Dreher A.U. (2004). What is psychoanalytic conceptual research?

Anna Ursula Dreher è membro della società psicoanalitica tedesca (Deutsche Psychoanalytische Vereinigung DPV) e lavora come psicoanalista e supervisore a Francoforte.E’ autrice di varie pubblicazioni sugli aspetti concettuali e epistemologici della ricerca in psicoanalisi, fra cui:

– Sandler J., Holder A., Dare C., Dreher A.U. (1997). Freud’s Models of the Mind: an Introduction. London, Karnac Books. Trad.it. I modelli della mente di Freud. Milano, F.Angeli 2001.

– Sandler J. & Dreher A.U. (1995). What Do Psychoanalysts Want? The Problem of Aims in  Psychoanalytic Therapy. London, Routledge.

– Dreher A.U. (2000). Foundations for Conceptual Research. London, Karnac Books.

– Dreher A.U. (2003). What does conceptual research have to offer? Pp. 109-124 in Leuzinger.Bohleber M., Dreher A.U., Canestri J. (2003). Pluralism and Unity? Methods of Research in Psychanalysis. The International Psychoanalysis Library. Trad.it. Pluralismo e unità? Metodi di ricerca in psicoanalisi. Perugia, Borla Ed. 2007.


– Dreher, A. U. (2010). Pluralism in theory and in research – and what now? A plea for connectionism. In: M. Leuzinger-Bohleber, J. Canestri, and M. Target (Eds.), Early Development and Its disturbances: Clinical, conceptual and empirical research on ADHD and other psychopathologies and its epistemological reflections (239-266). London, Karnac Books.

– Dreher, A.U. (2014) Psychoanalytic research with or without the psyche? Some remarks on the intricacies of clinical research. In: Boag, S. & Brakel, L. (eds.) Philosophy, Science, and Psychoanalysis. London, Karnac Books.



Anna Ursula Dreher

What is psychoanalytic conceptual research?


Testo originale della relazione presentata al Centro Psicoanalitico di Firenze nell’Aprile del 2004, che pubblichiamo per gentile concessione dell’Autrice. Nel sito web del C.P.F. è presente la traduzione del testo.

vai alla versione italiana del testo.


Introductory remarks

Conceptual research is no doubt en vogue. But if I look at all that goes by that label meanwhile, I think that there are some fundamental misunderstandings. And today for this reason I would like to consider the concept ‘conceptual research’, and say something about its origin and development – and also talk about how this research approach might cooperate with other research activities in psychoanalysis.

For this it is indispensable to first of all have a closer look at the research subject, namely our analytic concepts, and discuss why they are not so easy to grasp – and why, according to my opinion, we actually need concept-clarifying research.

In psychoanalysis, it is mostly the clinical phenomena which we seek to grasp theoretically. We speak with our patients and reflect about them. However, we do not merely speak with, but also about our patients – and our interactions with them – and we publish case studies. Language thus has a double function: as ordinary language in the talking cure, but also as scientific language when we discuss case studies or introduce our ideas at conferences, or when we present research findings. In all that, we normally use concepts like transference and countertransference, trauma and unconscious processes, defense and resistance and many more. And sometimes today we would already be content, if these concepts we use to communicate about the phenomena of interest to us, would not provoke all too many misunderstandings.

The complaints about the semantic vagueness of our concepts, sometimes about their inconsistent use and occasionally about an ‘anything goes’-use, are legend. Yet, arbitrary use overextends concepts and thins out their precision. Do these concepts have the same meaning for all of us; and do we mean the same thing when we say the same thing? The question arises, therefore, whether there can be a systematic way to grasp, structure, discuss and evaluate the differing and changing uses of our concepts.


Working definition

Let me put the following working definition at the beginning: Psychoanalytic conceptual research is such a way to clarify concepts. As a working definition I would suggest: conceptual research is concerned with the systematic and methodic investigation of the explicit and implicit meanings of analytic concepts and conceptual fields – in their clinical and extra-clinical use. For example, we are used to seeing clinical phenomena – liketransference patterns – as the subject of scientific investigation. But research is not limited to investigate such phenomena, it can equally encompass our analytic language which we use to grasp such phenomena, it can investigate the changing use of a concept – for example, oftransference. Conceptual research is so-to-speak only indirectly focused upon the phenomena themselves, as it focuses on our theoretical language, our concepts; and has the aim to clarify the rules of a reasonable concept use meeting clinical and scientific requirements.

For this purpose it is useful, to reconstruct the important points of change in the history of such conceptual fields under study. But it is equally indispensable to investigate the current use. Here one has to keep in mind the pressure for change, brought about by empirical – including clinical findings – and by theoretical developments, inside as well as outside psychoanalysis. The aim of such conceptual studies is not to write a history, but to describe and evaluate a concept in all its relevant meaning aspects. This could not only help us to learn from old mistakes and thus avoid their repetition, but also to preserve the reasonable elements there are in a concept.

