Fluxury by Sergio Benvenuto

SIMPLISTIC COMPLEXITY. A Discussion on Psychoanalysis and Chaos TheoryMar/19/2017


In the eighties Paul Watzlawick held a series of seminars in Italy.  He brought two amusing clinical examples to illustrate his psychotherapeutic “philosophy”. 

          One patient had been seriously agoraphobic for some time.  He practically never left his house. After a series of psychotherapies, all failed, he decided to end it all.  He got into his car and headed for the Pacific, meaning to plunge himself into the ocean in his car (not a surprising choice of transport, given that in California the private car is the only feasible transportation).  As he was driving he realized that he felt no distress...  He could drive!  This, it appears, was a turning point for him.  The moral of the story according to Watzlawick: “A proper psychotherapist would never have prescribed to his patient something like ‘jump into your car and kill yourself!’  But in this case it might have been the most appropriate prescription.”

The other case.  A couple, after twenty years of marriage, is no longer able to have sexual intercourse.  He, in particular, no longer feels like it.  So they go to Palo Alto’s Mental Research Institute to see what they can do about it.  Watzlawick’s comment: “After all, I think it’s normal for a couple not to feel like having sex anymore after 20 years of marriage!”  [My own comment: if he thought it “normal” for a husband and wife to be bored with each other, why go in for a “cure”?  In other words, should psychotherapy be palmed off like some sort of cosmetic surgery?  You get a facelift or nose job not because there is something pathological about your face or nose, but to meet your demand.  And, just as cosmetic surgery is second-rate medicine, isn’t this palliative psychotherapy also somehow second-rate?  But this is an issue that would take us down a very long road.]

          One evening the couple spend the night at the house of some friends, and sleep together in a bed with one side against the wall—a change from their habitual bed, open on three sides.  During the night the husband gets up to pee, but to get out of the bed he has to climb over his wife.  While sleepily in the process of doing this, he feels underneath him “something of value”, a naked woman!  Followed by an unexpected coitus.  It is the rediscovery of his wife as an “erotic value”: they go back to legitimate married couple intercourse.  And they lived happily ever after.  This is what Watzlawick squeezes out of this story: “One should not necessarily believe that for complex and profound situations—like a marriage crisis—the right answer need also be complex and profound.  A classical analyst would start sounding out the earliest childhood libidinal experiences and feelings, whilst in this case... all they had to do was sleep on a bed set against the wall.  Occasionally minimal causes can trigger off huge effects; while, on the contrary, huge causes can produce irrelevant effects ”.



          Watzlawick was one of the first psychotherapists to understand the importance of chaos and complexity theory.  The two clinical examples he cites[1] evoke the meteorologist Edward Lorenz’s famous sentence, by now a sort of post-modern ‘proverb’: “A butterfly flapping its wings in Japan could cause a hurricane in Argentina”.  In scientific terms, it is said that many natural processes—including atmospheric ones—are extremely sensitive to initial conditions, in other words, very dependent on them.  If we alter ever so slightly the beginning of a long process, we will end up with radically, even overwhelmingly, different effects.  This means that many natural processes—including social ones—are in fact unpredictable.  In the same way that “heads or tails” is unpredictable when tossing a coin: in fact, all that is necessary is to alter the initial conditions imperceptibly—by varying the thrust at the moment of tossing by an infinitesimal measure—and the result will be the opposite.  Precisely because it is unpredictable, we call this a chance result.  So the basic philosophical question is the following: is it unpredictable because it is chance, or do we call it chance because it is as a matter of fact unpredictable?  And is the unpredictability due to the fact that we can’t predict it, or because the process itself is “free”?

However, the contrary of Lorenz’s statement is often forgotten: 20,000 huge ventilators running in Japan do not necessarily create a draft in Argentina or elsewhere.  If it is true that minimal initial variations can cause spectacular effects, it is also true that maximum initial variations may not create any relevant effect for us.  This is because, as chaos theory teaches us, strange attractors are often operating: variations may be huge, the process may appear chaotic, yet in the long term a sort of secret, underlying regularity, masked by unpredictable variations, establishes itself.  Short-term results are unpredictable, but in the long term an order is established (an order masked by chaos), even if not a very strict one.  True, no weather forecaster will be able to tell you with absolute certainty whether it’ll rain in West Hartlepool tomorrow or not, but then again you don’t need to be a weatherman to bet that next winter it will be colder in West Hartlepool than next summer.  A winter and a summer can be more or less cold—the change from one year to the next is unpredictable—but an attractor stranger than one might think legitimizes the prediction that “next January you’ll have to wear an overcoat, next July even a t-shirt will probably be too hot”.

