject which princ.i.p.ally concerns us here. For further information about time-structuring, the author's book on group dynamics should be consulted.12 REFERENCES.

1. Berne, E., Transactional a.n.a.lysis in Psychotherapy, Evergreen, 1961.

2. Spitz, R., 'Hospitalism: Genesis of Psychiatric Conditions in Early Childhood', Psychoa.n.a.lytic Study of the Child, 1: 5374, 1945.

3. Belbenoit, Rene, Dry Guillotine, Cape, 1938.

4. Seaton, G. J., Scars on my Pa.s.sport, Hutchinson, 1951.

5. Kinkead, E., Why they Collaborated, Longmans, 1960.

6. French, J. D., 'The Reticular Formation', Scientific American, 196: 5460, May 1957.

7. The 'colloquialisms' used are those evolved in the course of time at the San Francisco Social Psychiatry Seminars.

8. Levine, S., 'Stimulation in Infancy', Scientific American, 202: 8086, May 1960.

Levine, S., 'Infantile Experience and Resistance to Physiological Stress', Science, 126:405, 30 August 1957.

9. Huizinga, J., h.o.m.o Ludens, Routledge, 1949.

10. Kierkegaard, S., A Kierkegaard Anthology (ed. R. Bretall), Princeton University Press, 1947, pp. 22ff.

11. Freud, S., 'General Remarks on Hysterical Attacks', Standard Edn, n, Hogarth Press, London, 1955.

Freud, S., 'a.n.a.lysis of a Case of Hysteria', ibid., VI, 1953.

12. Berne, E., The Structure and Dynamics of Organizations and Groups, Pitman Medical, 1963.

PART ONE.

a.n.a.lYSIS OF GAMES.

1 Structural a.n.a.lysis.

OBESERVATION of spontaneous social activity, most productively carried out in certain kinds of psychotherapy groups, reveals that from time to time people show noticeable changes in posture, view-point, voice, vocabulary, and other aspects of behaviour. These behavioural changes are often accompanied by shifts in feeling. In a given individual, a certain set of behaviour patterns corresponds to one state of mind, while another set is related to a different psychic att.i.tude, often inconsistent with the first. These changes and differences give rise to the idea of ego states.

In technical language, an ego state may be described phenome-nologically as a coherent system of feelings, and operationally as a set of coherent behaviour patterns. In more practical terms, it is a system of feelings accompanied by a related set of behaviour patterns. Each individual seems to have available a limited repertoire of such ego states, which are not roles but psychological realities. This repertoire can be sorted into the following categories: (1) ego states which resemble those of parental figures; (2) ego states which are autonomously directed towards objective appraisal of reality and (3) those which represent archaic relics, still-active ego states which were fixated in early childhood. Technically these are called, respectively, exteropsychic, neopsychic, and archaeopsychic ego states. Colloquially their exhibitions are called Parent, Adult and Child, and these simple terms serve for all but the most formal discussions.

The position is, then, that at any given moment each individual in a social aggregation will exhibit a Parental, Adult or Child ego state, and that individuals can shift with varying degrees of readiness from one ego state to another. These observations give rise to certain diagnostic statements. 'That is your Parent' means: 'You are now in the same state of mind as one of your parents (or a parental subst.i.tute) used to be, and you are responding as he would, with the same posture, gestures, vocabulary, feelings, etc' 'That is your Adult' means: 'You have just made an autonomous, objective appraisal of the situation and are stating these thought-processes, or the problems you perceive, or the conclusions you have come to, in a non-prejudicial manner.' 'That is your Child' means: 'The manner and intent of your reaction is the same as it would have been when you were a very little boy or girl.'

The implications are:.

1. That every individual has had parents (or subst.i.tute parents) and that he carries within him a set of ego states that reproduce the ego states of those parents (as he perceived them), and that these parental ego states can be activated under certain circ.u.mstances (exteropsychic functioning). Colloquially: 'Everyone carries his parents around inside of him.'

2. That every individual (including children, the mentally r.e.t.a.r.ded and schizophrenics) is capable of objective data processing if the appropriate ego state can be activated (neopsychic functioning) Colloquially: 'Everyone has an Adult.'

3. That every individual was once younger than he is now, and that he carries within him fixated relics from earlier years which will be activated under certain circ.u.mstances (archaeopsychic functioning). Colloquially: 'Everyone carries a little boy or girl around inside of him.'

At this point it is appropriate to draw Figure 1A, which is called a structural diagram. This represents, from the present viewpoint, a diagram of the complete personality of any individual. It includes his Parental, Adult, and Child ego states. They are carefully segregated from each other, because they are so different and because they are so often quite inconsistent with each other. The distinctions may not be clear at first to an inexperienced observer, but soon become impressive and interesting to anyone who takes the trouble to learn structural diagnosis. It will be convenient henceforth to call actual people parents, adults or children, with no capital letters; Parent, Adult and Child, capitalized, will be used when ego states are referred to. Figure 1B represents a convenient, simplified form of the structural diagram.