Even though it may not be possible to standardize our concept use, it is nonetheless important to find a clarifying communication about commonalities and about differences, a precondition for a constructive dialogue within our different analytic cultures, as well as with other sciences. One point in advance: There is no such thing as conceptual research as a unified and standardized procedure, just as empirical research as such or the experiment as such do not exist. Conceptual research is rather to be understood as a research program: if there is research interest, projects for the clarification of a concept may be initiated. They would be individually structured and the application of the appropriate methods would result from a methodological discussion. Conceptual research thus defines itself not so much by its methods, but by its subject, the use of concepts, and its aim, the clarification of concepts (for a detailed discussion see Dreher 2000; 2002).

In some way Freud outlined the frame for such a systematic conceptual reflection, too, when he described psychoanalysis as a science built upon the interpretationof the empirical [Deutung der Empirie]. He said that such a science will

“not envy speculation its privilege of having a smooth, logically unassailable foundation, but will gladly content itself with nebulous, scarcely imaginable basic concepts, which it hopes to apprehend more clearly in the course of its development, or which it is even prepared to replace by others. For these ideas are not the foundations of science, upon which everything rests: that foundation is observation alone. They are not the bottom, but the top of the whole structure, and they can be replaced and discarded without damaging it” (Freud 1914, p. 77). Our analytic concepts do change; they live just like scientific language games in general live.


The integration of the ‘new’ and the ‘meaning space’ of a concept

Every living science has the problem of continually integrating the ‘new’ that arises, whereby that ‘new’ is always confronted with established concepts. The ‘new’ is encountered in psychoanalysis every day in various forms – in clinical work, in classifying diagnostics, or also in research:

  • The handling of new material, which a patient brings to treatment, is a determining moment in everyday clinical activity and could lead to a question like: how does one integrate the new perceptions and experiences into the previous ideas about mode and pattern of transference in this actual case?  
  • A second example: what happens to our established diagnostic categories, when, as has variously occurred in the past, specific new forms of auto-aggressive behavior, self-mutilation, or eating disorders come into the focus of interest. Do we use our familiar concepts like hysteria, or do we develop new diagnostic concepts for these symptoms, and if we do, what is the relationship of the new to the old concepts?
  • A third point should be mentioned: ‘new’ knowledge presented by our research. This, too, may have consequences for the meaning of concepts: We have been seeing this for some time now in infant research, where some of our developmental concepts – especially the phase theory, based upon drive development – are being critically questioned by attachment research. We also see this where new findings from imaging procedures might change our analytic ideas about how memory functions.

Sometimes it is possible to integrate this ‘new’ into the familiar system of concepts, sometimes it may make sense to modify the established use, sometimes it may be necessary to give up old concepts and introduce new ones. Such change is more clearly seen, the greater the historical distance is. Thus today for example – in the light of recent research findings from developmental psychology regarding the different competencies of the infant – we would no longer speak of a ‘phase of infantile autism’, a concept that Mahler still suggested in the 70ies (Mahler, Pine, Bergman 1975).

Comparable mechanisms like those that are known from the history of science, also apply in relation to psychoanalysis. Joseph Sandler outlined structurally similar thoughts on the change of our concepts, similar to those, which Thomas Kuhn had about the dynamic change of theories as a whole. Sandler (1983) presumed that changes do not show in the same way at all places of our theoretical edifice, but that there are always concepts which are subject to particular tensions and criticism, and which therefore are of particular interest for conceptual research. Based on his experiences with the London Hampstead-Index-Project (Bolland and Sandler 1965; Sandler 1987), Sandler proposed to pay special attention to two phenomena:

  • the elasticity of the meaning-space of a concept and
  • the implicit conceptualizations by the analyst at work.

An analytic concept can be presented best in an elastic, multi-dimensional meaning space, where the weighting of aspects of meaning may change over time. ‘Elasticity’ means that concepts can ‘react’ to changes, take up, endure and integrate them. Such concepts take up the strain of theoretical change, absorb it, while more organized newer theories or part-theories can develop. This is of course different from the conviction, that our concepts have an unchanging meaning, which can be presented in a formalized form. The way analysts ‘play’ with concepts in an elastic meaning-space, in a context-dependant manner – better: case-dependent manner – is also a central aspect of the clinicians’ creative work.

But, elasticity does not mean elasticity at any price. To achieve an adequate degree of elasticity means, walking a fine line between a fixed or even dogmatic definition and a quasi-private concept use. No concept should become as arbitrary in its meaning, as has happened lately, for example, to the analytic trauma concept. Blum describes the situation sharply, when he sees the trauma concept situated between two unsatisfactory poles:

“The concept of psychic trauma has been both unduly compressed and stressed beyond the confines of consensual definition. Trauma may be so narrowly defined that the ego is considered to be totally overwhelmed, with no possibility of adequately registering the trauma or responding to it. At the other extreme, trauma may be loosely identified with any noxious experience or developmental interference” (2003, p. 416).

