Matteo Selvini, Editor, interviewed by Germana Cavallini

By adminpsy
4 May 2014

GC: Welcome, everyone. We are at the Mara Selvini Palazzoli Psychotherapy Training School to tackle the topic of research studies in the field of psychotherapy talking with Dr. Matteo Selvini. to start off  we will ask Dr. Selvini: "What might facilitate the creation of a new approach in psychotherapy? And how does this come about?

MS: Well, I believe that at the heart of it is the pain and discomfort of the therapist himself, that is to say the expert who gets involved in the cases he’s following, with the people he is trying to help, often has to deal with his own failures and his own sense of helplessness. So naturally he will attempt to come up with new ideas to help him avoid those feelings of failure and helplessness. I believe that all advances throughout the history of psychotherapy were made on this basis, when a therapist finds himself reliving something he has already experienced and that brought with it consequences that, in time, proved to be negative. So, when thinking back to our own history, my mother’s history comes to mind:  Mara Selvini Palazzoli, was a psychoanalyst, a therapist, and she worked with anorexic girls (I say "girls" because anorexic clients were just female). She dedicated many years to her work with these girls, feeling that same distress when patients would interrupt their therapy after perhaps only a few months, while others, even after five or six or seven years of therapy, were still very seriously ill. Thus, family therapy as a new method was born from this distress, this discomfort that my mother and many other professionals in the field felt in dealing with their lack of success with a particular means, such as psychoanalysis, and their desperate struggle to improve their means. This, indeed, was the case in my mother’s history: she began to question classic, orthodox, drive theory-based psychoanalysis, which is  useless with anorexic girls -it doesn’t work – and thus turned to other models for inspiration. Other models that she had turned to included Sullivan, Karen Horney and existential phenomenology. In the end, however, she realized that something even more radical than these practices was required, which was when she made the leap to family therapy. This was the first example that came to mind, going back fifty years, but it is still just as true today as it was then.

Taking much more recent experiences into consideration, we realize that even within family therapy  and other certain kinds of therapy, there is perhaps a risk of them becoming too cognitive, too intellectual, too reliant on speech. At some point, we come to realize that these very verbal practices do not, in fact, produce actual change and this leads to that feeling of futility, of helplessness, rising once again, that actually inspires new ideas. Something that we find often today in the field of family therapy is the use of techniques that provoke strong emotional reactions within the sessions themselves, moving away slightly from the much too cognitive, intellectual bent they have acquired. This is a time of great creativity in coming up with new, somewhat experiential approaches. However, out of all the classic techniques, and there are many of them, perhaps one of the most famous in the field of systemic therapy is working with sculptures, i.e. having the family members “sculpt” themselves as they were in the past and how they might see themselves in the future. This has often given significant results, it turns the session around completely, moving from an overly intellectual dimension to one where emotions are unleashed, and we see things that we would not see otherwise. Sometimes, the way people position themselves can open up new worlds and I believe that to be the answer we seek. In any case, when all is said and done, therapists work most of all on themselves and on their own emotions, both positive and negative. And when the pain of being unable to help one’s patients overshadows all else, that’s when something truly needs to be done, isn’t it? You might begin with a single case that can then be generalized into a technique to be used with others.

GC: I would now like to ask you how we can categorize clinical cases in order to benefit research studies.

MS: There is a great complexity at work here and the problem we face in our field is that we are immersed in it, making it quite intolerable for us simply because we can’t wrap our heads around it. This is why we need categories, to simplify this complexity and make it manageable. As a result, all schools of psychotherapy must either implicitly or explicitly come up with ways to categorize this extremely complex reality that needs to be dealt with. The model we’ve come up with for this purpose works on –let’s say- seven levels or seven dimensions and it helps clarify… Quite often diagnosis is considered useless as far as therapy goes. Culturally and scientifically, we battle against this attitude because, even though -as many of our colleagues state- we don’t want to be tied down by diagnoses and we want to preserve the uniqueness of each case; however, those who follow this particular line of reasoning make use of categories, and they do implicitly and this doesn’t sit well with me. Better for us to make it overtly clear what categories we’re using so as to avoid situations where we pretend to challenge and refuse diagnosis due to the stigma attached to it, only to then use it without being fully aware, which I consider a major dysfunctional aspect of psychotherapy.

In light of this, we’ve made a strong effort to clearly state which diagnoses we use and we have come up with a framework  in order to have “workable” diagnoses. So we came up with seven types of “diagnoses”.  

