I suggested the slightly austere title ‘Levels of Proof’ simply because Bertrand Hanin asked me for a title. But allow me to give this presentation the form of a wider ranging talk rather than hold me to that restricted theme, though it will come up at a certain point. I speak first and foremost in honour of Daniel Widlöcher who has for years deepened the theme of ‘debate in psychoanalysis’, which is essential. Before listening to the introductory talks I was asking myself once again whether those of us who still speak about debate are not dinosaurs. Every day the mail floods us with programmes for congresses, meetings and conferences on the most tempting themes—classic as well as new—yet one can see that for the most part there is no real question of debate. Even when what is proposed are ‘roundtables’: they are not really roundtables; the talks are parallel and never come into contact with each other; and, finally, the time given for discussion is minimal. But it doesn't matter. Speakers do not try to convince each other; they come, do their thing and then leave.
Is this what's called postmodernism? I agree with the idea that the truth always remains out of sight, but on the other hand this doesn't mean that ‘anything goes’, as Feyerabend (1975) put it in a depressing phrase.
Shall I say a word or two about truth? I have no need of a capital ‘T’ in order to speak about it, but truth is nevertheless at the heart of debate. Let's say truth or simply reason, a certain discursive coherence. This requirement for coherence applies not only to the so-called hard sciences but also, if one wants to engage effectively in debate, to the human sciences. The ‘hard’ sciences are frequently invoked apropos of the ‘uncertainty principle’. But one would be quite wrong to use this as a weapon in the service of irrationalism or relativism; for the uncertainty principle is itself mathematised; it is itself something argued through and refutable precisely by virtue of being a principle. Even if it
asserts an uncertainty in the object (or in the observation of the object—the debate remains open) the assertion of the uncertainty principle is a rational assertion which itself is not presented as being uncertain.
Psychoanalysis can often get bogged down in an excessive transposition of relativism into our own domain. ‘Your transference as against mine’, says one; ‘your mother as against mine’, says another. This is the tendency that for my part I call ‘postmodern’ and which has become breathless in its horror of ‘the theoretically correct’.
Not that there is no debate these days; but, it must be said clearly, debate is dominated by the media. Of course we have our groups, our seminars, our private meetings, but ultimately the real debate takes place in the media. And here, two points return consistently: one concerns therapy as a ‘technology’, the other concerns the question of normativity. As to therapy as a technology, it seems to me that our predecessors bear the greatest responsibility, and did so from the moment when the official organizations—I am thinking particularly of the International Psychoanalytic Association—agreed to modify Freud's own position which defined psychoanalysis above all as method and as science before defining it as a therapy. From that moment the therapeutic aim has been made the primary goal. We are de jure an association of practitioners. And this moves us away from the scientific goal of truth, which was Freud's constant goal. With my experience as a translator, since I am immersed in Freud's original terminology, I simply point to the German term Erfolg, which can mean ‘success’ or ‘result’. Thus an Erfolg may be an adverse result. One could say, as one did in old French, cette aventure a eu un succès déplorable [‘this venture has had an unfortunate outcome/“success’”]. Which means—and Freud himself testifies to this—that any kind of ‘tendentiousness’, even the tendentiousness which claims that this or that is the correct procedure for therapy, is alien to Freud. At least, it is not his major concern. If psychoanalysis wishes to be a part of the scientific community, where I believe it certainly belongs (Laplanche 1995), it must accept that there are some propositions with no practical consequences. There are results of our work which are failures from the point of view of a certain norm of healing, of happiness, even of life. The theory of gravitation, to refer to Newton, is verified by a bridge which collapses just as much as by one which stays up. And yet none of us escapes this fascination with
the technology of therapy. This includes those who, in France, have succeeded in inscribing psychoanalysis in the corpus of psychotherapies and, in doing so, have only extended the general aim of the International Psychoanalytic Association; as well as those who claim to represent ‘absolute knowledge’ by the ways of the matheme. In this manner they have succeeded—I am referring to the current situation which everyone could confirm—they have succeeded by means of an outcry in having a Government Minister censor the text of a report on the evaluation of psychotherapies that was published on the internet.2 I haven't read this report, I cannot judge its worth; but it is astonishing that the problem of truth should be resolved by an outcry, by an assembly, so we're told, of a thousand psychoanalysts. Censorship, pure and simple, takes the place of argument and debate.
