Here's Lookin at You, Id

Projective Identification as a language game (via Wittgenstein and Humphrey Bogart)

(Image above by Jean Dubuffet)

Early in my career, I interviewed for an internship at a public clinic for lower-income patients, run by a prominent Kleinian analyst. During the interview, I described a couple of cases and their difficulties. The interviewer interrupted me and, calmly sipping her tea, told me that any difficulty arose from my hostility toward the patients: I was clearly enraged and wanted to destroy them. (And, did I care for a biscuit.)

I stayed away from object relations for years—assisted by the fact that Klein was the bête noir of a psychoanalytic institute which centered on relational and intersubjective-systems theory. With the latter came an emphasis on phenomenology and contextuality; in reading Donna Orange I discovered the work of Ludwig Wittgenstein, a helpful way of understanding how clinical dialogue works (and doesn’t.)

But Kleinian ideas still hold interest for me, for the ambience they invoke, specifically that of dread, danger, and borderline terror. Klein hearkens to the black and white, moody shadow-worlds of film noir, classic films with Bogart, Mitchum, and Stanwyck. Here in the grime and shadow, as in the Kleinian world, it is everyone for themselves, greed, envy and destruction run rampant; no one is trustworthy. In fact noir’s classical era coincides with Klein’s apex.

In Klein’s descriptions of projective identification (or PI), arguably her signature idea (with the corresponding paranoid-schizoid position), it is the infant who becomes the femme fatale, seeking to destroy what it cannot have, or what (it is implied) the greedy little baby might not even deserve—but who still plows ahead like any good “bad” fatale.

In this column, I take a closer look at Klein and PI, given my growing interest in Winnicott and Bion: fruits from the Kleinian tree.

Both Winnicott and Bion, in their later work especially, moved away from the starker binaries of Klein, toward a more nuanced view of a holding environment or dyadic congruence (or emotional truth) in O. Quite a few contemporary authors—Lew Aron, Philip Bromberg, Thomas Ogden—expand upon the mutual aspects of PI. I am particularly taken with Stephen Mitchell and Robert Caper’s ideas of the non-verbal “dance” between infant and caregiver, patient and analyst, the waltz of the unsaid.

But for now let us return to the unfiltered original, the black and white rather than the color version, where I find near-gothic anxiety portrayed with the thick, scratchy rawness of a painting by Jean Dubuffet.

What draws me to PI is that it often seems to be happening much as Klein described; here, in rough-edged clinical moments, lifeis suddenly seen as limitlessly paranoid-inducing: one is on guard against fatality at every turn. Such rawness has vitality and punch, albeit stubbornly hard-to-articulate (and often menacing—yet still unnamabile, and therein lies the rub.)

In such unspoken moments, a persecutory anxiety and edginess abounds from an unknown source. I sense I am undoubtedly “screwing up,” failing or hurting the patient (or both.). I feel inadequate, even ashamed, given my alleged (possibly phony) expertise. It is so abrasively primitive that it resists naming or articulating, as if doing so were dangerous.

Or suddenly I appear naive, a sap even, in thinking that patients may want to “go deep” rather than get a quick fix and run, especially those patients ensnared in addictive activity. Just give em the “tools” and cash the check, Doc, says my Bogart-like voice: hand over some or another false reassurance; in the end we all want the shortcut, heh-heh.

The noir-like paranoia of the moment renders analytic thought and dialogue ineffective, tilting toward the subtly savage: eat or be eaten—as if anything I might say will devour hope or enrage the patient (it is, again, paranoiacally implied), as if I am cast against my will into a fatale-like role: the bad breast on steroids, with a restriction on the dialogue that might help us out of the jam, as if language itself is laughable. Talk is cheap, my friend. Barkeep, another round.

It is precisely this stubborn inarticulation, the refusing non or pre-verbalized denseness of the moment I focus on here; characterized by its gaping distance from the ability to think or find mutually verbal sustenance or traction, or to “locate” oneself theoretically, relationally, amidst a flood of affective angst or instability (whose is whose?)—as if the mask of my usual analytic persona has been washed away.

I have to come to think of such moments as a sort of language game, a variant on Ludwig Wittgenstein’s famous metaphor for our spoken forms of life, as he put it. This is a more primal form, a kind of shadow dance, an unspoken yet powerful affective moment contrasting our the usual verbal patterning between us.

