Richard L. Rubens, Ph.D.



It is my belief that psychoanalytic supervision is a psychoanalytic endeavor, rather than a didactic one.  Put succinctly, I view such supervision as psychoanalytic therapy for one’s work.  I shall first state, in the following two paragraphs, what I believe to be the essence of the position, and I shall then proceed to take up and expand upon various of the specifics of how I understand and undertake to structure and participate in such endeavors.

I view the supervisory process in much the same way I do the analytic process itself:  it is an intimate, intense relationship, conducted within the safety of the relationship’s confidentiality, and following its own, mutually determined course.  In this relationship, both participants bring to bear the fullness of their experience of the material under consideration –and of the process of considering it together.  Each is challenged to be as open to –and with–  his or her experience as he or she is capable; and each is also challenged to bring all of his or her skills at detailed inquiry to bear on the examination of all aspects of that experience.

The fundamental difference between analysis and analytic supervision lies in the agreed upon scope of the endeavor:  in psychoanalysis, the field of inquiry specifically needs to be unlimited; while in analytic supervision, the agreed upon field of inquiry is essentially predetermined as that relating to the supervisee’s analytic work.

Personally, I am someone with highly developed theoretical interests and philosophical concerns.  My writing and teaching on the theories of Fairbairn and on the interactive nature of  transference and countertransference express the central thrusts of this theoretical interest.  Nevertheless, I do not deem supervision to be the appropriate forum for the pursuit of such interests –any more than I believe psychoanalysis itself to be an appropriate arena for such foci.  While the specific developments of the course of a supervision (or of an analysis, for that matter) may bring the theoretical positions of one or both of the participants into consideration at any particular time, such considerations have no intrinsic roles in the process.  To reiterate what I have stated earlier, analytic supervision is not a didactic process in my view.

In fact, I go so far as to assert that a psychoanalyst’s theories and metapsychology are nothing more –or less– than an aspect of his or her countertransference.  They are a sometimes highly intellectually elaborated countertransference, to be sure; and, at times they are a potentially useful countertransference.  But, to the extent that theory and metapsychology are relatively enduring patterns with which the analyst approaches his or her experiences and through which he or she shapes and understands that experience, they must be considered to be part of the realm of countertransference.[1]

Thus, in supervision, there will be an interaction between the theoretical positions of  supervisor and supervisee (in much the same way as there will be such an interaction between the theoretical positions of analyst and analysand).[2]  The interaction of these positions represents, however, nothing more or less than an aspect of the overall interaction of the personalities and personhoods of the two participants in the process.  As with any aspect of the interaction, the interaction between the theoretical positions represents an appropriate subject for inquiry.  As with all psychoanalytic inquiry, however, the most important level of the inquiry will relate to the transference-countertransference interaction embodied and expressed in the subject.

Psychoanalytic supervision must never have as its aim the inculcating of any particular theoretical position.  Just as psychoanalysis must place no value above that of the mutual pursuit of the unique experience of the relationship,[3] so too must analytic supervision recognize and respect the inherent potential for two participants adhering to different views and theoretical positions to create growth from their psychoanalytic exploration of their joint experience.

Since I hold that psychoanalytic theories and metapsychologies are countertransferences –that is, that they are aspects of personality and forms of experiencing and understanding one’s interactions with others, it is my conclusion that the value of such theoretical positions lies in their utility for the individual holding them.  It is my opinion that the true measure of such theories is how well they fit or do not fit with the personality of a particular analyst.[4]  There is simply nothing worse than an analyst attempting to utilize a theory that is not congruent with his or her personality.  And certainly one of the most common experiences I have had in which theory directly enters the focus of analytic supervision, has been that of a supervisee progressively realizing that some aspect of his or her technical or theoretical commitments does not fit with his or her deeper beliefs about the nature of human beings and human interactions.

