An edited version of this paper appears in Modern Psychoanalysis 25, 1 (2000): 43-51.

Leaving Development to the Developmentalists:
Sticking to What We Know in Psychoanalysis

Donald L. Carveth, Ph.D.


Even if it were true that before around 14 months infants are incapable of coding experience in ways that are recoverable in analysis, nonverbal communicative processes of emotional induction or projective identification may play an important role in our work with certain, perhaps even all, of our patients. But it may be an error to assume that such NONverbal processes equate to or dervive from primitive, PREverbal processes and experiences. They may involve subtle and intelligent nondiscursive symbolic forms. Since we are analysts and not researchers in developmental psychology, perhaps we ought to remain within our area of competence, working in depth with the unconscious in the here and now of the analytic relationship, recognizing our speculations about the past as merely that, and leaving the issue of their validity to those who are interested in and capable of conducting scientific research in this field.

On Saturday, December 5th, 1998, the Manhattan Center For Modern Psychoanalytic Studies sponsored a conference at which Dr. Phyllis Meadow presented a paper entitled "Through the Heart or Through the Mind" to which Drs. Charles Brenner and Peter Neubauer responded as invited discussants.

In the following, I will not comment on Dr. Neubauer's interesting discussion because I wish to focus upon questions raised by Dr. Brenner's discussion of Dr. Meadow's paper.  Although Drs. Brenner and Meadow did not exactly enjoy a meeting of minds, after delivering his admirably clear but fundamentally critical discussion, Dr. Brenner returned to his seat beside Dr. Meadow and warmly patted her shoulder as he sat down.  As Dr. Mimi Crowell later commented from the floor, their intellectual and verbal differences concerning the role of nonverbal behaviour and communication in psychoanalysis did not appear to stand in the way of their own nonverbal connection.

Dr. Meadow, like other Modern Analysts, is interested in the "preverbal" or "preoedipal" patient who communicates more through emotional induction or projective identification than through words.  (I am placing the words "preverbal" and "preoedipal" in inverted commas here for reasons to be explained below.)  As is suggested by her title, "Through the Heart or Through the Mind," it was clear that Dr. Meadow thinks Modern Analysis seeks to reach the "preverbal" patient through the heart--i.e., through emotional communication--while classical analysis seeks to work through the mind, through verbal interpretation, with the result that classical analysis cannot work effectively or at all with the "preverbal" patient.  Dr. Meadow gave examples of emotional communication with "preverbal" patients and she indicated that she views their transferences, enactments and symptoms as repetitions of "preverbal" trauma and of "preverbal" drive tendencies toward life, connection and involvement on the one hand, and toward death, disconnection and withdrawal on the other.  (I leave aside here other interesting metapsychological issues raised in Dr. Meadow's paper, such as her suggestion that if the ego psychologists had not abandoned an id psychology of life and death drives for an emphasis upon the ego and its conflicts with the id, they would have been in a better position to understand the fundamental conflict of the "preverbal" patient between the competing forces of Eros and Thanatos.)

Dr. Brenner responded to all this with one basic point.  Evidence from neurobiology, neuroanatomy, neurophysiology, etc., indicates clearly, according to him, that the infant brain is just not capable of recording, either prenatally or for a considerable time postnatally, the kinds of memories that Dr. Meadow and others think their adult patients are repeating.  He affirmed analysis quite literally as a "talking cure" and insisted that it could not address experience prior to the rise of speech.  Initially he specified three years of age, but later he seemed to reduce this to 14 months.  As far as Dr. Brenner is concerned, experience prior to, say, 14 months is literally a dark continent because such experience cannot have been encoded linguistically and, therefore, is unavailable to a talking cure, if indeed it could have been encoded at all.  Just as it is impossible for a baby to walk before a certain age when the brain has developed to the point of enabling this, so it is impossible for experience to be encoded in ways recoverable later in a talking cure before, say, 14 months.

Many of the Modern Analysts I overheard reacting to and discussing this debate during discussion from the floor, in the corridors, and in the afternoon workshop I attended, appeared to find Dr. Brenner's position frustrating because they felt that they have had experiences of working productively with "preverbal" patients.  They have tended to view the transferences, symptoms and enactments of such patients as repetitions of preverally encoded mental processes communicated to the analyst through emotional inductions or projective identifications--the very encodings Dr. Brenner says are impossible.

In the face of this contradiction, one possibility is that Dr. Brenner may simply be wrong. Evidence might be obtained of preverbal encoding that later emotional inductions and enactments may repeat, as both Kleinians and Modern Analysts tend to believe.  For example, a recent paper published in the British Medical Journal (and subsequent discussions in its ejournal eBMJ) offers empirical evidence showing that the suicide rate is higher in adults who had a painful birth (Jacobson & Bygdeman, 1998).  To some, such evidence might suggest that assumptions about amnesia for early painful events may be unfounded.

