Response to Commentaries

Almond

I confess to having some difficulty responding to Almond because as far as I can tell he is criticizing an argument I did not and would not make. In fact, if I had written the paper he thinks I wrote I would probably agree with most of what he has said, including his characterization of me as “naive” (p. 40).

At the risk of claiming to be misunderstood (which has been characterized as the defense of the guilty), let me spell out what I see as Almond’s misreading. The problem begins in his first paragraph. “What,” he writes, “is the epistemological status of our theoretical models? Greenberg’s answer: they are powerful fictions, shared among communities of believers, controlling our thought and action” (p. 33). I must say that the image of models that exert control over thought and action is evocative if somewhat terrifying, but this signals the start of a conceptual slippage that runs throughout his paper.

The slippage is this: As I have said, my goal is to spell out the way theories are structured, top down, from the overarching and controlling fiction to the elements of the psychoanalytic situation that are derived from it (Blass accurately characterizes this intention [p. 47]; Boesky [pp. 70–71] falls victim to the same misreading as Almond). Almond, in stark contrast, is talking about the way in which clinicians use theories. This mistaken impression runs through much of his critique. I won’t belabor the point, but do want to note a couple of instances: “Greenberg . . . retains the sense that those who hold to a given model claim its exclusive and exhaustive reach” (p. 36; emphasis added). And again: “Another assumption that underlies Greenberg’s idea of controlling fiction is that every analyst adheres to a particular fiction” (p. 38). Finally (actually not finally, but sufficiently), Almond reports on his experience in study groups, from which he concludes that “analysts working from different theoretical bases can over time arrive at a common understanding of a case” (p. 39). This last observation is true (assuming, as I’ve noted above, that the analysts have something in common to begin with) if commonplace, but has no relevance to the points I am trying to make in my paper. The issue of how analysts wield their theories is complex and interesting. It is, among other things, an empirical question that, as Almond notes, has been studied by Hamilton and Tuckett among others. In my paper I tried to steer clear of it for the most part, though I did nod to it in writing that “the analyst’s theory of therapeutic action is most often implicit (Sandler 1983), and it will not necessarily match any of the formalized theories that are written about” (p. 28; emphasis added). My concern here, to repeat, is precisely the structure of “the formalized theories that are written about” and the place that theories of therapeutic action occupy within these formal structures.

Consider Almond’s discussion of my comments on Civitarese’s vignette. I wrote: “I hope it is clear in what I am saying that I am not assuming that one way of understanding the dynamics of this exchange is more ‘correct’ than another, nor do I have any information that would lead me to believe that one intervention is more analytically or therapeutically effective (this is, of course, true of virtually any clinical vignette)” (p. 25). In Almond’s rendering this becomes: “The idea that the analyst may be distracted from the ‘right way’ by a fiction/theory, or by the press of the situation, oversimplifies a complexity that Greenberg acknowledges . . .” (p. 42).

To my eye this reflects Almond’s attempt, perhaps unwitting, to impose his controlling fiction on mine and illustrates how doing so can inhibit generative conversation. The attempt to make things pivot on judgments of “right” or “wrong” assumes a “correct” psychoanalytic technique that can be specified a priori. Along with concepts like “complete” or “exact” interpretation or “resolution of the transference,” these are elements derivative of the controlling fiction that shaped mainstream psychoanalysis in North America until the 1980s or so, vestiges of which remain in some but far from all contemporary points of view. It fits poorly with the point I was trying to make, which is that any intervention—any piece of analytic process, actually—can be understood only in the context of the assumptions that guide the treatment as a whole. Almond nods to the value of this idea (p. 38) but then undermines its central importance (p. 39) by alluding to the theoretical convergences I have noted, thereby salvaging a concept of “correctness” that transcends theoretical structures, as well as the possibility of adjudication. But that, of course, is his project, not mine.