This sort of process is most easily demonstrated in terms of technical issues (only because the broader, theoretical versions are more deeply and intimately bound up in the personhood and privacy of the supervisee).  Take, for example, the case of a supervisee who has come  with a technical commitment to not sharing emotional reactions with patients.  She has been working for three years with an analysand who, while obviously extremely positively attached to the analyst, needs to deny any emotional connection to the analyst and is particularly insistent in pointing out the analyst’s complete lack of emotional attachment to her.  Now it should be understood that this analysand is, in fact, very dear to the analyst:  the analyst often has extremely warm, empathetic responses to this analysand, and actually feels very emotionally close to her on many levels.  A sad event occurs in the analysand’s life, and, while the analyst is very capable in her working with the issues of the event, she feels bound by her technical commitments not to express the rather intense sadness she feels for the analysand’s loss.  This is a moment where the analyst feels acutely a divergence between her human impulses and her understanding of  the requirements of technique.  Were this situation simply one in which her deep, theoretical commitments forced her to do something personally difficult, she most certainly should have been supported in her technical stance.  In actuality, however, the analysand had been wrestling throughout much of the supervision with the fact that she was operating with technical positions dictated by  her prior classical analytic training that no longer seemed warranted given the shifts in her theoretical  stance.  The immediacy of this moment in the analysis led her to decide to return to the following session with a more emotional acknowledgement of her feeling in the situation.  While specifically this acknowledgement was little more than a statement of how sad she had felt thinking about the analysand’s loss, it was understood correctly by the analysand as a declaration of emotional involvement and attachment on the part of the analyst.  While it was clear that the analysand had always on some level known this to be true, the analyst’s ability directly to deal with it for the first time changed a very fundamental transference-countertransference balance that resulted in major progress in the analytic relationship.  Moreover, the analyst began to recognize that her adherence to the more distancing aspects of classical technique was not only incongruent with her changing theoretical beliefs, but in fact represented a particular resistance she had to fuller engagement and intimacy in the analytic relationship (and, with the analysand in question, to identificational elements she had previously been avoiding).

Lest this initial discussion be misleading, allow me to remind the reader that I am merely making the point that theoretical issues can appropriately enter the arena of supervisory exploration –while it is my more basic assertion that theoretical and didactic discourse have no intrinsic place in the process.  In the actual conduct of  the analytic supervision I do, the focus I attempt to create is elsewhere.

I have stated earlier that I believe the supervisory process to be the mutual creation of the two participants therein.  As in psychoanalytic process similarly conceived, however, this does not suggest that the supervisor has no role in shaping and structuring the enterprise.  There are any number of requirements I have in order to make the work possible from my end, and any number of requests I make on anyone who undertakes to enter supervision with me.  Conversely, I recognize that there may be any number of requirements, needs and preferences that the supervisee may have.  As in analysis, these are facets of the contractual agreement between the two parties, and in no way compromise the mutuality of the enterprise.  And while there may be many areas of asymmetry in the relationship,  they in no way compromise its equality.

I am adamant, in structuring analytic supervision as well as analysis, that those things that I require, need, and prefer reflect my own needs in undertaking the work –rather than representing some ‘correct’ or more generally ‘necessary’ principles.  My need to collect a fee, my policies about missed sessions, my setting a specific time period aside on a recurring basis– all of these represent manifestations of my personal needs.  I believe that it completely valid for me to have and seek to satisfy such needs, but I do not attempt to represent them as having any special standing or to justify them as having any absolute validity.