The other offshoot of this current debate—a media ‘debate’, to be sure, but a debate all the same—is the issue of normativity. I shall simply offer one thought about this: psychoanalysts are now invited to give their judgment on all the current social issues. ‘Which psychoanalysts?’ one might ask. They are ex-officio members on all the committees on ethics, violence, civil marriages, cloning, homophobia, etc. In each instance they intervene in the name of a norm, and it must be said that the norm to which they return is a Lacanian norm, which passes as the psychoanalytic point of view. The Symbolic, the Name of the Father, the Law. Psychoanalysis is not reduced to the Ten Commandments but to two or three. Psychoanalysis becomes a sort of Law of the social.
Freud says repeatedly: ‘bias is quite alien to me’. All ‘tendentiousness [Tendenz]’—this is his word—is alien to me (Freud 1914: 89).
After these excursions into therapy as a ‘technology’ and normativity, I return to what is the central concern for me today, a problem which is tangential to that of debate—the problem of scientificity, of verification or proof: ‘psychoanalysis in the scientific community’. Although I don't have time to go into detail, I allude here to what we might call the parallel trajectories of Freud and Popper. These two men of genius, even if one doesn't measure the genius of one against the other, are at cross-purposes to an incredible extent. There is a lack of awareness on the side of Freud, who I don't think ever cited Popper; and on Popper's side there is misunderstanding. One of Popper's most important works is Conjectures and Refutations (1963). A conjecture is a
model that is devised to account for observed facts, which are themselves already partially constructed on the basis of other, pre-existing models. As such, for Popper science is never drawn from facts, is never inductive. Here, Popper takes up Hume's old argument about habit. I leave my office, I open the door. I go into the adjacent room. I can open the door a hundred times, a thousand times, ten to the power ten or however many times, and I will always come out on a floor. What would happen if one day the door opened onto the void? Nothing permits the assertion that the door will not, one day, open onto the void. This is the proposition that Hume and Popper defend with vigour. Induction cannot lead to a certainty. One can speak of statistics, but no law is ‘statistically true’. One cannot say, ‘This is 95% true.’ One must say: it is 100% true that this phenomenon will happen 95% of the time. Which is very different. There is no such thing as statistical truth and the attempt to rebuild induction on the basis of statistics is a logical fallacy.
However, Popper does not embrace the notion of an absolute truth in the Platonic or any other sense. On the contrary, for him the only truth is a provisional, even imaginary truth. He starts from the idea of ‘conjecture’—subsequently taken up under the names of ‘model’ or ‘paradigm’—as something which remains provisional until it is refuted, falsified. Is this so far away from Freud? There are certain fragments of epistemology in Freud. There is the familiar passage in ‘Instincts and their Vicissitudes’ (1915b)(in which Freud focuses more perhaps on concepts than on the relations between them, and in which the concepts are considered provisional. When they are no longer adequate, he says, we change them. In contrast to Popper, he formulates things in terms of inadequacy rather than in terms of refutation. The idea that one day the theory will be contradicted by what it entails: this aspect is sometimes present in Freud. It is interesting to detect the rudiments of Popperian thought in Freud. Think, for example, of the title of a text which is single-handedly Popperian in its implicit programme, ‘A Case of Paranoia Running Counter to the Psycho-Analytic Theory of the Disease’ (1915a) (Whatever the fate and the truth of this text, its purpose is to test a falsification of a theory, in Popper's sense, within a case of paranoia—a case of female paranoia in which the underlying homosexuality had not been apparent.