Thus, not only may minimal initial variations generate entirely different and unpredictable processes, but phenomena that come across as purely chaotic turn out to be “attracted” by a sort of inertia that’s difficult to change.  This is the case for all living, homeostatic, systems: the living system tends to cushion all fortuity and to maintain the organism’s general structure.  We could say that psychopathologies reveal homeostases that are difficult to deconstruct—insofar as a psychopathological disorder means repeating the same symptom, the same mistakes, over and over.  It is, so to speak, an incapacity to learn from experience.

          Watzlawick tries to bring this complexity mentality into the field of psychotherapy.  Human processes are no less chaotic than physical ones: nothing ascertains that cause A leads linearly to effects B, C and D in a certain (in both its senses of particular and sure) sequence.  This is what today discredits certain classic psychoanalytic explanations for particular pathologies, behavior or destinies: they are too linear (viz. simplistic).  For example: why in the same family do we get one son who becomes an active, promiscuous homosexual and an other an extremely faithful and happy husband and father of several children?  In fact, no one has ever managed to explain what specific initial conditions are needed for one son to evolve towards homosexuality rather than towards other tendencies, even if they share the same parents and have passed their childhood in the same family (even if, in some cases, “you can already tell” who’s homosexual—or hysteric, psychotic or a mathematical genius--even in a six-year-old).  The process that brings a subject to roll towards one side—be it homosexual, perverse, heterosexual, hysteric, psychotic, or of political commitment, a psychoanalytic career, etc.—is largely unpredictable.  In the same way, nothing can tell us that a patient’s therapeutic improvement—or metanoia, conversion—has resulted from certain profound interpretations of the analyst, rather than from other interpretations (quite the contrary, there is the legitimate doubt that it is not the analyst’s interpretations—whatever analytic school s/he belongs to—to produce effects in the subject).



          All this is true. But it is also true that Watzlawick’s solution isn’t too convincing precisely on the basis of the theory to which he refers.  I want to make it clear that I am not hostile a priori—in a moralistic way—to psychotherapeutic prescriptions.  A pluralist like me is not upset by the multiplication of approaches, and therefore also accepts those based on prescriptions, whether they be paradoxical, systemic, cognitivist, behaviorist, etc.  It is truly naive to believe that the analytic approach is different from these others insofar as it is rigorously non-prescriptive.  First of all, the analyst prescribes a setting and the rules connected to this—and this is not just a trifle.  Those like me who think that the analytic setting is responsible for at least 80% of the effects of analysis do not consider irrelevant the often very strict rules imposed by the analyst.  Furthermore, one would have to be really psychologically illiterate to believe that the analyst, because he doesn’t explicitly prescribe, does not prescribe in his/her own way!  We know that a cough, a yawn, a crossing of the legs are enough to “prescribe” to patients what to say, how to say it, whether to say it, etc.  The famous Bion, I’ve been told, late in life set up his supervisions in such a way as to prove that the subject had the very dreams and expressed the very thoughts that the analyst, deep down, wanted to hear.  “How Kleinian your patients’ dreams are!”, he sarcastically told the Tavistock Clinic missionaries who told him about their cases[2].  Every analysand, like every subject, has a sixth sense that makes her aware of what the other wishes or doesn’t wish.  So, in the end, psychoanalysis is as prescriptive as any other psychotherapy—except that it prescribes in a more refined, elegant, sophisticated and a bit hypocritical and politically correct way.  The differentiating point does not lie in the prescription.

          But it is precisely this chaos-complexity approach that makes us see how the single prescription doesn’t mean much.  For example, does the case that Watzlawick cites mean that for most cases of severe agoraphobia the analyst should prescribe “get in your car and kill yourself? ”  But because there is no linear relation between prescription and behavior, the former could lead to totally catastrophic effects.  The agoraphobic might actually succeed in killing himself—or, more likely, he would give up psychotherapy.  By the same reasoning, should one say to every couple who can no longer manage intercourse to “just sleep in a bed with one side against the wall?”  In other cases, a husband, forced to sleep between wife and wall, might have an attack of claustrophobia...  An anti-linear approach like Watzlawick’s should not lead one to believe that one carefully chosen prescription is enough to obtain the desired effect.  Just as it would be laughable to make a butterfly in Japan flap its wings in the hope of causing a hurricane in Argentina.  Precisely because chaos theory teaches us that macroscopic effects are unpredictable, it is vain to hope that certain prescriptions can change a repetitive situation hardened by time.  Prescriptions may be as strange as you wish, like the bizarre Zen enigmas (koan): the strange attractor—known as psychopathology—can still reaffirm itself.  But then, what can the psychotherapist, whether a “prescriber” or not, do?  Whatever he uses—whether prescriptions, interpretations, the timing of sessions, silence, or acting—what guarantee can a shrink really have that what he’s doing will have a desirable effect?