Before we leave the subject of structural a.n.a.lysis, certain complications should be mentioned.

1. The word 'childish' is never used in structural a.n.a.lysis, since it has come to have strong connotations of undesirability, and of something to be stopped forthwith or got rid of. The term 'childlike' is used in describing the Child (an archaic ego state), since it is more biological and not prejudicial. Actually the Child is in many ways the most valuable part of the personality, and can contribute to the individual's life exactly what an actual child can contribute to family life: charm, pleasure and creativity. If the Child in the individual is confused and unhealthy, then the consequences may be unfortunate, but something can and should be done about it.

2. The same applies to the words 'mature' and 'immature'. In this system there is no such thing as an 'immature person'. There are only people in whom the Child takes over inappropriately or unproductively, but all such people have a complete, well-structured Adult which only needs to be uncovered or activated. Conversely, so-called 'mature people' are people who are able to keep the Adult in control most of the time, but their Child will take over on occasion like anyone else's, often with disconcerting results.

Figure 1. A Structure Diagram.

3. It should be noted that the Parent is exhibited in two forms, direct and indirect: as an active ego state, and as an influence. When it is directly active, the person responds as his own father (or mother) actually responded ('Do as I do'). When it is an indirect influence, he responds the way they wanted him to respond ('Don't do as I do, do as I say'). In the first case he becomes one of them; in the second, he adapts himself to their requirements.

4. Thus the Child is also exhibited in two forms: the adapted Child and the natural Child. The adapted Child is the one who modifies his behaviour under the Parental influence. He behaves as father (or mother) wanted him to behave: compliantly or precociously, for example. Or he adapts himself by withdrawing or whining. Thus the Parental influence is a cause, and the adapted Child an effect. The natural Child is a spontaneous expression: rebellion or creativity, for example. A confirmation of structural a.n.a.lysis is seen in the results of alcohol intoxication. Usually this decommissions the Parent first, so that the adapted Child is freed of the Parental influence, and is transformed by release into the natural Child.

It is seldom necessary, for effective game a.n.a.lysis, to go beyond what has been outlined above as far as personality structure is concerned.

Ego states are normal physiological phenomena. The human brain is the organ or organizer of psychic life, and its products are organized and stored in the form of ego states. There is already concrete evidence for this in some findings of Penfield and his a.s.sociates.1, 2 There are other sorting systems at various levels, such as factual memory, but the natural form of experience itself is in shifting states of mind. Each type of ego state has its own vital value for the human organism.

In the Child reside intuition,3 creativity and spontaneous drive and enjoyment.

The Adult is necessary for survival. It processes data and computes the probabilities which are essential for dealing effectively with the outside world. It also experiences its own kinds of setbacks and gratifications. Crossing a busy highway, for example, requires the processing of a complex series of velocity data; action is suspended until the computations indicate a high degree of probability for reaching the other side safely. The gratifications offered by successful computations of this type afford some of the joys of skiing, flying, sailing, and other mobile sports. Another task of the Adult is to regulate the activities of the Parent and the Child, and to mediate objectively between them.

The Parent has two main functions. First, it enables the individual to act effectively as the parent of actual children, thus promoting the survival of the human race. Its value in this respect is shown by the fact that in raising children, people orphaned in infancy seem to have a harder time than those from homes unbroken into adolescence. Secondly, it makes many responses automatic, which conserves a great deal of time and energy. Many things are done because 'That's the way it's done.' This frees the Adult from the necessity of making innumerable trivial decisions, so that it can devote itself to more important issues, leaving routine matters to the Parent.

Thus all three aspects of the personality have a high survival and living value, and it is only when one or the other of them disturbs the healthy balance that a.n.a.lysis and reorganization are indicated. Otherwise each of them, Parent, Adult, and Child, is ent.i.tled to equal respect and has its legitimate place in a full and productive life.

REFERENCES.

1. Penfield, W., 'Memory Mechanisms', Archives of Neurology & Psychiatry, 67: 178198,1952.

2. Penfield, W., and Jasper, H., Epilepsy and the Functional Anatomy of the Human Brain, Churchill, 1954, Chapter 11.

3. Berne, E., 'The Psychodynamics of Intuition', Psychiatric Quarterly, 36: 294300, 1962.

2 Transactional a.n.a.lysis.

THE unit of social intercourse is called a transaction. If two or more people encounter each other in a social aggregation, sooner or later one of them will speak, or give some other indication of acknowledging the presence of the others. This is called the transactional stimulus. Another person will then say or do something which is in some way related to this stimulus, and that is called the transactional response. Simple transactional a.n.a.lysis is concerned with diagnosing which ego state implemented the transactional stimulus, and which one executed the transactional response. The simplest transactions are those in which both stimulus and response arise from the Adults of the parties concerned. The agent, estimating from the data before him that a scalpel is now the instrument of choice, holds out his hand. The respondent appraises this gesture correctly, estimates the forces and distances involved, and places the handle of the scalpel exactly where the surgeon expects it. Next in simplicity are Child-Parent transactions. The fevered child asks for a gla.s.s of water, and the nurturing mother brings it.