Here, though, the following question might be posed: if the case
really did contradict the theory, would everything collapse? Would the whole of metapsychology come crashing down? Here is where the question of the levels of statements and the coherence of theories arises. It is true that in Popper as well as Freud one sometimes encounters an absolutism which may lead one to believe that ‘the throne and the altar’ ultimately depend on a certain number of shibboleths, such that the refutation of one small point of the metapsychology might imperil the entire system. This is clear in Popper because he takes the Einsteinian revolution as his model. Yet this revolution is actually a very unusual case, where ultimately a single experiment, that of Michelson and Morley, was the origin of the falsification of Newtonian theory. But is it the case that things always happen this way? I think not; and along with many epistemologists who have followed up Popper's thought, developing and nuancing it, I accept that there are intermediate levels between what one might call ‘psychoanalytic doctrine’, to use Freud's term, or general metapsychology, and something I shall shortly call the ‘mytho-symbolic’. Between these two levels there are the intermediate levels of theories of conflict, of psychopathology, of symptoms, of Witz, etc., even the theory of dreams. Would a modification of the theory of dreams necessarily entail a falsification of the whole of metapsychology? I don't think so.
You see that the idea of a theory which is ultimately supple and open to reworking, a theory containing on the one hand a relatively dense kernel and on the other hand statements which are not deduced from the general theory, is valid not only for psychoanalysis but for every science. A hard kernel, a more supple periphery. What would be the hard kernel of psychoanalysis for which we could claim to devise (is this possible?) Popperian procedures? Let there be no mistake, what causes scandal today, the hard kernel, what's constantly being left to one side, consists of infantile sexuality, the unconscious and the paths of access to it, and repression. These may well be objects, but above all they are constructed models and their centre is the infantile sexuality which Freud first formulated in the Three Essays (1905). I cite a short passage from the Three Essays, which involves some simple epistemology:
The direct observation of children has the disadvantage of working upon data which are easily misunderstandable; psycho-analysis is made difficult by the fact that it can only reach its data, as well as its conclusions, after long detours. But by co-operation the two methods can attain a satisfactory degree of certainty in their findings. (Freud 1905: 201)
You can see the image of a tunnel being dug from both ends and by way of rigorous understanding. I emphasize this because it is not simply a question of ‘observation’; it is observation itself which must be understood—on the one hand the observation of infants, and on the other hand psychoanalytic experience.
All of which seems simple. And yet what a concatenation of discord is thus created. Prominent among the voices are the categorical positions for or against the psychoanalytic observation of children, that is, for or against one end of the tunnel. In his 1979 article ‘L'enfant modèle’ (I don't think he has changed his opinion on the subject), André Green immediately contrasts two types of science: ‘sciences of observation’ and ‘sciences of interpretation’. Things start out from an epistemological dilemma of the most radical kind. The Popperian epistemological sequence, which would be observation, conjecture, refutation, is displaced into an opposition between ‘observation’ and ‘interpretation’. An opposition which itself corresponds to two forms and domains of knowledge: on the one hand, psychological knowledge of the child, observations which would be devoted to what is purely ‘perceptible’; and on the other hand, psychoanalytic knowledge of the adult, which would be the domain of representation and interpretation.
I am revisiting this question because it is important to reject this opposition between the science of observation and the science of interpretation, which is posed as a general opposition pertinent to all knowledge, and not specifically for psychoanalysis. First of all, apropos of observation, there is no such thing as a science of observation, not even observation of the stars. Observation functions only if guided by conjecture. The idea of direct observation is a poor one in respect of nurslings just as much as galaxies. There is no observation which is not oriented by hypotheses: there is no domain of pure observation. Even the observation of the stars by the ancients presupposed reference points, charts, suppositions, conjectures.