          An answer I think can be found in the very concept of the attractor in chaos theory.  Personally, I am not a fan of long or endless psychoanalysis.  On the contrary, I believe that psychoanalysis should manage to find a way not only to shorten times, but to get tangible results in non-geological time-spans.  Having said this, in contrast to hurried psychotherapies, a certain duration in the analytic relationship is necessary precisely in order that it tap into chaotic, subjective processes.  The strange attractor actually reveals—and may modify—itself only in the long-term.  This means that, if to cure signifies to get rid of the strange attractor, a small variation is not enough—for example, quickly eliminating an annoying symptom—to obtain a full cure.

By this I do not at all mean that, being a long relationship, psychoanalysis can be reduced to a pedagogy, or rather a paedo-orthopaedics.  It is true that analysts, over time, somehow educate analysands: they teach them to analyze themselves.  But analysis cannot be reduced to pedagogy: in my opinion the analyst doesn’t behave like that “good enough mother” who, according to the theory, the subject never had in childhood.  (The problem remains: do I reject these approaches because they are not real analysis, or simply because I ethically reject them?  In this field, truth, goodness and effectiveness are hard to disentangle.)  The long duration of analysis is not a pedagogical duration but rather a duration of weaning[3]—or, if you prefer, of mourning. 



          Thinking in terms of complexity means thinking in terms of webs (Freud too, in his 1895 Project for a Scientific Psychology, had perfectly understood—well ahead of his time—that the brain and the mind are a web).  Even in psychotherapy—especially analytic—what counts is to create a web.  In individual analysis as well, a web of relationships is woven between the two players.  But this weaving takes time.  Bad analysis—the type that ends quickly or goes on forever—is one where the network becomes oversimplified: the analyst begins to react predictably, one-on-one, ego-against-ego, and so the subject reintegrates analysis into the web dominated by the usual attractor.

So, precisely because in a web the effects of any input are unpredictable and largely uncontrollable, a multitude of inputs is required for a mutation to occur.  But what do we mean by mutation?  We mean that a subject changes attractor.  Or, even better, that he has many attractors in his life.  A neurotic is someone who cannot “change tune”, who always repeats, in short, the same mistakes.  It’s what the later Freud called the death drive: to always end up, in the search for pleasure, with the same displeasure.  Whatever the sujet à problèmes does, in one sense or the other, the result will always be the same.  This is the inanity of “cosmetic” therapies that cure only individual symptoms: one wonders whether they really change the strange attractor that is causing the neurosis, and whether the subject will really stop repeating his zero sum game.

          But what is capable of changing an attractor?  No one can say exactly.  So that there are serious symptoms that disappear in just a few sessions, while other apparently minor symptoms persist even after a ten-year analysis.  For example, Freud cured Gustav Mahler of an impotence symptom in one session.  Is that Freud's genius?  No, it is just that the butterfly flapped its wings at the right moment.  But when this does not happen, then forms of intervention have to be multiplied in time, until one day—there’s no telling why or how—the modification happens.

          In the evolution of life, according to modern biology, things take place in a very similar way.  Mutations, which in Darwinian theory are always random, may take place often but are mostly unsuccessful and are eliminated.  Then, suddenly, a tiny mutation takes place that enjoys an overwhelming success (often due to entirely contingent environmental circumstances): and a new species appears, like Homo Sapiens.  And if analysis were a bit like the evolution of species?  Time is needed for the analyst’s persistent but flexible inputs to be able to finally modify the attractor, to constitute a new speciation, as the biologists say, that is to say the determination of a new species.  I wonder, therefore: does this persistence consist of an accumulation of inputs in the same sense, just as one swallow doesn’t make summer?  It is not a rhetorical question.