For Wittgenstein (also a contemporary of Klein’s), language games reveal the character or “personage” of our verbal activities or dynamically lived co-existence—within contexts employing some or another “genre” or variety of idiom or dialect. Language shapes and illuminates context, and vice-versa; our metaphorical games are as varied as chess, poker, or pin-the-tail-on-the-donkey. (In PI, the analyst often feels like the donkey.)

Just as music is defined by the players, genre, arrangement, and defining era, language takes on meaning in specific times and instances of use. To believe that words adhere to pre-set definitions, including or especially psychoanalytic terms (such as “transference,” “unconscious,” “projective identification”)— a free floating a priori definition—is illusory, like saying A sharp or B flat have “meaning” outside of any song or score.

I would add that analytic theory is the score, clinical dialogue the performed interpretation. (If we wanted to stretch it, this column is my “cover” of PI; I often think of free jazz when I read Klein.)

There is ambiguity in all theories, including Freud, Stolorow, or Klein’s. Theory often reads as a kind of ideal, the aspiration of a successful treatment.

We might for instance ask what Klein (or any theorist) meant when saying projective identification is a “phantasy,” given that we cannot “read the minds” of infants to begin with. Nor does behavior alone give us a “window into” the psyche. This points to how our entire field is one of inference, reflection, and speculation, not direct observation. (This was Wittgenstein’s issue with Freud’s occasional sojourns into positivist intrapsychic mechanics, language more experience-is than experience-as; relational theorists, including myself, see experience as more inclusive and not entirely separate from cognitive interpretation, especially in clinical settings.)

Wittgenstein urges us to look at the specific use of words, and the words surrounding the word, in understanding what he called the “grammar” of meaning, as “to understand a sentence is to understand a language”. And to understand a person’s language, I would add, is to understand their world.

To understand a theory is to understand something about our own subjective therapeutic or analytic (or philosophical) stance, as outlined in Faces in a Cloud by Stolorow and Atwood.

Bion (a dyed-in-the-wool Kleinian, Grostein reminds us) says something similar about language, making him a kind of sibling to Wittgenstein; the two were roughly contemporaries, and both served in highly dangerous scenarios in World War I. Bion well understand the limitations of language in conveying the fullness of traumatic experiences. He believed that dreams were our most advanced communications.

Bion said he “died” on the battlefield in France; Wittgenstein survived an insanely dangerous sentry post on the front lines, and (more perilously) a tyrannical father who drove three of Ludwig’s brothers to suicide.

Both of these powerful thinkers understand the shattering of language and the impossibility of clear thought under impossible conditions—an issue germane to contemporary analysis (and our current political moment). They knew words cannot protect our psychic vulnerabilities, are “disobedient” in spite of what we want or hope them to do, often wandering in errant directions like restless children (especially in our digitized age.)

One simple example. A patient once took me to task for suggesting he try medication. In fact I had suggested, the previous session, that he look into meditation.

Bion frequently asked “what language” any particular patient was speaking. We might ask the same of our theorists. So what kind of “language” or language game is PI? Klein says surprisingly little about it in her published work, though it is reported she made much use of it in supervision and in her unpublished notes (Spillius, 2007.)

What she does say in her famous paper, “Notes on some schizoid mechanisms,” is that in PI the infant “expels his harmful excrements [or] split-off parts of the ego…into the mother. These…bad parts of the self are meant to not only injure the object but also to control and take possession of it” (quoted in Spillius, 2007, p. 102.)

Here the mother comes to resemble Fred MacMurray in “Double Indemnity,” easily manipulated by the beautiful, scheming Barbara Stanwyck; this classic Billy Wilder flick was released around the same time as Klein’s paper.

Klein says in that paper, “Since the projection derives from the infant’s impulse to harm or control the mother he feels her to be a persecutor” (p. 102, my italics.)

An apt description of Stanwick or other persecutory fatales. The baby here sounds almost manipulative; eventually, in time, it (like Fred MacMurray) will understand that the truest corruption always lies within, that greed and envy are our human defaults.