Thus Green puts in the same bag psychology, child psychology and an empiricism that operates without reference to any principle. And yet modern psychology, such as that of Stern or Dornes, can only function when supported by conjectures which often are quite elaborate. What he places in opposition to this are the sciences of interpretation. But here the ambiguity becomes total, since ‘interpretation’ is a word with which psychoanalysis is well acquainted. Is it really psychoanalytic interpretation which is at stake here? Psychoanalysts interpret in their practice; but they have never wanted to make that interpretation—psychoanalytic interpretation—the path of science. For example, one cannot say that the Freud's metapsychology—the first or the second topography for example—is the result of an ‘interpretation’ in the psychoanalytic sense of the term. It is a conjecture in Popper's sense. Yet, by a slippage of meaning, Green gets very near to maintaining that psychoanalysts who treat adults would constitute a model for science because they think by means of interpretation: conjecture, construction, scientific hypothesis.
This debate is important. It has been somewhat distorted by Freud and by all those who, a little imprudently at certain moments, have spoken of the direct observation of the child as a scientific finding.
In my opinion, the central problems of psychoanalysis remain the unconscious and the ways we gain access to it (for Freud this is the very definition of psychoanalysis), and the position and genesis of infantile sexuality or, if one prefers, of the sexual drive. I speak of infantile sexuality in the sense of the Three Essays, that is to say in the sense of what I call the ‘sexual’ [le sexual]3 beyond the ‘sexed’, beyond the difference between the sexes and beyond even the diversity of genders.
It is interesting to read the Three Essays of 1905. Once separated from the later additions, the original essays emerge as a composite being, an enigmatic sphinx, which is often ambiguous and sometimes contradictory. Nevertheless, these essays are consistent in their affirmation of the notion of an enlarged sexuality—the major problem which is also the most denied and the most repressed, even though we sometimes dedicate study days to it. Freud begins his ‘Second Essay’ with two subchapters, entitling the first ‘Neglect of the Infantile Factor’ and the second ‘Infantile Amnesia’. This immediately underlines the repression which affects not only the child with respect to his own sexuality,
but also the adult parent and the adult observer. If we take account of this neglect of the infantile factor and take account of infantile amnesia, what we discover is not the root of attachment theory, but the root of its hypertrophy and its tendency to occupy the entire domain of the infantile. My concern is not that attachment theory should have no place in this domain, but that it takes up all the space. Infantile sexuality is, for its part, watered down and barely mentioned. If paediatric and perinatal specialists overrun the whole field—albeit with certain concepts that have a psychoanalytic flavour, such as ‘drive’ or ‘attachment drive’—certain psychoanalysts would, for their part, be ready to abandon the whole field to these theoreticians of attachment since, according to one conception, infantile sexuality is only a retrospective projection into the past by the adult (a re-signification, as it's sometimes called).
I do not have the time to dwell upon this distorted conception of afterwardsness [après coup]4 as a matter pure back-projection. The notion of afterwardsness is completely transformed when one assumes a particular point of view that seems to have been neglected by everyone, including Freud (Laplanche, 2006). In Freud himself and among all those who have studied the child and the adult-child relationship, child and adult are always envisaged from the beginning and primarily in terms of a succession: the child is the antecedent of the adult, the prehistory of the adult; the adult regressively finds the child and the traces of the child within himself. The point I have tried to hammer home is that the child and the adult—before the first becomes the antecedent of the second, and before the second finds within himself the traces of the first—must be conceived by psychoanalysis as being from the beginning simultaneous, in dialogue, engaged in an exchange of messages. The adult in the presence of the child is an adult who sees re-burgeoning within himself all the pre-genital, partial sexuality which is aimed at partial excitations and pleasures. Here is how Freud in 1908 summarizes infantile sexuality as it emerges in the Three Essays: ‘in man the sexual instinct does not originally serve the purposes of reproduction at all, but has as its aim the gaining of particular kinds of pleasure’ (Freud 1908a: 188).