          If this hypothesis of mine is of any value, we can then say that what distinguishes an analytic therapy from any other psychotherapy is simply the fact that it is long.  It is not really distinguished by the theoretical model it refers to, and therefore to the content of interpretations—whether the model be Freudian, Jungian, Bionian, Lacanian, or Kohutian, etc.  What really cures is time.  That is to say, what with time causes caesurae, cuts, modifications, the “new cycle” as M. Balint called it.  It is with time that—thanks to the presence of the analyst—a subject’s life can find emancipation from the neurotic attractor and get into new cycles.



          But then, someone might say, if the content of interpretations does not count at all, why interpret?  Why doesn’t the analyst simply keep his mouth shut?  I think that, sooner or later, analysts can (or have to?) say something—no matter what their hermeneutic system of reference is, if they have one at all—simply to signify to their subjects that they are there.  That they are not pieces of stone, that the patient is confronting himself with a subject who is interpreting in his own way.  It is as if analysts, when they speak, were only and always saying: “I’m here, I’m listening”.  And I would add: “I’m not applying a pre-established scheme of interpretation to what you’re telling me, I’m trying to understand who you really are!”

          Therefore, the analyst, thanks to the setting he imposes, creates a simple attractor—regular sessions, benevolent neutrality, renunciation of acting out, etc.  Around this spider a web is spun: this, one hopes, will modify the strange attractor that makes up the neurosis.  Thus it is not an analyst’s interpretations that work, but rather what I would call his shirking away from transference.  And an analyst does shirk away from transference—i.e. he’s not offended if you insult him, he doesn’t have sex with patients if they try to seduce him, etc., etc.—because a part of him always remains outside the web.  His staying out of the web is the only show of caritas (love) that can be given.  In other words, psychoanalysis is not a question of truth, but of charity.

Transference is nothing other than the subject’s repetition in analysis of the usual modalities—i.e. those regulated by the strange attractor.  If, for example, the analyst—maybe because of his age or because he wears a moustache—reminds the subject of his/her surly father or neglectful mother, s/he’ll treat the analyst like this surly father or neglectful mother.  If the analyst reacted like an ordinary person, like another Ego—as any relative, friend, lover, or colleague of the subject most certainly would—the attractor confirm itself.  If, for example, the analysand began making sly remarks against the analyst, the latter would never behave like someone else who felt “bitten”:  he wouldn’t bite back nor would he run away.  The analyst, thanks to the simplicity of the setting that protects him, can afford not to “act or react”: he doesn’t fight back or give up.  With his presence, silent or more talkative, he presents another possibility.  It is as if the analyst said: “you’re treating me like a surly father... but I am something other”.  But what other?  I would say other and no more.  Another who doesn’t act or react, but persists.  By maintaining this persistence, by keeping on flapping the wings a little, the perturbation can change direction.

          Thus, the analyst neither forms nor conforms—rather I would say unforms—the subject: he allows the neurotic coherence to disintegrate.  And he succeeds with time—through his persistence—to un-do the strange attractor.  But not because he has revealed some truth to the subject—even if the latter often has the sensation that something has been “revealed” to him.  Until now the analytic tradition has still been too mired in a “platonizing” vision, where “truth heals”—like in the Gospel, “the truth will set you free”.   Perhaps the subject is transformed simply because the analyst, in his persistence, insisted.

                    In fact, chaos theory is above all a mathematical theory which deals with quantities and numbers; an attractor is an average found in a crowd of numbers.  But, at least until now, psychoanalysis has not been quantitative: it deals with qualitative forms.  This means that once we have a real non-metrical complexity theory we will better be able to apply it to the analytic process.  We already have non-quantitative mathematical approaches, such as topology—which is why the (rare) analysts who make use of mathematics prefer topological tools.  In fact, the analytic process itself refuses a pure numerical treatment; rather, it needs a qualitative version of Chaos Theory.


Sergio Benvenuto

[1] The first example too, insofar as there is no linearity between the cause of the action and its effect: the agoraphobic jumps into his car to commit suicide... instead he is cured of his phobia.  Systems and complexity theory rightly highlight the non-linearity between causes and effect in living systems.


[2] I thank Bice Benvenuto, at the time student at the Tavistock Clinic, for having reported that to me.


[3]See my  “L’analisi: uno svezzamento lungo?”, Psicoterapia e Scienze Umane, year XXXIV, n. 3, 2000, pp. 27-50.


Flussi © 2016Privacy Policy