With the ‘implicit conceptualizations’, Sandler attempted to grasp a particular process: the usefulness of a concept like e.g. transference must show its validity in clinical practice, where each analyst is continually subsuming relevant phenomena under concepts. Their explicit, codified meaning often turns out to be unsatisfactory for one’s work with the actual case. As an experienced clinician one may therefore be tempted, to adapt a concept to one’s needs, gradually developing a kind of subjective meaning-space from case to case. This subjective meaning-space overlaps with the established one – the one socially shared with the reference group, but, nevertheless possibly showing interesting divergences.


These processes are implicit in so far as they are, often for a long time, not precisely put into words. But, it is necessary and not unusual for scientists to reflect upon these adaptation processes of concepts and to systematically attempt to make the boundary between ‘explicit’ and ‘implicit’ more permeable – and thus integrate this specific ‘new’ into the general theoretical reflection.



An example: about the meaning space of ‘transference’

Let me briefly illustrate, what is to be understood by this meaning-space, with the concept of transference, surely one of our most important clinical concepts with a long and complex history (see e.g. Sandler, Dare, Holder 1992). The concept lends itself well to demonstrate a problem that presents itself similarly with many of our concepts, namely, the reference problem: what are we referring to in the world, when we speak of transference?

According to Freud: to a burnt-in cliché, that is, to an unconscious, ontogenetically acquired relationship pattern, which is being actualized in the here and now of the analytic encounter. For psychoanalysis, transference is primarily not an observable behavior and not simply a manifest narrative. Rather, transference is a concept, under which an analyst subsumes a number of clinical perceptions and experiences. There is, therefore, no behavior that as such is transference. There are only very different phenomena, which in the clinical situation can be seen as indicators of transference. And these are essentially unconscious patterns.

If one were to subject the concept of transference to a systematic meaning analysis, the following dimensions or aspects of its meaning space – based on clinical phenomena – would result (see also Dreher 1999):

  • the aspect of repetition, addressed by Freud in his idea of new editions of indelible impressions of a stereotype template originating early in a patient’s life history;
  • the change over time, concerned with the question of if and how transference patterns change over a lifetime, how new editions are elaborated through ‘deferred action’ (Nachträglichkeit);
  • the form of appearance, which deals with the problems of whether transference phenomena are expressed in narrative or enacted form, and by what pattern of affects they are accompanied;
  • the meaning of unconscious elements in narratives as well as in enactments;
  • the relevance of interactive processes, which looks at the interplay of transference, countertransference, and the analytic situation;
  • the psychic functionality, which asks how the ‘here and now’ is connected with the ‘there and then’.

This list of aspects is certainly not exhaustive, although it names the most important aspects of the meaning space of ‘transference’. Only the consideration of all aspects, i.e. a holistic view of the concept, makes it useful for our work in the clinical situation. And it is this very complexity and richness of the concept, which have made it so powerful and effective in our history – a complexity, which is sometimes lost, when the concept is applied in research contexts. So, for example, not infrequently a pars-pro-toto mistake shows up, when analytic concepts are operationalised in empiristic research technologies: one manages to grasp one, certainly important aspect of many; however, one fails to do justice to the complexity of the concept. There is often the danger of “oversimplification” (Green in Sandler et al. 2000), the danger of an empiristic flattening of analytic concepts in the research process.


Specific features of psychoanalytic concepts

What all makes our concepts so complex and sometimes ambiguous?

To start with, there is the interwoven-ness with the place, where many concepts not only have their historical origin, but which, above all, is the central place, where they must prove their worth: the analytic situation. There, analysts gain a special kind of experience with other people and with themselves, and some even consider this experience unique. Two subjects, patient and analyst, meet each other, and it does not satisfy a psychoanalytic consideration to merely objectively register the behavior of the two actors. Psychoanalysis, of course, is above all concerned with the understanding of two dynamically intertwined inner worlds. A particular role therein is played by the tension between psychic reality and external reality – two realities that are not just simply in a representational relationship, which could be constructed by objective methods.

Analysts attempt to put the totality of these experiences into scientific language, to make them communicable, replicable, and criticizable via the concepts used. The meaning of these clinical concepts refers particularly to the subjectivity of the analytic experience including unconscious aspects – a subjectivity, again not easily accessible from an objectivizing perspective.

Another specific feature of our concepts is that their complexity is increased by the fact, that their meanings are embedded into the theoretical presuppositions of the analytic Weltbild. This is irritating sometimes, when they are used in other than clinical contexts. Our ideas, by the way, about conscious and dynamically unconscious processes, our thoughts about how object relations are sedimented, or the analytic convictions about the effects of drives and affects, about fantasies and dreams, are certainly not shared by everyone; there are indeed still fundamental differences between the analytic model of the mind and, for example, empiristic, cognitivistic, and neuroscientific models.