I shall not unpack here my own ‘conjecture’—the theory of messages which are compromised by the sexuality of the adult and then translated by the child. It is a theory of the genesis of the unconscious
and of the genesis of the apparatus of the soul, which for me is at the heart of metapsychology.
What truth is there in what here must be named the ‘general theory of seduction’, what falsification is possible? Surely we are a long way from having found the means of falsifying Freud's first or even second topography, and no less so the theory of seduction as I propose it. But this is so precisely to the extent that one is in no position to consider the adult and the child as a simultaneous and asymmetrical couple. Not an interaction but an asymmetry of communication. That is, to consider in particular the analysis of the parental adult (or caregiver) as being no less pertinent to the situation than conjectures about the genesis of the psyche of the child. This taking into consideration of the unconscious of the adult in the adult/child situation is what is lacking and what will always be lacking in the infinite developments of ‘the experience of satisfaction’ (nursing) which is habitually reduced to a single person: namely, the child.
How is it possible to set forth and perhaps refute or falsify a process which bears on messages (primarily nonverbal messages), their reception, their translation, and even the failure of this translation?
In a sense, observations which are at once individual and collective, such as those of Roiphe and Galenson in The Infantile Origins of Sexual Identity (1986) show us one possible path. In any case they demonstrate that, in a given adult/child situation, it is a matter of putting conjectures to the test, and not of lining up meaningless ‘observations’. Furthermore, where they fail somewhat is where it comes to conjecturing about the impact and the evolution of parental messages. Where they fail is precisely in the observation—by which I mean the psychoanalytic observation—of adults, the caregivers or the parents of these children.
I shall now return to the question of levels before coming back to the possibility of some kind of observation and conjecture about the processes leading to the formation of the ‘apparatus of the soul’.
Popper showed us the way forward by opposing, somewhat schematically, the scientific and the metaphysical. The scientific is that which is susceptible to being falsified and indicates the paths of its possible falsification. The metaphysical is that for which no falsification is possible. Don't suppose that Popper devalues the metaphysical, however. He considers Darwinism, for example, to be a fruitful metaphysical system,
some of whose consequences can themselves be falsified, but which as a system cannot. Since Popper's own writings, Popperianism has been significantly modified and improved. The all-or-nothing of great changes of system, such as Newton being eliminated by Einstein, is now considered to be exceptional. In fact, as I indicated above, in each theory there is most frequently a hard kernel—the most difficult to endanger—as well as an entire series of hypotheses subject to testing and falsification but whose refutation does not directly endanger the centre.
In psychoanalysis, this can be found in what I call levels of truth or of proof. I spoke of the intermediate level. I have already cited the example of the ‘case of paranoia’ but I could give another example of the same level. Interestingly Freud employs the term ‘negative case’ here. The discovery of a single ‘negative case’ would allow a hypothesis to be brought down. It comes up in respect of neurasthenia. In his early texts, Freud says, ‘neurasthenia is always caused by a genital dysfunction’ (he is thinking essentially of masturbation), and ‘so far as the theory of the sexual aetiology of neurasthenia is concerned, there are no negative cases’ (Freud 1898: 269). Here we are at what I would call an intermediate level: a psychoanalytic theory which does not, of course, call into question the hypothesis of the unconscious. In this we encounter something quite close to Popper, and above all to his successors.
We have thus discerned the existence of one or several intermediate levels, but the major opposition is the one between the two extreme levels, the opposition founded by the theory itself. There is on the one hand the Sexual Theory or the Metapsychology, the theory of the unconscious and its genesis, centred on infantile sexuality; and on the other hand there are the ‘sexual theories’ of children or adults. Yet Freud left a certain ambiguity—which some have rejoiced in—between the sexual theory of the Three Essays, which is the theory of the scientist, and the sexual theories of every individual, the theories which the individual creates for himself about his own existence and his own sexuality. We thus have psychoanalytic theory at one extreme, and at the other extreme the way in which the human being ‘theorises himself’, on the basis of messages from the other which he has to translate and which he must integrate by assimilating. The most conspicuous example in the
article on ‘The Sexual Theories of Children’ (1908b) is the castration complex, which is a way for the child to represent for itself in a ‘story’ or a narration the difference between the genders as communicated by the parental messages that assign gender to the child (cf. Laplanche 2003).