When doing analytic supervision, I insist that the work be about the supervisee’s relationship with a single patient.  This preference reflects my belief in the crucial importance of examining the nuance of the transference-countertransference interaction –on the first level, between the supervisee and his or her analysand; and, ultimately more importantly, between the supervisee and myself.  It also contains within it my belief that the goal of supervision is the growth of the analyst, not of the analysand.  It is actually an unavoidable fact that the supervisory process interferes in the relationship between the supervisee and the analysand.  (There are, fortunately, many instances in which this interference falls into the range of neutral to positive; but it must be remembered that it is not uncommon for the presence of a third party in the intimacy of the analytic relationship to adversely affect that relationship, at least in the short term.)  So, while hopefully the supervisee’s work with all of his or her patients will improve and grow as a result of the supervisory experience, the improvement of the work with the particular analysand upon whom the supervision is focused is not primary, and certainly is not assured.  There is a direct analogy here to the psychoanalytic process itself:  although an analyst may focus for extended periods of time on an analysand’s relationship with a particular individual in the analysand’s life, the outcome of the relationship with that particular person is not ultimately what is most at stake, but rather the growth that is possible in the analysand’s relating more generally.[5]

Contrary to the current legal realities, responsibility for what transpires between an analyst and analysand can only be the responsibility of that particular partnership.[6]  To see the supervisor as “responsible” for what actually transpires between his or her supervisee and that supervisee’s patient defies logic.  In the first place, all that a supervisor really ‘knows’ about the patient in question is what he or she learns second hand though the report of the supervisee.  Furthermore, it is illusory to believe that a supervisor in actuality has any real ‘control’ over the actions of the supervisee.

My insistence that supervision focus on work with a single patient (particularly in situations where this requirement is not already dictated by training requirements) also serves to throw into immediate focus some of these issues about the nature of analytic supervision.  In trying to decide which analysand to present in supervision, the supervisee must grapple with the issues implicit in why one would choose the work with a particular analysand to present.  Does one choose work that is going particularly well, or particularly poorly; does one choose a relationship that is relatively typical in ones practice or atypical– the choice reflects much of one’s preconceptions about the process.  It also begins to sound themes about how a supervisee feels about his or her work and about the process of supervision.

My second requirement is seemingly somewhat paradoxical.  I request that a supervision work as much from spontaneous recall as possible; and, at the same time, I expect that the supervisee present the work in as nearly verbatim form as possible.  I do not ever agree to listen to tapes, I most strongly request that supervisees do not take notes during analytic sessions,  and I prefer (although I do not insist) that he or she not present from notes (viz., those reconstructed after the analytic session) in the actual supervision session.  Since my intention is to assist in the growth of the supervisee’s actual functioning as an analyst, I attempt to recreate the experience as it actually exists during the process.  The point is not at all to minimize the necessity of detail and accuracy, but rather to structure a way to make this sort of detail and accuracy more organically a part of the supervisee’s ongoing functioning.  In truth, there are exceedingly few practicing analysts who really work in the way much of supervision encourages:  we remember what detail we can during the actual encounter of the sessions, we do not refer to our notes.  The immediacy of our reactions to our analysands comes not from a written catalogue of notes, but from our living memory.  I shall return later to the second half of this apparent paradox, the importance I place on specific details.

I first feel that I must admit here that my aversion to having supervisees take notes during sessions represents a specific prejudice more that just a technical position.  Whereas many of the positions I have enumerated in the preceding paragraph can be understood as attempts to replicate the actual process of recall in analysis outside of supervision, this position can not be justified on these grounds.  (Analysts can and do take notes during sessions aside from when they are in supervision.)  It is simply true that my own bias is that note-taking during sessions is so necessarily distancing from the immediacy of the relationship, and that the immediacy of the connection between analyst and analysand is so essential to the way I work, that I have strong feelings on the subject.  I include this extended discussion of this anomaly because I believe it serves to demonstrate another aspect of my approach to supervision:  I believe that it is best to have such prejudices and feeling out in the open as much as possible.  I would not ‘forbid’ a supervisee from taking notes during sessions –and, in fact, some whom I have successfully supervised have chosen to continue to do so.  I do, however, want the supervisee to know that I have a strong opinion in the matter.