Both these transactions are complementary; that is, the response is appropriate and expected and follows the natural order of healthy human relationships. The first, which is cla.s.sified as Complementary Transaction Type I, is represented in Figure 2A. The second, Complementary Transaction Type II, is shown in Figure 2B. It is evident, however, that transactions tend to proceed in chains, so that each response is in turn a stimulus. The first rule of communication is that communication will proceed smoothly as long as transactions are complementary; and its corollary is that as long as transactions are complementary, communication can, in principle, proceed indefinitely. These rules are independent of the nature and content of the transactions; they are based entirely on the direction of the vectors involved. As long as the transactions are complementary, it is irrelevant to the rule whether two people are engaging in critical gossip (Parent-Parent), solving a problem (Adult-Adult), or playing together (Child-Child or Parent-Child).

The converse rule is that communication is broken off when a crossed transaction occurs. The most common crossed transaction, and the one which causes and always has caused most of the social difficulties in the world, whether in marriage, love, friendship, or work, is represented in Figure 3A as Crossed Transaction Type I. This type of transaction is the princ.i.p.al concern of psychotherapists and is typified by the cla.s.sical transference reaction of psychoa.n.a.lysis. The stimulus is Adult-Adult: e.g., 'Maybe we should find out why you've been drinking more lately,' or, 'Do you know where my cuff links are?' The appropriate Adult-Adult response in each case would be: 'Maybe we should. I'd certainly like to know!' or, 'On the desk.' If the respondent flares up, however, the responses will be something like 'You're always criticizing me, just like my father did,' or, 'You always blame me for everything.' These are both Child-Parent responses, and as the transactional diagram shows, the vectors cross. In such cases the Adult problems about drinking or cuff links must be suspended until the vectors can be realigned. This may take anywhere from several months in the drinking example to a few seconds in the case of the cuff links. Either the agent must become Parental as a complement to the respondent's suddenly activated Child, or the respondent's Adult must be reactivated as a complement to the agent's Adult. If the maid rebels during a discussion of dishwashing, the Adult-Adult conversation about dishes is finished; there can only ensue either a Child-Parent discourse, or a discussion of a different Adult subject, namely her continued employment.

Figure 2. Complementary Transactions The converse of Crossed Transaction Type I is ill.u.s.trated in Figure 3B. This is the counter-transference reaction familiar to psychotherapists, in which the patient makes an objective, Adult observation, and the therapist crosses the vectors by responding like a parent talking to a child. This is Crossed Transaction Type II. In everyday life, 'Do you know where my cuff links are?' may elicit: 'Why don't you keep track of your own things? You're not a child any more.'

Figure 3. Crossed Transactions The relationship diagram in Figure 4, showing the nine possible vectors of social action between an agent and a respondent, has some interesting geometrical (topological) qualities. Complementary transactions between 'psychological equals' are represented by (11)2, (55)2 and (99)2. There are three other complementary transactions: (24) (42), (37) (73) and (68) (86). All other combinations form crossed transactions, and in most cases these show up as crossings in the diagram: e.g., (37) (37), which results in two speechless people glaring at each other. If neither of them gives way, communication is finished and they must part. The most common solutions are for one to yield and take (73), which results in a game of 'Uproar'; or better, (55)2, in which case they both burst out laughing or shake hands.

Simple complementary transactions most commonly occur in superficial working and social relationships, and these are easily disturbed by simple crossed transactions. In fact a superficial relationship may be defined as one which is confined to simple complementary transactions. Such relationships occur in activities, rituals and pastimes. More complex are ulterior transactions those involving the activity of more than two ego states simultaneously and this category is the basis for games. Salesmen are particularly adept at angular transactions, those involving three ego states. A crude but dramatic example of a sales game is ill.u.s.trated in the following exchange: Figure 4. A Relationship Diagram Salesman: 'This one is better, but you can't afford it.'

Housewife: 'That's the one I'll take.'

The a.n.a.lysis of this transaction is shown in Figure 5A. The salesman, as Adult, states two objective facts: 'This one is better' and 'You can't afford it'. At the ostensible, or social, level these are directed to the Adult of the housewife, whose Adult reply would be: 'You are correct on both counts.' However, the ulterior, or psychological, vector is directed by the well-trained and experienced Adult of the salesman to the housewife's Child. The correctness of his judgement is demonstrated by the Child's reply, which says in effect: 'Regardless of the financial consequences, I'll show that arrogant fellow I'm as good as any of his customers.' At both levels the transaction is complementary, since her reply is accepted at face value as an Adult purchasing contract.

Figure 5. Ulterior Transactions A duplex ulterior transaction involves four ego states, and is commonly seen in flirtation games.

Cowboy: 'Come and see the barn.'

Visitor: 'I've loved barns ever since I was a little girl.'