However, let us assume for the sake of argument that Dr. Brenner is right.  What then?  Over the past few years I have found myself repeatedly wondering whether Freud, Klein, Meadow and Brenner, together with virtually all psychoanalysts, yours truly included, may have been altogether too caught up with developmental concepts in the first place and too preoccupied in psychoanalytic work with adults and older children with attempts to link what we observe in the here and now of the transference and the countertransference with empirically ungrounded assumptions and speculations about the past.

Although Freud himself did not include the genetic perspective in his own listing of "the points of view and assumptions of metapsychology" (Rapaport & Gill, 1959), it was certainly implicit in all of his work and it can be said with justice that the developmental metaphor has "possessed" the psychoanalytic mind ever since.  But the fact is that the analytic situation offers no opportunites to empirically test the validity of our analysands' reported memories, let alone our joint constructions of the past and, hence, with rare exceptions, we have little way of knowing how accurate or distorted such memories and constructions may be.  Despite this fact, psychoanalysts have continued to extrapolate from observations in the present to a hypothetical past.  In fact, our most central clinical concept, transference, itself suggests that what we observe in the present is a more or less distorted repetition of past object-relations, a complex of attitudes, images, wishes and feelings deriving from the subject's experience of past objects displaced or transferred onto present ones that in this way come to function psychologically as their substitutes.

Is it feasible to simply study what goes on between analyst and analysand consciously, preconsciously and unconsciously in the here and now of the therapeutic interaction, observing both words and nonverbal actions and interactions, including emotional inductions or projective identifications, while foregoing empirically unjustified (at least in the clinic) conclusions that the patterns we see in the consulting room necessarily repeat experiences or even phantasies from childhood or infancy?  Although it often seems useful to speculate about such linkages, perhaps we should be clearer than we have been that speculating is what we are doing.

As Langs (1980) and Gill (1982) among others have suggested, our preoccupation with the past may well serve as a defensive, intellectualizing retreat from the threatening immediacy of the therapeutic relationship--that is, from what in our preoccupation with the idea of the past in the present we have called the transference and countertransference.  Although it seems useful when analysands themselves speculate about the possible childhood and infantile origins of what we and they observe and experience directly in the analytic relationship, perhaps we should have a sharper awareness that, in reality, the question of origins is simply not scientifically answerable on the basis of the type of data the consulting room provides.  Perhaps, at least in this sense, we should leave developmental psychology to the developmental psychologists.

It is in no way my intent to devalue developmental psychology.  My point is only that most analysts are not developmental psychologists.  What we see in adult patients and older children may or may not be a repeat of something these patients experienced, or believe they experienced, in childhood or infancy.  Freudians working with adults interpret the observable "transference neurosis" as a repetition of an infantile neurosis, the Oedipus complex, assumed to have occurred sometime between two and a half and five or six years of age.  Even if developmental psychology presents evidence that this complex really does occur in children of this age, we still have no evidence that the triangular conflicts we observe in our adult patients are repetitions of this childhood complex.  Admittedly, the hypothesis of repetition is both plausible and appealing on a number of grounds, but it remains merely an hypothesis that moves far beyond the observational data derived from the analysand's current life both in and outside the analytic relationship (the "transference").  Similarly, Kleinians working with adults or children as young as two and a half years of age interpret their behaviour in terms of phantasies alleged to have occurred, without any corroborating evidence to subtantiate the claim, during the first six months of life.

From this point of view, we are on insecure ground in assuming that the nonverbal processes studied by Modern Analysts derive from supposed "preverbal" processes.  (This is why above I placed the term preverbal in inverted commas.)  In addition, this very reduction of the NONverbal to the PREverbal reflects not only our attachment to the developmental metaphor, but also the endemic bias toward the verbal in psychoanalysis and its primitivization of the nonverbal.  The NONverbal is made PREverbal, a primitive precursor of the verbal viewed as mature, a mere preliminary to what is truly civilized and valuable--i.e., mere foreplay rather than "the main event."

The nonverbal processes studied by Modern Analysts may or may not reflect preverbal processes.  While we do know of the importance of nonverbal emotional communication in the analysis of adults and children, the truth is that we do not know to what extent if at all such nonverbal communication reflects preverbal processes.  We do not know the history of the nonverbal behaviors and communications of our patients.  We are highly inclined to speculate about the history of these processes.  But we should recognize such speculation for what it is.

And whatever its history, we may be displaying little more than our own subjective countertransference bias as "people of the word" when we label the nonverbal or nondiscursive communication in which these patients often appear to be skilled as primitive.  The fact that such patients behave in ways that conflict with our subjective bias in that they do not employ words--the one, limited mode of symbolic communication that we appear to regard as mature and healthy--does not in and of itself justify their devaluation or pathologization.