Beyond all that, the person of the analyst is relevant for the use of concepts from a rather different perspective: concepts are in a special way tied to the analytic identity of analysts. Analytic concepts are our everyday, usually unquestioned toolbox, the inventory of terms, which is taken for granted to describe reality. And it requires some effort to take a distanced or even critical attitude toward one’s own concept use. Perhaps it is this very matter-of-course-ness, which is a source of resistance against other conceptual understandings, particularly, as the use of the language of their analytic group, often is identity-creating for analysts. And not infrequently the rules of the narcissism of the small difference apply here: the closer the conceptual understanding of two discussants, the more pronounced their criticism and delineation may become. Such mechanisms may be obstacles, of course, for any communication about concepts and about proposals for their change.


Further sources of variety in concept use

All of this only explains to a certain extent the great variety in the use of our concepts; let me shortly add some further factors:

  • one factor would be the long history of analytic traditions, mostly linked to prominent names like Freud, Jung, Klein, Bion, Winnicott, Kohut, Lacan, etc.;
  • another one would be the history of the worldwide spread of psychoanalysis with its differentiated regional cultures. From a conceptual research viewpoint it would be a useful endeavor to work out an up-to-date taxonomy of the regional analytic dialects;
  • then there would be the different convictions as to which status psychoanalysis has in the canon of sciences. The range of opinions goes from a strictly empirically oriented, natural science-based understanding to a hermeneutic one, with an orientation at the single clinical case, up to opinions, attesting psychoanalysis a status of independence and uniqueness – each with a specific view about function and meaning of concepts;
  • furthermore there are the consequences due to embedding psychoanalysis into the different medical, and thus economic, systems of various countries. One should name here the often criticized so-called medicocentrism (Parin 1986), namely the subsuming of analytic thinking and concepts under a strictly medical view of nosology, diagnostics and therapy. One may mention the possibly varying answers to such important questions like: what is normal; or from which degree on, mental disturbances can be understood as an illness; and what are the intended aims of an analytic cure (Sandler and Dreher 1996; Dreher 2002).

Whatever other factors may have generated the variety in our concept use, this variety is judged differently: some complain about the Babel of languages, others do not. Some recognize a creative evolutionary potential here, others see a danger for the unity of psychoanalysis. And there are good reasons, by all means, for both points of view. Yet, what remains the indispensable basis and motor for good and precise concepts has been known since Freud: ‘observation alone’ – that is, the observations in the clinical situation and the constant attempts to conceptualize them adequately. But between these two spheres – the phenomena and the language used to describe them – lies, what Green calls, a “practical-theoretical gap” (Green 2003, p. 29). The question is, however: does this gap have to remain so broad, a mysterious gap for ever? And, what would a systematic methodic approach look like that could be useful in reducing the long-standing complaints about conceptual confusion?


History and practice of conceptual research in psychoanalysis – some examples

If one was to list all that might be helpful for the clarification of a concept, one might find the following thematic areas, each with corresponding questions:

  • The investigation of the historical context of a concept’s origin – which problems was the concept intended to solve, when it came into being?
  • The history of a concept’s use, viewed against changes of psychoanalytic theory – which were the main changes in the understanding of the concept?
  • The current use of a concept in clinical and extra-clinical practice – which are the most important variants of current usage and to which extent are they compatible?
  • A critical discussion and, possibly, formulation of a suggestion for a different concept use – which well-founded suggestions could be made for a better use?

In order to integrate these aspects and to do justice to each of them, we must of course apply different research methods. A systematic literature analysis, for instance, would help the reconstruction of the development of a concept, mapping out the crucial points of its change in analytic history since Freud (as demonstrated e.g. by Sandler et al. 1992, regarding concepts relevant to the therapeutic process; see also Sandler et al. 1997). In addition, such studies can contribute to a preliminary clarification of the concept, or help identify and work out any residual ambiguities.

Furthermore, systematic conceptual research comprises procedures, which make it possible to generate evidence for the explicit as well as implicit use of a concept in clinical practice, e.g. by way of expert interviews. Here, one particular aspect turns this kind of research into a psychoanalytic one: a simple inventory of the present use could be put together by applying behaviorist or philological methods, but it is not sufficient, when we want to include aspects of a concept’s current use in clinical work, especially those which have not yet been made explicit – the above mentioned implicit conceptualizations.

As a historical prototype of such conceptual research one can see the work of the already mentioned Hampstead-Index. There, a research team of analysts – on the basis of small units of observation from analytic treatments – attempted a categorization of these observations under analytic concepts. All this took place in close discussion with the analysts treating the respective cases. At times, these categorizations turned out to be not so simple; causing quite controversial discussions about how the treating analysts themselves understood the material of their cases and which concepts would best grasp the respective units. And this in turn provoked discoursive clarification attempts regarding the concepts themselves.