Blass

Blass begins her discussion by claiming that our views are not “ungrounded fictions.” Rather, she writes, “they are expressions of my best effort to grasp the reality in which we live . . .” (p. 47). The crucial shift—from the concept of fiction to ungrounded fiction—fails to acknowledge that all serious fiction reflects the attempt to “grasp the reality in which we live.” Note in this connection my discussion of Davidson’s view of metaphor, which he says “draws something to our attention” (1978, p. 218). That “something” is a way of course of viewing the world; there is nothing “ungrounded” about either successful metaphor or successful fiction. Freud certainly knew this as well as anybody ever has, even as he struggled to establish psychoanalysis as a science.

Blass stakes out a very clear position regarding the relationship between observation and theory, a position she presents, inaccurately, as uncontroversial: “Our concepts are responses to reality” (p. 50). Certainly the most famous alternative to this epistemological point of view in the history of psychoanalysis comes from Freud himself: “Even at the stage of description it is not possible to avoid applying certain abstract ideas to the material in hand, ideas derived from somewhere or other but certainly not from the new observations alone. Such ideas—which will later become the basic concepts of the science—are still more indispensable as the material is further worked over” (1915, p. 117).

The relationship between observation and basic concepts as psychoanalysis has unfolded over the hundred years since this was written is particularly illuminating, because the “abstract idea” to which Freud was referring was, of course, instinct. Instinct structured psychoanalytic observation for a very long time, certainly throughout Freud’s lifetime, even though he significantly modified the concept shortly after writing this passage. Today the concept of instinct is highly controversial; some analysts retain it in one way or another, while others have long since discarded it. And yet the concept itself facilitated both observations and further theoretical elaborations that have become foundational for all of us. Can we say that the concept of “instinct” is both fictive and generative? That possibility doesn’t trouble me, though it does seem to bother Blass and many others.

Blass claims that there are different ways in which to array theories, and she lists several. To my eye, each of these is an old warhorse that has been trotted out too many times in the service of belittling adherents of points of view with which one disagrees. I will consider several.

First, theories operate on “different levels of inquiry.” Feinting toward fair-mindedness, Blass says that if we look at things this way “several theories can be true without there being any clash between them” (p. 52). Here the contrast between Blass’s point of view and my own could not be more apparent. She believes that theories can be “true”—and presumably that they can be shown to be—a proposition I find problematic at best. But a more important question for us to ask ourselves is whether we find theories different from our own interesting, which as I read Blass’s argument I can’t imagine she does.

Consider: “some theories focus on the person as more whole and conscious and others on the person as more divided and unconscious”; the whole person “can exist alongside an understanding of the person insofar as he is influenced by unconscious determinants . . .” (p. 52). Would Blass be interested in having a conversation with a psychoanalyst who does not believe that people are influenced by unconscious determinants, or who believes that this influence is not of great concern or interest? Would she even agree that the conversation between them was addressing competing psychoanalytic points of view? Shortly, we learn from Blass that in fact she would not, because there is, it turns out, a “level” that is a “truly psychoanalytic one” (p. 53).

In addition to its dismissive implications, there is a logical problem with Blass’s idea that there is a “level” of theorizing that is “truly psychoanalytic.” The idea assumes that “psychoanalysis” is what philosophers call a “natural kind,” that is, a grouping that exists in nature rather than one that is brought into being as a result of its definition by observers. Although I am far from an expert in these matters, I doubt very much that any human invention—psychoanalysis, for example—can possibly be considered a natural kind. To say that something is “truly psychoanalytic,” then, is simply to claim membership in a group that agrees on a definition of psychoanalysis. Or, to put it another way, it amounts to claiming membership in a group that subscribes to a particular controlling fiction. Contra Blass, I don’t think the problem is that deciding “which theory is most psychoanalytic is not a question to be readily resolved” (p. 53); the problem is rather that asking whether one theory is more psychoanalytic than another is what philosophers call a “poorly formed question.”

From here Blass goes on to suggest that theories differ on the basis of “competing world views of the person.” She acknowledges that adjudicating among these poses a problem; she doesn’t want to characterize either Kleinian theory or Kohutian self psychology as not psychoanalytic. So far so good, but then she accuses me of having bailed out of the problem; according to her, I claim “that the understandings offered by the theories are narratives detached from reality and that there is no way of assessing their value and validity” (p. 54; emphasis added).