To put the point I am making into the language of split-brain research, right-brain communicators should not be devalued as primitive by left-brain analysts.  We should attempt to break our automatic association of the non-verbal with the "early" or the "infantile" or the PREoedipal.  Nonverbal or nondiscursive codes may be viable alternative modes of coding experience that are not necessarily deficient or inferior and, in some cases, may even be superior to verbal codings.  The mystics of all religious traditions have always attempted to tell us this in emphasizing the importance of experience that is beyond, not beneath, words in that it is too subtle or complex to be reduced to crude and reductive verbal categories.  And through what possible combination of words may a lover convey to the beloved what is communicated nonverbally in the act of making love?

As Suzanne K. Langer (1951) explained a half century ago--I think the fact that she is that rare phenomenon, a woman philosopher, is not without significance in this context-- it is a mistake to equate language or discursive symbolism with symbolism as such.  Nondiscursive symbolic forms such as music, dance, mime and visual art are equally, if not more sophisticated expressions of our uniquely human symbolic capacities as are linguistic or discursive forms.  So ingrained is our prejudice in favour of words that one often finds language defined in terms of the verbal symbol, thus excluding by definition such nonverbal languages as American Sign Language and Morse Code, as well as the languages of mathematics, cybernetics, dance, music and art.  Not only should we question the association of nondiscursive symbolic forms with primitive or primary processes, but equally the association of discursive or verbal symbolism with secondary process thinking, health and creativity.  The fact is that much verbal symbolism is mere verbiage, what Heidegger (1927) called "idle talk" and Lacan (1977) called "empty speech".

Of all contemporary psychoanalytic approaches, Modern Analysis, with its distinctive emphasis upon emotional communication, has been most alert to the seductions of intellectualization.   In this connection I am reminded of the anecdote reported by the Marshalls regarding the analysand who "having invested several years in an extensive psychoanalysis resignedly quipped about its effect, 'When I went into analysis I was a horse's ass.  I'm still a horse's ass, but at least [now] I know why'" (Marshall & Marshall, 1988, p. 43).  But the Modern Analytic emphasis upon the use of emotional communication and mirroring and joining techniques in situations where interpretation is unproductive or destructive has in no way displaced its emphasis upon the importance of "saying everything" or putting everything into words.  The use of induced feelings, emotional communication, mirroring and joining techniques, etc., is thought to be in the service of resolving "preverbal" resistances to progressive verbal expression (as distinct from insight or self-understanding) and it is often implied that it is through such verbal expression that the cure is achieved.

In actuality, Dr. Spotnitz (1985) did not claim that "saying everything" is in and of itself therapeutic.  He wrote:

Getting the patient to express his anger in language or behavior is viewed by some therapists as a major problem to work on ....   This view is incorrect; the cathartic approach is not curative.  The problem is, rather, to study and resolve the forces that prevent the patient from expressing the anger in language (p. 215; original emphasis).

Thus, for Dr. Spotnitz, it is not putting everything into words that is therapeutic.  For one thing, that simplistic formula fails to exclude "idle talk" and "empty speech."  Clearly, for Dr. Spotnitz, what is therapeutic is resolving resistances to progressive emotional communication.

While there is no doubt that such progressive emotional communication often takes the form of putting everything into words of an emotionally meaningful sort, I think we need to remain open to the possibility that progressive emotional communication may also occur in nonverbal and nondiscursive forms.  In a chapter entitled "The Unconscious Dialogue Based on Emotional Communication," the Marshalls (1988, ch. 4) provide clinical examples of such nonverbal, communicative processes.  Occasionally we hear puzzling reports of  cases in which very little ever gets put into words and yet the analysand improves without either party to the analysis having the foggiest idea as to why.  The answer may be that progressive emotional communication is taking place, but in nonverbal or nondiscursive forms.  Whereas the classical psychoanalytic tradition emphasized the goal of remembering rather than repeating and viewed acting-in, acting-out, and enactments in general as resistance, recently some relational, self psychological, and even contemporary classical theorists (Aron, 1996; Lazar, 1998; McLaughlin, 1991) have suggested that enactment may be viewed as an alternative mode of interaction and communication within the analyst/analysand dyad.

Another part of the answer to the puzzle as to how certain patients manage to make progress without saying very much at all, let alone saying everything, may lie in another principle of Modern Analysis, one that seems not to receive the same emphasis as the principle of putting everything into words, perhaps because of our lingering commitment to the verbal bias intrinsic to the psychoanalytic tradition as such.  I am refering to the Modern Analytic principle of recognizing the analysand's need to experience particular feelings in the analysis, generally in relation to the analyst and, in addition, the analysand's need to at times encounter the analyst's (processed and sublimated) countertransference feelings both positive, as in what Dr. Spotnitz (1983) called the anaclitic countertransference, and negative, as in what he called the toxoid response.  For, as Winnicott (1949) understood, "If the patient seeks objective or justified hate he must be able to reach it, else he cannot feel he can reach objective love" (p. 199).