This brings me to yet another request I make very directly of my supervisees:  that they know to expect that I give my opinions and share my reactions rather strongly and directly, but that I do not expect them to treat them as anything other than opinions.  I expect that there is good reason for  people with whom I work to take seriously the opinions and reactions that I have –and I believe that there is equally good reason for me to treat their opinions and reactions in the same way.  But I certainly do not believe that either of us should take the other’s reactions and opinions as having any more validity than our own.[7]  Just as I believe that theory and metapsychology are countertransference, so, too, do I hold that all that I think or say as a supervisor is nothing more or less than a secondary countertransference to that which the supervisee is bringing.[8]  This belief is one I make explicit to supervisees at the very outset of our work together.  It is my ongoing experience, however, that it only becomes a fully appreciated, believable aspect of our relationship once it has been lived through in some meaningful way.  (And, as one might imagine, this happens at different points in supervision for different supervisees.)  It is one of my fondest beliefs, however, that almost all of the supervisees I have worked with over the years have come –at least in the end– meaningfully to know the reality of this position.  I have no desire to have supervisees agree with or adopt my opinions.  My main objective is to engage in an experience in which they confront these opinions in a way that hopefully will result in a deepening and refining of their own opinions –however different they may be from my own.

I believe this to be so important, because I am convinced that the direct sharing of one’s opinions and reactions represents the first of two underlying dimensions that comprise the essence of the analytic interaction –and, therefore, for me, of the supervisory interaction as well.  This first dimension is the spontaneous, associative one.  It consists of all of those feelings, thoughts, associations, speculations, day dreams, fantasies, and every other form of psychic activity we have in the presence of another particular person.[9]  One can never know if such phenomena actually relate to the person in whose presence they occur, or, even if they do, whether they say much about that person, as opposed to what they say about the person having them.  Nevertheless, it was with good reason that Freud made it his basic rule of psychoanalysis that patients should share these thoughts and feelings without subjecting them to censorship of any kind, for he knew that to provide any rationale for dismissing them would mean that patients would use that rationale to exclude virtually everything important from analytic examination.  While analysts have many varied positions on how much of their own spontaneous thought feelings to share directly with their patients, increasingly there has been fairly general recognition that these thoughts and feelings must be closely attended to, at very least internally.  (It should be noted that much that passes for interpretation, and essentially all that falls into the category of dream interpretation, is  in reality a version of this spontaneous, associative dimension –no matter how many attempts are made to disguise the fact or authoritarian claims are made to deny it.)

While one would think this issue would be less difficult in supervision, it often is still highly charged.  There is often a similar reluctance to acknowledge that the majority of the supervisor’s reactions and ideas are essentially feelings and thoughts of this spontaneous, associative kind. While it might be reasonably claimed that a supervisor’s thoughts and feelings should receive a particularly attentive hearing by reason of the supervisor’s experience and expertise, there is still no reason to accord them any absolute authority.  In fact, the main reasons they deserve attention is, first, because the supervisor is a legitimate participant in the dyad, and, second, because by not being subject to the same repressions as the supervisee, he or she should be able to notice things that are outside of the supervisee’s awareness.  It should be noted, however, that these same reasons apply with respect to the value of the supervisee’s thoughts and feelings![10]  One of the central thrusts of my work as an analytic supervisor is to attempt to make clear the utility of recognizing, acknowledging and sharing the thoughts and feelings of both parties.  I insist on two things in relation to these thoughts and feelings:  one, that they have value simply because they occur in the context of relating to this other particular human being; and, two, that they have no special status beyond the simple fact that they have so occurred –which is to say that they are nothing more or less than spontaneous, associative material.