The problem here is that effective fiction is not and cannot be “detached from reality”; its effectiveness in fact depends upon its being anchored in reality, though differently from the way scientific hypotheses are anchored in reality (the symptoms of tuberculosis predict with certainty the presence of the tubercule bacillus), and differently from how mathematical formulas are anchored in reality. Is Oedipus Rex detached from reality? Are Picasso’s cubist portraits? Is Sandburg’s image of the fog creeping in “on little cat feet”?

I suggest not only that these fictions must be anchored in reality, but that they can generatively interrogate other fictions addressing similar themes. Thus, Sophocles’ Oedipus interrogates Aeschylus’ (mostly lost, though enough is known of it to make it clear that his version of the story is very different); Picasso’s portraits are in conversation with Matisse’s, and vice versa; and Sandburg’s depiction of the fog can be in conversation with other metaphors used to describe it.

It goes without saying that there are bad tellings of the Oedipus story, there are bad portraits, and there are bad metaphors. How different creations (and different psychoanalytic theories) work themselves in and out of the conversation is another interesting topic. It is not the topic I am addressing in my paper, but it is worth mentioning briefly because Blass introduces and then flounders around in it.

She takes me to task on two issues. First she asserts that “there may be little value in Kleinians and Kohutians jointly discussing clinical material with the intent of understanding a specific patient” (p. 55). But despite what “Greenberg seems to imply,” she writes, there is value in having a conversation between adherents of these positions because it would “provide a sense of what the worldviews are about in an experiential kind of way” (p. 55).

It is difficult for me to see how this last point in any way stands “in contrast to what Greenberg implies.” As I said in my paper and have reiterated here, these sorts of conversations are precisely what I am suggesting, whether they lead to changing anyone’s mind or not. I think that alongside our clinical work (with which they share many similar characteristics) they constitute the vital core of psychoanalysis. It is, in fact, Blass who seems not to value the sort of clarifying discussions she’s endorsing here; recall that at the beginning of her paper she writes that there is “no real value” in dialogue that doesn’t lead to adjudication. So here it seems that she is arguing more with herself than with me.

Because she acknowledges that both Kleinian theory and self psychology have some psychoanalytic respectability, they can be welcomed into our discourse. But some ideas cannot (e.g., archaic ideas about homosexuality and female sexuality). Guessing correctly that I wouldn’t relish spending a great deal of time debating these ideas, in a kind of reductio ad absurdum argument she wonders whether I actually support my own position (p. 56). My answer, perhaps surprising to Blass, is that I do. But I agree that she has put her finger on a vexing problem, because I imagine that there are some conversations that both she and I would be interested in joining and others that neither of us would be eager to participate in.

Where does each of us draw the line? How do some ideas fade (Freud’s views on homosexuality and female sexuality, his definition of pleasure as discharge, his notion that a mother’s bond to her son is the only human relationship that is free of ambivalence, etc.) and how do some find their way into the conversation after years of exclusion (the informative role of countertransference, the influence of real people in the lives of our analysands, the power of the analyst’s participation in the analytic process, etc.)? These questions, of course, have long histories in the philosophy of science, but they are not central to my argument. It comes down to this: we (I refer here both to individual analysts and to the marketplace of ideas) do make evaluative judgments about some ideas, but there are others that have proven clinically useful and conceptually generative to large groups of analysts and yet are so anchored in basic assumptions (controlling fictions) that they resist adjudication. We have different kinds of conversations about these ideas.

Blass seems not to recognize what is involved in these different kinds of conversations. Consider her response to Civitarese’s vignette. She writes that “simultaneous reflection on theoretical claims and specific clinical instances in light of our experience of reality, including our own inner reality, can give us some sense of the value and validity of Civitarese’s clinical work” (pp. 57–58; emphasis added). To me this claim seems self-refuting: If we must include our own inner reality in our reflections, and if Blass acknowledges that each of us has his or her own inner reality and that these realities differ substantially, how can we expect that we will always or even reliably reach any kind of consensus on the value or validity of a treatment?