In this light, Modern Psychoanalysis may be as much about analysands experiencing needed feelings, of their own and those they have induced in the analyst, as it is about putting such feelings into words.  A good deal of the significant progressive emotional communication that takes place in certain analyses--perhaps in all analyses to a degree--may take a nonverbal or nondiscursive form.  It seems fruitful to conceive of mental health as integration of the heart and the head, the right brain and the left brain, feeling and insight, verbal and nonverbal communication.  In this light, it would seem likely that for some patients emotional growth will entail the deeper experiencing of their emotions and of emotional interchanges with the analyst, while for others it will entail differentiating and putting their emotions into words.

Perhaps, leaving developmental psychology to the developmental psychologists, we should not be so quick to assume that NONverbal equates to PREverbal and to subtly primitivize and pathologize what may often amount to quite subtle and intelligent nondiscursive communicative forms.  In other words, even if Dr. Brenner were right (although there are suggestions that he may not be) and before around 14 months of age infants are incapable of coding experience in ways that are recoverable in analysis, Dr. Meadow and other Modern Analysts may also be right about the nonverbal communicative processes, such as emotional induction, projective identification, and other forms of emotional communication that play an important role in our work with certain, perhaps even all, of our patients.  At the same time however, it may be an error to assume that such nonverbal processes equate to or dervive from primitive, preverbal processes and experiences.  They may or they may not.  Since we are analysts and not researchers in developmental psychology, perhaps we ought to remain within our area of competence, working in depth with the unconscious in the here and now of the analytic relationship, recognizing our speculations about the past as merely that, and leaving the issue of their validity to those who are interested in and capable of conducting scientific research in this field.


Aron, L. (1996).  Enactment, interaction, and projective identification: the interpersonalization of
    psychoanalysis.  In L. Aron.  A Meeting of Minds: Mutuality in Psychoanalysis.  Hillsdale,
    NJ: The Analytic Press, pp. 189-220.
Gill, M.M. (1982).  Analysis of Transference.  Vol. 1: Theory and Technique.  New  York:
    International Universities Press.
Heidegger, M. (1927).  Being and Time, trans. J. Macquarrie & E.S. Robinson.  New  York:
    Harper, 1962.
Jacobson, B. & M. Bygdeman (1998).  Obstetric care and proneness of offspring to  suicide as
    adults: a case-control study.  British Medical Journal 317 (14 November 1998): 1346-1349.
    This study is available online here:
     See also the subsequent discussion in eBMJ here:
Lacan, J. (1977).  Ecrits: A Selection, trans. A. Sheridan.  New York: Norton.
Langer, S.K. (1951).  Philosophy In a New Key.  Cambridge, Mass.: Harvard University  Press.
Lazar, S. (1998).  Optimal responsiveness and enactments.  In H.A. Bacal (Ed.).  Optimal
    Responsiveness: How Therapists Heal Their Patients.  Northvale, NJ:  Jason Aronson,
    pp. 213-233.
Langs, R. (1980).  Interactions: The Realm of Transference and Countertransference.
    New York: Jason Aronson.
Marshall, R. & S.V. Marshall (1988).  The Transference-Countertransference Matrix: The
    Emotional-Cognitive Dialogue in Psychotherapy, Psychoanalysis and Supervision.
    New York: Columbia University Press.
McLaughlin, J.T. (1991).  Clinical and theoretical aspects of enactment.  Journal of the American
    Psychoanalytic Assn., 39: 595-614.
Meadow, P.W. (1996).  Selected Theoretical and Clinical Papers. Modern Psychoanalysis 21, 2.
Rapaport, D. & M. Gill (1959).  The points of view and assumptions of metapsychology.   The
    Collected Papers of David Rapaport, ed. M. Gill.  New York: Basic Books.
Spotnitz, H. (1983).  Countertransference with the schizophrenic patient: value of the  positive anaclitic
    countertransference.  Modern Psychoanalysis 8: 169-172.
-----. (1985).  Modern Psychoanalysis of the Schizophrenic Patient: Theory of the Technique.
    Second Edition.  New York: Human Sciences Press.
Winnicott, D.W. (1949).  Hate in the countertransference. Through Pediatrics to Psycho-Analysis.
    New York: Basic Books, 1958, pp. 194-203.  Reprinted from  the International Journal of
    Psycho-Analysis 30: 69-75.

Copyright (c) 1999  D. Carveth

Donald L. Carveth, Ph.D.
Training & Supervising Analyst
Canadian Institute of Psychoanalysis
York University, Glendon College
2275 Bayview Avenue
Toronto, Ontario, Canada M4N3M6
Phone: 416-964-1499
Fax: 708-570-2816

Click here for Carveth Homepage