The second of the two underlying dimensions that comprise the essence of the analytic interaction –and, therefore, for me, of the supervisory interaction– is detailed inquiry.  In order to be truly meaningful, the spontaneous, associative material needs to be subjected to a thoroughgoing process of detailed inquiry.  In the mutually conceived relationship of analytic supervision, both parties bring to the process their spontaneous, associative material and together subject the various aspects of their separate and joint experience to a rigorous examination.  They must set about to understand what their respective experience actually is as well as what the experience of the other is –each in the terms of the personal experience of each participant.  The goal is not consensus, but rather an in depth exploration of how their experience relates –in its similarities as well as its differences.

I have mentioned earlier my belief in the importance of presenting clinical material in supervision in great detail –as close to verbatim as possible.  The process of detailed inquiry necessitates this level of detail.  The awareness of this reality is one of the most important legacies of the Interpersonal Psychoanalytic tradition.  Without the detail, one simply cannot begin to know what the experience is truly about.  My eschewing the use of tapes and notes is meant in no way to lessen the level of detail, but rather to shift its nature.  The subjective element, even of detailed recall, is no small part of the material.  For example, I find it more important to have a supervisee recognize where the lacunae are in his or her recall, then to have an artificially complete record separate from his or her personal experience.

While the apparent subject matter of an analytic supervision is the supervisee’s work relationship with his or her analysand, the actual subject matter eventually is the relationship of supervisor and supervisee.  In this way, the supervisory relationship almost completely parallels the analytic one.

Which brings us again to the central and crucial difference between psychoanalysis and psychoanalytic supervision: whereas in the former, the scope of the mutual exploration is by definition without preset boundaries, in the latter, the scope of the exploration is limited by contractual agreement to the supervisee’s work.  Now, to be sure, the full understanding of a person’s analytic work potentially can include every aspect of his or her personality, since all of an analyst’s personality most certainly comes into the analyst’s work in one way or another.  And, in any particular supervisory dyad, there may be great latitude appropriately permitted in what is open for exploration.  Nevertheless, the only expectably agreed upon area of exploration for any supervisory relationship is a more limited version of the area of psychoanalytic work.  If the exploration is to go beyond these immediate limits, it has to be with the explicit agreement of both participants, and with the absolutely explicit permission of the supervisee.  In this matter, the supervisee must have absolute veto power. Should the supervisory relationship be such as to make attractive the following of some issue beyond the more generally accepted limits of supervisory exploration, it must be because the trust, closeness, and previous positive history of that relationship supports going farther.  But, even then, the supervisor must respect absolutely the supervisee’s decision to call a halt or even sound a retreat.

Since trust and the reasonable expectation of safety plays such a crucial role in any analytic exploration –be it supervisory or therapeutic, confidentiality understandably is a crucial concern.  Confidentiality is by definition an issue in supervision, even only because the basic material the analyst is bringing to the supervisor from his or her work already requires such protection by law.[11]  But the confidentiality of the supervisee him- or herself also needs protection if the process can be successful in progressing in an analytic way.

There is, of course, a built-in limitation to the confidentiality of the supervision conducted as part of any training program:  in addition to the growth function of the supervision, there is also a necessary evaluative function.  The legality of this component is, of course, of no consequence:  it simply is made clear as part of the particular contractual agreement between supervisor and supervisee that this evaluation and reporting component will exist.  The more difficult issues involve the extent of the agreed upon exception and its subject matter.  In my understanding, at the very most, the exception to confidentiality relates only to those explorations having directly and specifically to do with work issues, and never to any explorations that are undertaken beyond or outside of that narrow boundary.  I always show a supervisee any report I am going to submit in advance of submitting it, and I am also quite explicit as to what I intend to say verbally.  Moreover, I always make sure to share with a supervisee exactly what I  have said about them in any evaluation meeting, after the fact.  It is a compromise that I discuss openly and in great detail with my supervisees, and it has functioned quite satisfactorily for me and them over the course of my many years doing analytic supervision.

As I imagine is obvious from what I have presented, I am rather extreme in my defense of the confidentiality of those with whom I do any form of analytic work.  The reason, too, should be clear:   I believe this condition to be an absolutely necessary condition for the process of psychoanalysis and of psychoanalytic supervision.  Accordingly, I have chosen to present a basically theoretical description of my views on psychoanalytic supervision and been most sparing in my presentation of  clinical examples.