With similar interests a group of researchers at the Sigmund-Freud-Institute in Frankfurt/Germany instigated a project in the 1980s, which investigated the concept of psychic trauma (Sandler et al. 1987 and 1991). At the time, the concept had been chosen, because it played a central role in the formulation of the PTSD-syndrome (Post Traumatic Stress Disorder), which the Vietnam veterans’ associations had fought for. It was also central in the debates of the false-memory phenomenon in the context of sexual abuse; and essentially in the attempts, to understand experiences from the treatment of Holocaust victims, linked to the questions of if and how it is possible that traumatizations could be transferred from one generation to another.

In all these considerations and discussions the use of the concept had undergone a significant extension.Along with questions, concerning contents, the main purpose of the Frankfurt pilot study was, however, to develop an inventory of methods for conceptual research. This inventory included the already mentioned literature analysis, and semi-structured interviews by experienced analytic clinicians, members of the research project (in the role of conceptual researchers) of experienced analytic clinicians (in the role of ‘experimental subjects’). The latter were asked, to present three case studies in which ‘trauma’ had played a central role, and to discuss them with the interviewers.

The data gathered were then analyzed and evaluated in group discussions by the researchers of the project. Eliciting implicit aspects of the concept’s use from the explicit statements of the trauma experts was a central task. As I am not speaking about trauma today, but about conceptual research and its justification, I can – concerning the results – only refer you to the pertinent literature in the bibliography. But, of course, the results reflect the state of the discussion 20 years ago – and thus have a predominantly historical value. Today, as the trauma concept is much more in the focus, an update would certainly be warranted, especially against the background of neuroscientific findings on how trauma is dealt with.

But of course there are many more current ‘hot spots’: the concept ‘unconscious’ for example, which is being increasingly used by memory researchers. But in which sense do they use this term? In psychoanalysis the meaning of ‘unconscious’ is based on Freud’s considerations and has been clinically confirmed over decades. In memory psychology the meaning of ‘unconscious’ is based upon manifold experimental results and is closely related to the concepts of implicit memory and implicit knowledge, of those memory contents not accessible to consciousness. Psychoanalysis, which in the past was certainly not unjustly called the science of the unconscious, has evidently lost its monopoly on the concept ‘unconscious’, and it would be of great current interest to clarify the conceptual field surrounding this concept and contour the specific analytic connotation of ‘dynamic unconscious’.

There have been many studies focusing on concepts, and it is useful to consider them definitely as expressions of a ‘research program’ with a common methodological justification. Precursors of such systematic conceptual investigations can already be seen in Freud, of course, when he practiced the integration of empirical investigation and conceptual differentiation (as we can see for example in the concepts of ‘sexual’ or ‘unconscious’ (see particularly Lecture 20 in Freud, 1916-17). Indeed, conceptual studies have a long tradition and I would just like to mention a few briefly:

  • McIntyre’s (1958) work, which advises against the use of ‘the Unconscious’ as a noun, because of the danger of an ontological misunderstanding, suggesting instead an adjectival use as in ‘unconscious processes’;
  • another is a study by Pulver (1970), who warned against an inflationary use of the term narcissism;
  • and certainly there are the prominent dictionaries of psychoanalysis, which are the results of careful investigation by groups of researchers. These encyclopedias are useful tools to orientate oneself better among the variety of concept meanings. Laplanche and Pontalis (1973), then also Moore and Fine (1990) have been reliable companions; the new volume by de Mijolla (2002) will perhaps play this role in future. But what becomes evident in such a chronological sequence is, that even excellent dictionaries, though they are competent aggregations of the understanding of their decade, do need an update from time to time. The modified understanding of such a central concept like narcissism, for instance – an understanding initiated by Kohut’s work and not yet included by Laplanche and Pontalis but systematically integrated by Moore and Fine – may serve as an illustration.

Attempts to clarify concepts in discourse are, especially in the English-speaking world, nothing new. One must only think of the concept-study-groups at individual American institutes which see their work as just that: the study of concepts. An excellent example is the renowned Ernst Kris Study Group of the New York Psychoanalytic Institute (its work is documented in the Monograph Series of the Kris Study Group, e.g. in Fine et al. 1971; or in Abend et al. 1983).

The explicit intention to discuss concepts controversially, as well as seeking consensus or recording dissent, can be found especially in panel discussions. The aim of these panels is the clarification of concepts through the simultaneous inclusion of clinical experience und theoretical convictions. They have a typical pattern with speech, reply, and conclusion from all arguments and a closer look at this predominantly US-American tradition shows how imaginative and clarifying such investigations can be (see e.g. Blacker 1981, Escoll, 1983; Rothstein, 1983). Panels and the concept-study-groups are established and generally accepted models for activities which, so-to-speak, meet the minimum requirements for conceptual research.