The first and most useful of these is the diagnosis of  the request: we have drawn up a chart and found that all requests related to psychotherapy can be grouped into four categories. We speak of ‘individual request’, when the patient seeks help for himself, and ‘family request’, when a person asks for help on behalf of a family member. In both cases we are referring to a personal issue: one individual has a problem. The third case, which is a little different, we’ve called ‘relationship request’, that is where the request concerns a type of interpersonal crisis, between partners or father and daughter for instance. The fourth is the so-called ‘forced request’, which is when no one is expressing a request but rather an institution, a court of law, a school or some other form of authority decrees it mandatory to see a psychologist or therapist. All of the above are considered a part of this first classification, which is made within five to ten seconds during first contact with a patient. This leads to practices that can then vary wildly based on who is present at the first session, how information is gathered, who is called in and so on. Ultimately, this is only an initial level of diagnosis, elementary even.

Naturally, we don’t disregard classic psychopathological diagnosis, either: in fact, we have recently had the great fortune to carry out some research -something not all experts in the field are able to do- into many case studies of the same nature. As per cultural tradition, we see many different patients suffering from anorexia, restricting type, thus we can make our observations and keep a classification for example of the last thirty or forty of them.  Lately, we took the most recent thirty-four restrictive subtype anorexic girls that we worked with in therapy in order to study what they did and did not have in common, specifically looking for the so-called redundancies, or repeating phenomena. By studying the lives, illness and families of these thirty-four girls who share the same diagnosis, what do we see? This is yet another way to use categories for research purposes.

Then we have the third level in the diagnostic category, which is dedicated to attachment theory. With attachment theory, simply distinguishing between an ambivalent/resistant pattern and an avoidant pattern is very useful. For example, we are aware that avoidant patients tend not to ask for help, making it much more difficult for them to be dragged into psychological evaluation, while ambivalent patients do tend to ask for help but are left unsatisfied with the results. This is furthermore important as a criterion in conducting a first session since we know that, for avoidant patients, initial reception, especially acceptance, is key, whereas with patients exhibiting attachment patterns consistent with ambivalence require a more direct, guided approach, to provide a more immediate solution to their request for help. Quite simple guidelines.

As for the fourth level, it has proven to be quite innovative in the systemic field but not so much in the field of psychotherapy in general, by integrating personality diagnosis. The theme of integrating the individual and systemic family history has been and continues to be very important for us. With personality disorders in particular we can observe, for instance, some narcissistic traits in a patient at the beginning of a session that can, in turn, paint a general picture of the kind of family history they may have had. This helps to speed up the thinking process and make hypotheses more rapidly, even in regards to the fifth level of diagnosis, which has always characterized the systemic method and it is, of course, systemic diagnosis.

This is yet the fifth categorization. But what is it, really? Systemic diagnosis is all about reading a family’s pulse. If a sixteen year old is brought before me, what kind of world has he been living in? What position does he occupy within the family hierarchy? Here is a different level once again, categorizing the specifics of roles in a family setting: being the youngest of three brothers in relation to the problems that occur between parents, being the eldest child dealing with a mother whose marriage is going through a crisis, the consequences that arise from such situations and so on. We’ve now clearly seen how synergizing systemic diagnosis with individual diagnosis, with attachment theory, with psychopathology and so forth, is much more interesting from both a research and a clinical standpoint.

The sixth level is the so-called ‘Three-generational’ one. What does it mean? That each of us, before becoming parents, were children once and, as such, went through certain experiences, suffered particular traumas, and learned lessons that will inevitably impact on our behavior during parenthood. For instance, if I had been encouraged as a child to learn to mostly fend for myself, then obviously I would pass on the same lesson to  my children. However, this could lead to negative consequences, like underestimating my children’s hardships and thinking that, just as I had to deal with it on my own, so should they, when maybe what they actually need is a different kind of support to what I received. This example demonstrates how the three-generational level, the sixth level of diagnosis and categorization, can be vital.

Finally, and possibly most important of all, is the last level , which we already briefly touched on during this interview: the therapist’s own emotional state, which essentially concerns how the expert in question classifies the people he sees based on his own feelings . This is the reason working with a team is so valuable, indeed it is the key point of our chosen approach, because the therapist never truly knows how much involvement his personal history and resonances have with the people he or she meets or the way he or she feels about this patient or that family. Say I meet a certain gentleman who speaks in a Milan dialect and somehow, consciously or not, he reminds me of a cherished uncle, leading me to instantly take a liking to him. I might then realize, however, that a colleague in my team feels completely the opposite way about him. This is how interesting it is, making diagnoses on this level. On the other hand, there might be times where someone who seems quite unlikeable appears that way to the whole team, at which point it’s highly probable that this reveals something of the way this person interacts with others.

We shouldn’t fall into the common mistake, which we often observe in students of using diagnostic categories obsessively. A balance must be maintained, to ensure we only use diagnostic categories so long as they keep our vitality and creativity strong.

GC: Let’s talk about follow-ups. What methods do you suggest?