Return to Richard L. Rubens PUBLICATIONS Page.


[1] It must be noted that I utilize here –and elsewhere– an inclusive definition of transference and countertransference:  I understand these categories to refer to the whole of a person’s organization of experience and not to any “distorted” or otherwise specified segmentation of that experiential organization.

[2] I adhere to the belief that analysands have their theories and metapsychologies –albeit that in the case of the latter these may or may not be as intellectually or consciously conceptualized.  (One might note that this may also be true of certain analysts!)  But many of our analysands are themselves students of analytic theory.  Moreover, there exist analysands who are in no way conversant with the corpus of analytic theory and yet have highly developed theories of human personality functioning that put to shame the theoretical thinking of many analysts.

[3] I am opposed to any value being appended to the analytic endeavor.  One can be a Marxist (for which can be substituted “feminist,” “segregationist,” or “integrationist,” etc.) who is an analyst, but it is problematic in my opinion to identify oneself as a “Marxist analyst,” as this amalgamation suggests that there is some other competing value that is being given equal billing with the values inherent in the analytic enterprise itself.

[4] The question may reasonably be raised as to whether certain personality configurations may be so antithetical to the therapeutic endeavor as to be considered inappropriate as the basis for a theoretical style.  While this, of course, can be true, it does not mean that therapists with such personality configurations would be better off compounding the problem by covering them with an incongruent theoretical position.  It simply means that such people are not good therapists.

[5] This assertion is not to be construed as suggesting that the importance of any particular relationship in an analysand’s life is not to be seen as important.  It suggests only that it is a mistake for the analyst to become too focused on or dedicated to the maintenance of  any such particular relationship.

[6] While the analyst may have special responsibilities within the relationship that apply only to the analyst (e.g., for maintaining confidentiality and for not abusing the analyst’s position in the relationship, to mention two extremely important ones), the responsibility for the overall life of the relationship must be a joint one.  This in no way contradicts the fact that the two participants may have a whole range of individual responsibilities to which they have contractually obligated themselves.)

[7] In this regard, I have often been known to quote the Marx Brothers to a supervisee (or analysand):  on one of the many occasions that Groucho is trying to con Chico into agreeing to something against his better judgment, Groucho says to him, “Who are you going to believe –me or your own eyes?”

[8] To be completely accurate, I must distinguish that this is true of only my reactions to the supervisee’s patient.  My reactions to the supervisee him- or herself, of course, represent a primary countertransference in their own right.

[9] I say “in the presence of” another particular individual because I am specifically avoiding the phrase, “in reaction to.”  While, of course, I understand, that it includes things that quite literally are “in reaction to” another person, and while I recognize that not all of that to which I am referring is quite literally “in the presence of” the other, I am very pointedly trying to be clear that I believe it to be a tremendous error to assume, in general, that the phenomena I am describing are necessarily reactive.

[10] The reciprocal nature of the latter reason was best elucidated by Margaret Little, when she noted that viewing one’s unconscious was a little like trying to see the back of one’s own head –it was much easier to see the back of someone else’s.

[11] It is very disturbing to hear supervisor’s presenting clinical material gleaned from their supervisees.  As difficult as it is to present clinical material from one’s own psychoanalytic work in such a way as to insure the confidentiality of one’s patients, it is virtually impossible to do so with work that is an extra step removed.  Understanding that one simply cannot adequately disguise clinical material in a  way that insures it will not be recognized if the hearer knows the patient in question and knows that he or she has been in treatment with a particular person, it becomes very risky business indeed to present material from which one is that far removed: one simply does not have any idea to whom the situation may be familiar.  At very least, supervisors need to be more guarded about sharing material from their supervisees than from their own work.