An intermediate remark: conceptual reflection – conceptual research

The idea that an autonomous conceptual research could yield gains, for both clinical work and scientific endeavors, has become accepted in contemporary psychoanalysis. The Hampstead and later the Frankfurt project group had initially used the term ‘conceptual research’ only as a loose working title; in the meantime the label has established itself. But who is doing conceptual research? Every clinician who uses a concept to describe his case? Every empirical researcher who tries, as part of a study, to transfer aspects of a concept into measuring procedures for the purpose of operationalization? Anyone who thinks about a concept anywhere in any way, or who criticizes how others use a concept? Of course, ‘psychoanalytic conceptual research’ is not a protected term. However, I suggest, making a distinction between conceptual research and conceptual reflection – this distinction can be seen as complementary to the above mentioned working definition:

  • I would reserve the term conceptual research for such activities, which are exploring the use of concepts, including their function in conceptual fields, in a systematic way and applying adequate – empirical as well as hermeneutic – methods.
  • Whoever reflects on concepts (actually a routine activity for every clinician and researcher) is certainly doing something useful, yet not automatically research. Conceptual reflection is a necessary, but not itself a sufficient condition for conceptual research.


Some essentials of the research process

Let me take a closer look at some features of the research process: Beyond the ever important reflection in conceptual research there is first a need for appropriate empirical data. Such data alone can reflect a concept’s actual use in practice and how its users illustrate and legitimize its use. In order to get a comprehensive picture of both issues – factual use and justification of that use – not only a selection of suitable methods is needed; but in particular a choice of representative samples: historical authors, experts, or clinical practitioners.

The actual questions under study may of course vary considerably with regard to the perspective – either the development or just a current cross-section on a concept’s use. The inventory of possible methods thus is very broad, ranging from different forms of literature analysis to various interview techniques (from non-structured to structured interviews), to questionnaires, expert ratings, group discussions and the like. And as far as the logical consistency of conceptual fields is concerned, simulation models may indeed be useful. Especially when we are dealing with expert knowledge, the circle of specialists does not have to be limited to psychoanalysts. Thus, considering the canon of conceptual research methods, exactly the same applies that any other kind of research also claims for itself. Methods are, above all, means to an end, and the all-important question is, whether they contribute to reaching this aim: in our case, suggestions for an improved concept use for the scientific community.

That structuring reflection and discussion in a research team belong to the procedure of a research process is understood by itself. However, it is not so obvious that this discoursive moment is being systematically applied and considered as essential. But, without such attempts to communicate about the rules of concept use and to look for social consensus, the aim of conceptual research would hardly be attainable. In the course of such a research process a concept is repeatedly put on the test-bench of reflection, discussion, and criticism by the members of the research team – based on their previous knowledge and the systematically acquired knowledge base. The structure of this procedure is designed to gradually elaborate the meaning-space of a concept. Such working through, with the aim of continually approaching a better concept use, is at its best that “progressive spiral” described within the context of the Hampstead-Index (Sandler 1987).

In its claim conceptual research goes beyond the limits of school traditions and is often interdisciplinary. The bond to our common ground should not, on the other hand, be lost. Therefore the integration of clinical knowledge is indispensable. Making sure that this essential meaning of our concepts remains preserved and that there are no reductions thus becomes a gripping tightrope walk between preservation and change.


Conceptual research in the context of methodological problems

After this description of how conceptual research works there still remain – for a very last section – some issues, regarding its scientific status, to discuss. But you will see, that this will at the same time be a kind of summary:

  • Especially researchers from empiristic traditions hold the view that only observable or measurable phenomena can be the subject of research. Furthermore, the methods applied should be orientated toward the ideal of the experiment in the natural sciences. The aim should be theories and models as formalized as possible, which serve explanation and prediction. And for many it is above all the choice of the right method, which will determine whether something is seen as scientific. Conceptual research does not primarily define itself by its methods, but by its subject, the use of concepts, and by its aims, their clarification. Conceptual research is in its procedure empirical, as far as the description of the factual concept use is concerned; as well as historico-reconstructive, concerning a concept’s development; and also evaluative, regarding the critical discussion of the collected data and the elaboration of proposals for a clinically and scientifically adequate use. For this, it is self-evident that conceptual research has to use different methods from various research traditions.
  • Thus, conceptual research, just like empirical research, can recur to qualitative as well as quantitative methods. Nevertheless, it is clear that conceptual research works intensively with language use and meaning-spaces. Besides the data collection to describe the concept use, an essential aspect is the analysis and interpretation of verbal data (for instance from expert interviews or group discussions) and thus also the application of hermeneutic methods, well-known and familiar to us analysts. This central attention to language, as subject of research and as data source, can only be disturbing, when words are basically considered as opaque and numbers always as exact. Where conceptual research and empirical research do indeed differ is in their research interest, not necessarily in the class of methods applied and not at all in their intention to improve analytic knowledge.
  • Occasionally, when one finds analytic research activities divided, for instance, into empirical, clinical, and conceptual research, one can have the impression that these traditions stand rather unconnectedly side by side, or even in competition. This would be a reductionist view. Only constructive cooperation and mutual acknowledgement will do justice to our complex area of research. In that respect psychoanalytic conceptual research is always integrative, in the sense that it takes the different research results on a concept’s use into account – focusing on the question whether a concept meets clinical and scientific requirements equally well.
  • Let me come back to that already mentioned step, usually called ‘operationalization’, which requires the researcher to determine, what in a concept should be precisely observed and measured and with which methods these aspects of a concept should be grasped. From the perspective of conceptual research such detailed definition does present an advantage – compared to some diffuse or idiosyncratic concept use, for it makes its rules traceable. However, just as a concept can be understood differently, it can also be operationalised very differently. ‘Transference’ is a good example how researchers, each with their own technologies, try to grasp transference phenomena on the basis of different operationalizations. These operationalizations and the conceptual understanding they are based on can, of course, be – as a kind of concept use – compared and evaluated.
  • It is clear, that our concepts present some difficulties within this step of operationalization. There is always the problem, that our basic concepts, our model of the mind, must be taken into account, for instance, that we see many psychic processes as being dynamically-unconscious (and although they are not directly observable, they can definitely be disclosed as the result of a process of interpretation and inference). Such a problem, namely the dependence of the meaning of concepts from pre-assumptions about the functioning of the psyche, is not that of psychoanalysis alone. The convictions of many cognitive scientists and some neuroscientists are based on the pre-assumption, again not shared by all, that the human mind functions like a computer, that it actually even is one. Of course, the computer metaphor is considered more ‘modern’ than the system of our fundamental convictions derived from Freud‘s metapsychological ideas. But that this image of man is generally better, or clinically more adequate than ours may well be contested.
  • Conceptual research is characterized by its particular interest for concept changes. But – besides the progress in clinical experience and scientific research mentioned – what actually are the historical and social reasons for concept changes? It should be clear by now, that the relationship between language and world, between a concept and the pertaining empirical phenomena, cannot simply be seen as a two-digit relation, but according to old semiotic tradition as a triangular one. The person of the concept user namely comes into play and thus the dependence of the concept‘s meaning on historically grown, socially shared language games, embedded in different cultures and human life forms. To examine such, in the largest sense, cultural dependencies, conceptual studies can also take the perspective of cultural theory [Kulturtheorie]. For the changing trends in the research landscape do not principally withdraw from reflection. What exactly were the reasons two decades ago for the shift of analytic research interests toward empiricism? How do treatment aims change as a function of socio-economic changes in the health systems? Or what are the reasons for the current trend toward concepts?



Systematic reflections on a concept and its use do not automatically lead to a better understanding between different disciplines; they also do not bring a unified psychoanalysis back. But they could facilitate controversial discussions. Conceptual research can, by making its own suggestions, stimulate discussions, at best moderate the discourse, but cannot lead it alone. A reasonable concept use can in effect only be achieved through good arguments and consensus. Just as the empirical and the conceptual are in a dialectic relationship with each other, so psychoanalytic research activities which have either empirical phenomena or a concept as their subject are dependent on each other: conceptual research without recourse to the findings from empirical research can be a glass-bead game, empirical research without adequate conceptual understanding can be without clinical relevance. Cooperation is conceivable, based on an acknowledgement of the different methodologies, and not on claims of exclusiveness, as researchers and clinicians strive for a common aim: a theory on a solid empirical basis formulated with good concepts.




Abend SM, Porder MS, Willick MS: Borderline Patients – Psychoanalytical Perspectives, 3rd Edition. Madison CT, International Universities Press, 1986

Blacker KH (Reporter): Panel: Insight: Clinical conceptualizations. JAPA 29:659-671, 1981

Blum H P: Psychic trauma and traumatic object loss. JAPA 51:415-432, 2003

Bolland J, Sandler J: The Hampstead Psychoanalytic Index. A Study of the Psychoanalytic Case Material of a Two-year-old Child. New York, International Universities Press, 1965

Dreher AU: Was sollte man bedenken, wenn man Übertragung messen will. Zeitschrift für psychoanalytische Theorie und Praxis 14:260-283, 1999

Dreher AU: Foundations for Conceptual Research in Psychoanalysis. Madison CT, International Universities Press, 2000

Dreher AU: La ricerca concettuale in psicoanalisi. in Quale Ricerca per la Psicoanalisi?, Edited by Bonaminio V, Fabozzi P. Milano, Franco Angeli, 2002, pp278-303

Dreher AU: What are the aims of psychoanalysts? in Outcomes in Psychoanalytic Treatment, Edited by Leuzinger-Bohleber M, Target M. London, Whurr, 2002, pp17-29