MS: Well, follow-up methods require a fundamental principle that is quite difficult to put into practice very often; which should be to have multiple viewpoints on the same situation. At its core, there simply aren’t enough follow-ups happening. In my experience, the overwhelming majority of psychotherapists haven’t the faintest idea what happens to their patients five, ten, twenty years after the end of the therapy. If instead all therapists systematically kept tabs and checked up on their patients’ long-term outcomes, then it would be a great help to psychotherapy all around because there would be a more serious, more systematic approach to make something constructive out of the pain of failure. Follow-ups teach us so much, which makes it imperative to find a way to keep in touch with our patients a year or two after therapy ends and, if possible, over even longer periods of time. To tell our patients, perhaps, that we are more than happy to hear from them, even after a long time, is truly the best way that I believe we can learn many things. Undoubtedly the most difficult part of that, though, and something I personally always try to do: is to find more information from other collateral sources, so not just hearing from the patient himself but also from his general practitioner or a family member, in order to get the full picture. Sometimes the patient himself may tell us what he thinks we want to hear, so getting other opinions gives us the certainty of knowing what happened to the person in question and the long-term effects of our work. Perhaps something more should be done about this as a category, to fully back follow-up work.

GC: During scientifically conducted research studies, how does one move beyond a certain categorization to another?

MS: I would say this question is connected to the first one because there are, at times, categorizations that hinder more than they help. The most current case of this, as far as my line of work is concerned, is in the field of diagnosis and personality disorders, specifically making use of borderline personality disorder or "masochism". The borderline diagnosis has become vastly overused. When one realizes that nearly fifty per cent or more of all patients are diagnosed with BPD, one wonders what the diagnosis is even for anymore, if all cases are just "borderline" anyway. Quite fittingly, a recently released book by author Cancrini is titled “L’oceano borderline” (The Borderline Ocean), an incredibly apt description for the situation. This is one of the reasons that lead certain categorizations to be scorned, when they fail at recognizing specific phenomena. If something is equally applicable to everything then it becomes useless and this is simply what happens in general, isn’t it? The other criterion, on the other hand, within the boundaries of borderline or masochistic cases, is that these definitions have been stigmatized, that is in current parlance they are equated to an insult of sorts, like saying that someone is crazy, unreliable, aggressive and just a bad person all around.  If patients are already forced to recognize their psychological problems, having to deal with a disparaging label  as well will not help them and is not an ideal premise for therapy.  Not to mention the fact that names have importance. With this in mind, I have been working for the past few years to substitute this borderline label with a ‘post-traumatic personality disorder’ diagnosis. It seems to me to be much more constructive this way: getting the message across immediately, that these could be the consequences of trauma throughout a prolonged period of time is already a small step towards a slightly more empathic viewpoint, both for the patient in regards to himself and for his family towards him. Very often the traumas that these men, women and children went through are thoroughly trivialized and disregarded; which means that, as a first step, it becomes necessary for them to recognize that there have been traumatic events in their lives. With this acknowledgement, they have a good head start on their therapeutic alliance. Categorizing and changing the name also changes one’s way of thinking, changes one’s emotional reactions, making them more productive.

GC: Still on the topic of psychotherapy research studies, what can the qualitative and quantitative methods

offer us?

MS: For me, the problem with psychotherapy research studies is that, currently, there are only qualitative methods in the field. There are many real experts out there that do case studies, publish their findings on them and get results this way. Yet, I am also quite willing to read books by authors who follow approaches different from mine and reflect on certain issues, such as personality disorders, which have been the topic of my research in recent years. By contrast, quantitative research studies are basically non-existent. The very serious problem afflicting the field of psychotherapy is that the only research studies in existence are always connected to the pharmacological area, studying the combination of drugs and psychotherapy. Whereas there is, sadly, no one carrying out research studies that seriously compare one method, for instance the psychoanalytic one, with a cognitive method or a systemic method. A patient who, for instance, contracts restrictive subtype anorexia, might end up in psychoanalytic therapy just as easily as cognitive therapy or systemic family therapy, without anyone having the faintest clue of what could be the most effective means of treatment. Nor is any research being done into it at all. I find it to be completely absurd, that there are no organizations or institutions that could seriously make these comparisons. Apparently, it does not interest anyone in particular. The last I read about a study like this, it was done in London twenty years ago on depression and it compared couples therapy with pharmacological therapy and individual therapy as possible treatments. The results that came of it were very, very interesting and, speaking for myself and some of my colleagues, we were quite pleased to see just how superior couples therapy proved to be in this regard. However, the “London Depression Trial” carried out by Eia Asen and her colleagues is really the only research study I’ve heard being talked about in the last twenty years, by whatever miraculous circumstance it was even approved to begin with. This is a significant issue to my mind and I hope that, sooner or later, we will have the necessary institutions to deal with the problem, since no single individual could possibly carry out these studies alone. 

GC: Can I just say something? There are times, however, when one can find follow-up studies being done on the various psychotherapy techniques.