Of course, it is precisely the ways in which inner reality affects our clinical and theoretical sensibilities that is at the core of my argument that while adjudication is difficult even to define, other kinds of outcome are interesting and generative. To bolster my point of view, I would ask Blass to reflect on her own experience: Has she ever participated in a conversation in which an analyst abandoned or even significantly modified his or her basic assumptions? Toward the end of her discussion of Civitarese, she seems to acknowledge that such change is unlikely, suggesting that while we “cannot provide certain proof regarding the truth of a theory, what [discussion] can provide is a deeper and more precise understanding of one’s own theory as well as those of others” (p. 60). At the risk of repeating myself, this is precisely my point.

Let me add that many of the questions Blass raises about Civitarese’s vignette strike me as both important and interesting; my quarrel is not with asserting the value of asking those questions, but with her goal in asking them. Her goal is determining whether what Civitarese said was “a good analytic intervention” (p. 59), which strikes me as both tendentiously framed (good? analytic?) and impossible to achieve unless the group discussing the clinical material is (like Almond’s group) unusually homogeneous.

Boesky

Boesky begins by interpreting what I have said in an interesting but problematic way. He quotes my comment that “it is relatively easy to say what we are trying to accomplish when we meet with patients; put in the broadest possible terms, we are hoping to help deepen their experience of themselves and of the worlds in which they live” (p. 65). He then takes issue with this way of putting things, writing that “had he said that we wished to help our patients deepen their experience of themselves and of their own unconscious mind I could say I agree with him” (p. 65). But he doesn’t agree with me, evidently because “deepening experience of the world” does not seem to him to necessarily involve unconscious processes.

Boesky attributes this difference to my “epistemic convictions” (p. 65), an explanation I don’t exactly understand. To my eye the difference is theoretical; in accord with contemporary analysts working with a range of conceptual models, I believe that our experience of the world— object relations, history of trauma, adaptive strategies, and so on—is largely unconscious. Accordingly, deepening our experience of the world does indeed entail deepening our experience of our unconscious minds, as much so as learning about our drives and defenses.

Here Boesky, rather like Blass, seems to invoke the “depth” of our psychoanalytic probing. I would suggest to both that depth is par excellence a metaphor that comes into existence only within the context of a particular controlling fiction. Which is “deeper,” drive or relationship? Ask Freud. And then ask Kohut or Fairbairn.

That said, I have a hunch—only a hunch, to be sure, but one that intrigues me—that in other and more important respects Boesky and I may be more in agreement than he imagines we are. I say this because I suspect that his notion of “contextualizations” that ground observation and theorizing is not so different from my concept of controlling fiction. Consider this formulation: “Where Greenberg concludes that ‘nothing within the psychoanalytic situation can be understood . . . in the absence of some controlling fiction’ . . . , I would say that nothing in the psychoanalytic situation can be understood without understanding the context in which the analyst understands something” (p. 75). And, as he notes throughout, different analysts contextualize things differently; this accounts for the epistemological controversies to which he alludes.

Let’s examine how Boesky’s “contextualization” shapes the analytic process. In his discussion of Civitarese’s case, he notes that there is a basic epistemological question “of what counts as data” (p. 68). Like Blass, he goes on to raise a number of questions: Is this the first time the patient has worried about her parents’ marriage? If not, why is it scarier now? What is the analyst’s idea about why he felt a particular urgency? The list goes on, and each of Boesky’s questions strikes me as well worth asking. (Note that his questions are quite different from Blass’s, which focus more on the state of the transference-countertransference.) I can imagine many other questions as well, also worth asking. Some might be based in curiosity about the patient’s early experience of separation or her history of trauma, some might focus on the analyst’s personal fears of harming people in his care, some might address the events of the previous day’s session. And so on.

Of course not all questions can be asked at once, and so the data with which any analyst is working will be particular to that analyst and to the specific analytic dyad. How does any analyst choose which questions to ask; to put it another way, how does any analyst generate the data that will then be observed? Boesky says the choice is shaped by the analyst’s epistemology; I say that it is shaped by the analyst’s controlling fiction. ¹ Although the phrase I’ve chosen is certainly more provocative than Boesky’s, it seems that his “epistemology” and my “controlling fiction” play a similar role in the structure of our inquiry and of our clinical process. Both organize the analyst’s observations and determine what counts as data.