Escoll PJ (Reporter): Panel: The changing vistas of transference: the effect of developmental concepts on the understanding of transference. JAPA 31:699-711, 1983

Fine BD, Joseph ED, Waldhorn HF: Recollection and Reconstruction – Reconstruction in Psychoanalysis. The Ernst Kris Study Group of the New York Psychoanalytic Institute, Monograph 4. New York, International Universities Press, 1971

Freud S: On narcissism: an introduction (1914), in The Standard Edition of the Complete Psychological Works of Sigmund Freud, Vol 14, Translated and edited by Strachey J. London, Hogarth Press, 1962

Freud S: Instincts and their vicissitudes (1915), in SE14

Freud S: Introductory Lectures on Psychoanalysis (1916-17), in SE15-16

Freud S: Two encyclopaedia articles (A) Psychoanalysis (B) The libido theory (1923 [1922]). in SE18

Green A: The pluralism of sciences and psychoanalytic thinking. in Leuzinger-Bohleber M, Dreher AU, Canestri J, 2003, pp26-44

Kuhn TS: The Structure of Scientific Revolutions. Chicago, IL, University of Chicago Press, 1962

Lakatos I: Falsification and the methodology of scientific research programmes. in Criticism and the Growth of Knowledge, Edited by Lakatos I, Musgrave A. Cambridge, Cambridge University Press, 1970

Laplanche J, Pontalis JB: The Language of Psychoanalysis. London, Karnac Books, 1973

Leuzinger-Bohleber M, Dreher AU, Canestri J (eds.): Pluralism and Unity? Methods of Research in Psychoanalysis. London, International Psychoanalysis Library, 2003

Mahler MS, Pine F, Bergman A: The Psychological Birth of the Human Infant. New York, Basic Books, 1975

MacIntyre AC: The Unconscious: A Conceptual Analysis. London, Routledge and Keagan Paul, 1958

Mijolla A de: Dictionnaire International de la Psychanalyse. Paris, Calmann-Levy, 2002

Moore BE, Fine BD (eds.): Psychoanalytic Terms and Concepts. New Haven and London, The American Psychoanalytic Association and Yale University Press, 1990

Parin P, Parin-Matthèy G: Medicozentrismus. in Subjekt im Widerspruch, Aufsätze 1978-1985. Frankfurt am Main, Syndikat, 1986, pp 61-80

Pulver SE: Narcissism. The term and the concept. JAPA 18:319-341, 1970

Rothstein A (Reporter): (1983). Panel: Interpretation: Toward a contemporary understanding of the term. JAPA 31:237-245, 1983

Sandler J: Reflections on some relations between psychoanalytic concepts and psychoanalytic practice. IJP 64:35-45, 1983

Sandler J (ed.): From Safety to Superego. Selected Papers of Joseph Sandler. London, Karnac Books, 1987

Sandler J, Dare C, Holder A: The Patient and the Analyst. The Basis of the Psychoanalytic Process. 2nd Edition, revised and expanded by Sandler J, Dreher, AU. London, Karnac, 1992

Sandler J, Dare C, Holder A, et al.: Freud’s Models of the Mind. An Introduction. London, Karnac, 1997

Sandler J, Dreher AU, Drews S, et al.: Psychisches Trauma: Ein Konzept im Theorie-Praxis-Zusammenhang. Materialien aus dem Sigmund-Freud-Institut Frankfurt am Main 5, 1987

Sandler J, Dreher AU, Drews S: An approach to conceptual research in psychoanalysis illustrated by the consideration of psychic trauma. IRP 18:133-141, 1991

Sandler J, Dreher AU: What Do Psychoanalysts Want? The Problem of Aims in Psychoanalytic Therapy. London, Routledge, 1996

Sandler J, Sandler A-M, Davies R: Clinical and Observational Psychoanalytic Research: Roots of a Controversy. London, Karnac, 2000.


In addition:


Dreher, A. U. (2005). Conceptual research. In:E.S.Person, A.M.Cooper, and G.O.Gabbard, Textbook of Psychoanalysis (pp. 361-372). Arlington, VA: American Psychiatric Publishing, Inc.

Dreher, A. U. (2010). Pluralism in theory and in research—and what now? A plea for connectionism. In: M. Leuzinger-Bohleber, J. Canestri, and M. Target (Eds.), Early Development and Its disturbances: Clinical, conceptual and empirical research on ADHD and other psychopathologies and its epistemological reflections (239-266). London, Karnac Books.

Dreher, A.U. (2014) Wie heilig sind uns psychoanalytische Konzepte? In: Forum Psychoanal 30: 11-26.

Dreher, A.U. (2014) Psychoanalytic research with or without the psyche? Some remarks on the intricacies of clinical research. In: Boag, S. & Brakel, L. (eds.) Philosophy, Science, and Psychoanalysis. London, Karnac Books.