MS: Yes but it’s unfortunately quite rare. I am always on the hunt for these sorts of studies to use them and to gain some measure of comfort from them about the work we do. True, sometimes you can discover some interesting things in international reading material. My main complaint is just that it happens so very rarely. The most recent example that I can think of, which is not even that recent, was a research study done in an English hospital on restrictive subtype anorexia. The only problem was that it was not clear what form of family therapy was being used and consequently the therapeutic procedures were also not very clear. Despite this, the study produced data that I thought was illuminating: namely by demonstrating that the younger the anorexic girls were, say between the ages of about ten to twelve years, the more effective family therapy was over a short period of time. This was fascinating to me because it perfectly coincided with our own experiences on the matter. We don’t really get to see many girls in that age group, since the girls suffering from restrictive subtype anorexia that come to us are, on average, fourteen to seventeen years old; which is the more typical age range. The few we do see, however, such as a twelve year old recently, often give immense satisfaction as they show quick results. It is a shame that, as I’ve already mentioned, cooperation of this kind between researcher and psychotherapist is so rare. The majority of studies carried out for psychotherapy don’t really grab a specialist’s interest since they don’t really result in concrete ideas or even prompts. They mostly seem tailored to a very specific context, which means we can’t meaningfully benefit from them. I think this is the biggest limit to psychotherapy research studies in trying to somehow come up with new tools for experts to use in the field. As usual, the professional field and the research field tend to be too disconnected from each other.


GC: Incidentally, this leads into our last question: in what direction do you think or hope research studies in psychotherapy will go?


MS: In a way, I suppose I have already answered the question to an extent with what I already talked about, in that I truly hope that there could be some serious comparative work done in the near future. Psychotherapy is focusing on creating certain protocols at the moment, and rightfully so, something that we wholeheartedly agree on. It is important to establish the exact parameters of what a procedure is and can be. If we take, for instance, a teenager suffering from narcissistic personality disorder, what is the optimal approach here? What combination of individual and family therapy is required? How long should the therapy go on for? What are the therapeutic factors that need to be taken into account? What will we attempt to change on an individual level? I believe there is much potential for research studies in psychotherapy, as there are more and more methods where these things are specified. Actually weare currently working on making increasingly specific protocols available for our students and colleagues to use. At this moment in time, for example, I am preparing to publish an article on teenagers who are brought into therapy at their families’ behest. The best possible method to use in these cases involves an initial joint interview with both the children and the parents and, based on the results of this early evaluation, there may be different alternatives to follow up on. But if there must be some kind of protocol then I maintain that this one, having a joint first interview, is much more effective than the standard one, which is a one on one interview with the teenager in question. Many of my colleagues would, perhaps, meet only the parents first, before seeing the teenager individually. My belief is that this is not the right way to go about things in practically any case, but especially the ones where the initial interview with the teenager, who has been forced into it, is done alone. Granted, this is merely my personal experience based on my own work but it would certainly be intriguing to have systematic comparisons being made between different procedures: ie conducting a first interview while organically establishing diagnostic categories like teenagers with personality disorders, drug addiction, anorexia or other types of addiction and studying the different protocols to be used with each category. On the bright side, I see many different schools of psychotherapy constantly creating more and more precise protocols. Gone are the days of having patients lie on the couch and letting the chips fall where they may, the whole thing has become quite unacceptable at this point. These days therapists are called on to clearly state the criteria they will be using with a certain level of accuracy. Yet because there are still so many various ways to decide on what to do and say, which procedures to employ and so on, I believe that, as I was saying before, the future of research studies should hinge on comparing these different methods. Really studying the pros and cons, of course, but also the different settings as well.  Obviously we then have to also consider what we do in these sessions in-depth and it needs to be made as clear as possible, though I think it’s safe to say that psychotherapy itself is heavily divided on this between the various schools. I regret to say that comparing and contrasting research between different schools is simply not done, at least not at the moment.

Published online: 14 May 2014.

Copyright: © 2014 Matteo Selvini and Germana Cavallini. This is an open-access article distributed under the terms of the Creative Commons Attribution License. The use, distribution and reproduction of this article is permitted, provided the original authors and licensor are credited and that the original publication in E-Journal of Psychotherapy Research is cited in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.


Comments: 0

Leave a Reply

Your email address will not be published. Required fields are marked *


  • Editorial Board


    Matteo Selvini, Scuola di Psicoterapia della famiglia Mara Selvini Palazzoli, Milano

    Scientific Editors

    Grazia Attili, Sapienza Università, Roma

    Alfredo Canevaro, American Family Therapy Academy, Buenos Aires

    Juan Luis Linares, Università Autònoma, Barcellona

    Marco Vannotti, Cerfasy (Centre de Recherches Familiales et Systémiques), Neuchâtel