The castration complex and the Oedipus complex are mythic theories which do not require proof. In Popper's terms, they belong to the level of metaphysics; what I call the mytho-symbolic is presented to the human being largely by his cultural entourage. Unfortunately, however, Freud ended up regarding the sexual theories of children—the apparatus best suited to repressing the unconscious—as the very kernel of the unconscious.
The strength of the theory of seduction consists in its ability to account for the non-scientific function of psychoanalytic myths. It is thus necessary to situate the two Popperian levels carefully. Metapsychology is scientific in Popper's sense of the term. The mytho-symbolic is narrative and its principal function is both sublimatory and repressive, helping to give form to a personal history, which is clearly of crucial importance for the human being.
There is no denying the difficulties for observation which would arise here—for the observation of what I call the fundamental anthropological situation. Conjoint rather than successive observation, psychoanalytic observation, psychoanalytic investigation of the adult-child dialogue in its entirety, and of the resurgence of infantile sexuality in the adult protagonist. It is important to underline that the adult in the situation of relating to an infant experiences a resurgence of his own infantile sexuality. Freud underlines this several times and everybody realises it.
We should, then, pay close attention to the two warnings Freud puts at the beginning of his ‘Second Essay’ on sexual theory: ‘Neglect of the Infantile Factor’ and ‘Infantile Amnesia’. These are one and the same: the repression of the sexual.
Finding a joint methodology for the observation and interpretation of adult-child communication and its results would reopen the question of the ‘neglect of the infantile factor’: is it the result of repression or an absence? I believe that a large part of so-called psychoanalytic child psychology—which has existed for decades and has taken on a new élan
with ‘attachment theory’—however scientific the psychology may be, must begin by questioning (and trying to eliminate) the repression of infantile sexuality in the adult and of course, above all, in the ‘observer’.
Delivered at Débat en Psychanalyse, the sixth colloquium organized by the APEP (Association psychanalyse et psychothérapies), the CHU Pitié-Salpêtriere and the Association de santée mentale, Thursday, March 10, 2005. Discussion opened by Professor Daniel Widlöcher. Published in Sexual: la sexualité élargie au sens freudien 2000–2006 (Paris: Presses Universitaires de France, ).
[Trans.: Here Laplanche invokes his own neologism, sexual, which stands in pointed contrast with the standard French term sexuel. The neologism operates by appropriating the German prefix Sexual- into a free-standing word. (In German Sexual only appears as a component or proclitic in combination with a noun, e.g. Sexualtrieb– sexual drive; Sexualtheorie– sexual theory). This is an attempt to register terminologically the difference between the enlarged Freudian notion of sexuality (le sexual) and the common sense or traditional notion of a genital sexuality (le sexuel). This terminological innovation cannot be adequately captured in English, since the German term Sexual coincides exactly with the spelling of the standard English term ‘sexual’, rather than contrasting with it as in French. To mark Laplanche's use of the neologism we have therefore italicised the standard English word. Laplanche's most sustained discussion of his neologism to date appears in Laplanche (2003)].
[Trans.: ‘Afterwardsness’ is the term by which Laplanche prefers to translate Freud's Nachträglichkeit into English. James Strachey's Standard Edition translates Freud's term as ‘deferred action’. For a
critique of Strachey's choice and a discussion of ‘afterwardsness’, see Laplanche 1992].
[Trans.: The report to which Laplanche is referring was published in February 2004 by the Institut nationale de la santé et de la recherche médicale (INSERM), and was deeply critical of psychoanalysis. The report was initially published on a Government website but subsequently withdrawn from it, allegedly under pressure from certain figures in the French psychoanalytic community].