Another hunch: Boesky, committed to viewing psychoanalysis as a science and to the possibility of validation, would object strenuously to my offer of rapprochement. But consider his complaint about the field’s failure to think epistemologically: “few pages [in our literature] are devoted to the methodology by which we argue for conflicting methods of contextualizing” (p. 75; emphasis added). In other words, he thinks it is as daunting to adjudicate among conflicting epistemologies as I think it is to adjudicate among competing controlling fictions. Or perhaps he doesn’t, although surely he is aware that philosophical discussion over millennia has resolved very little. Boesky may be satisfied that the Hanlys’ critical realism has settled the issue, but only the professional epistemologists who agree with him will sign on to that point of view.

Boesky concludes his discussion asking whether I am “sounding the death knell for the enterprise of a comparative psychoanalysis” (p. 81). I find this a curious question, especially since in the my abstact I characterize my paper as a “venture into comparative psychoanalysis.” I do believe that as our comparative psychoanalytic projects cast a wider net (to include theories that have been developed within non-anglophone communities and that therefore are not rooted in basic concepts familiar to North American analysts), we must be increasingly careful to understand not only the overarching ideas that shape different approaches but also the ways in which particular terms are used within different traditions. In my paper I make quite a lot of the different meanings the concept of interpretation has in different theoretical models; in this response I have pointed to similar differences in understanding of what is meant by the dyad. Doing comparative psychoanalysis has never been easy, and it becomes even more difficult as our conceptual horizons expand. The project is more important today than ever, but it does demand that we be prepared to probe more deeply into the structure of the theories we are considering.

In this connection I want to address a paragraph in Boesky’s paper in which he makes a series of claims about what I have said that seem inaccurate to me. First, he suggests that I believe that “no one element in therapeutic action is more important than any other” (p. 72). As I have already mentioned, I see myself as more agnostic than that; in fact I have no idea whether one element is more important than another (Always with every patient? Sometimes with some patients? At different times with the same patient?). What I do know is that the writings of various authors sensitize me to different elements of therapeutic action: interpretation, containment, mirroring, confrontation, recognition, taking a second look, and so on. To the extent I can hold these in mind and allow them to interrogate each other without pressing for resolution (of course most of this goes on preconsciously or even unconsciously), my experience of the work I do is deepened and enriched.

Then, broadening his theme and leaving the argument of my paper far behind—Boesky claims that I “will later state that no one model is better than any other” (p. 72). Again, I don’t know how to answer this, in part because I don’t quite understand the assertion. Better for what? For whom? On what evidence? Given the current state of our knowledge, isn’t it “better” to acknowledge that many different points of view have been argued persuasively and that the more perspectives we can keep in mind the better off we are. This isn’t a plea for ecumenicism, which I know full well is often a rationale for muddled thinking, but it does argue for what I think of as a healthy analytic skepticism.

Concluding Remarks

As I read the paper I wrote in light of these three commentaries, it strikes me as something of a “report from the front,” detailing the state of psychoanalytic conversations as I understand them. My sense is that over the past quarter-century we are talking more to each other, that we are inviting adherents of a wider range of theoretical points of view into our discussions, that communication among analysts working in different cultural traditions is increasing rapidly. It is also my impression that these conversations have not typically led to adjudication among the different perspectives, to validation of any one point of view at the expense of others, or even to much synthesis. But the different approaches do interrogate one another, both in our public forums (journals, conferences, etc.) and privately, in our own minds. To me—I know not to everybody—this indicates the strength of psychoanalysis; as I’ve suggested, analysts who are aware of alternative points of view are better equipped to work within the theory they embrace. Similarly, I doubt very much that my discussants and I will find a great deal of common ground or synthesis; some of the issues that divide us are too fundamental and cut too deeply into our ways of seeing the world for that to happen. But I do hope that we will continue our conversation and that we will nag each other in ways that help us sharpen and refine our own thinking.