However. I can also imagine such an aggressive infant as emotionally desperate—so alienated as to desire to inhabit the mother’s mind, given a potentially unstable psychological oscillation between self-(m)other in the earliest of surrounds. Recall the baby is entirely dependent on its caregivers for survival. Perhaps the baby is already feeling compelled to know the mother’s mind, to be the mother, as in environments demanding pathological accommodation.

In the context of the consulting room, a paranoid sense of a persecutory or disaffected other can feel true, vividly so, to a beleaguered parent (or analyst); Winnicott presciently observed a caregiver’s occasional hate (exasperation?) for the demanding baby, as happens in clinical scenarios also. Those we care about can sometimes drive us crazy, due to a sense of “losing our minds” in the fraught surrounds I describe.**

Klein’s mechanisms (her word) are most often unilateral; she talks about the mother’s projections, but the rawness of the infant’s maneuvering is where the Kleinian “juice” is; even when more contemporary analysts talk about co-projection, the patient’s “evacuations” is still for the most part the center of focus.

One understands why Winnicott felt compelled eventually to pronounce “there is no such thing as an infant” (i.e., without the pairing of a mother.)

And there is no such thing as a patient—or a theory—without an analyst. At the same time, perception is powerfully influenced by affective chaos, an impact even analysts tend to underestimate. (This is an aspect of bewitchment also; we toss around the word “trauma” and in such repetition, benumb ourselves to its powerful psychic force.)

Recall some shadowy clinical moments where it appeared you were under some hypnotic spell: a pressure to act, speak, or be a certain way (your own way suddenly insufficient). This is how I felt with the supervisor mentioned at the start, though I now think the anxiety she picked up on had to do with my apprehension of meeting such a prominent analyst.

In the clinical moment, the patient is exasperated or enraged, it appears (at first), due to my not “coming through” with the goods: the offering of the good breast that would “make it all better.” To ask would be stupid—I should simply know. I appear precariously balanced upon what Ogden calls the primitive edge, as if looking into an abyss, where some unending Sophoclean agony awaits, a depressive position without end. The treatment is going bust, the patient about to exit injured, depleted of money and hope. Better luck next time, Doc.

It is tempting to take Klein’s imagery at face value, her persuasive picture of PI as holding some “quintessence” or definition, as if PI exists in pure form—an idea Wittgenstein (and I fancy the later Bion) would say is an error (thus leading to more contemporary revision).

But my point is about language itself; to say any analytic description is a definite example of this or that is highly misleading, as if there is only one way to perceive clinical phenomena: a metapsychology, arriving with the assumption that we can stand outside intersubjective dialogue with God’s-eye clarity.

I do not buy into the Platonic notion, clinically at least, that any psychological notion exists in “pure form”. We dialogue with theorists as we do with patients. Classical ideas are ever revised in contemporary eras, and every patient speaks with their own psychic “dialect”. Furthermore PI-like moments do not all occur in the same way, though it is often presented as such in first-wave Kleinian literature, via the projective “mechanisms” in the title of the above paper.

But the terms “projective” and “identification” do not carry inarguable, unfixed meanings. We might ask, what is being projected into whom and when, evoked by what: a chicken and egg dilemma. We might ask who is evoking what in the other. This is the ambience I speak of, often carrying unformulated (D.N. Stern) inferences.

There is not one “prototype” of projection; such moments are dense, rife with affectivity—anger, rage, and wounding, together with a longing for relief. They cry for dialogic understanding, a co-definition specific to an emergent means of communicating. (Bion: “participants devise a means of communication while communicating.”)

We posit that the infant in projecting into the mother; might this be expressing a longing to be “inside” the mother, a kind of reuniting or residence in her mind, a way of shoring up one’s own existential presence? The baby’s ferocious determination and maneuvering makes sense if this is an infant under threat, especially given the malattunement (it appears) regarding Klein’s passive mother.

The intensity of longing can lead to the errancy of words mentioned above; including analysts grasping at theory to help us in fraught moments. Whether this helps us understand the patient, and whether the patient feels recognized, remains to be seen. I acknowledge an urge to “get a grip” on the proceedings, re-ground myself, via good-enough theory language. Donna Orange remarked these language games are “like family” to us.

The language of these familiar family-worlds construct our analytic identity or sensibility, an “idiom” and purpose; in PI-type moments all of that sinks into quicksand, exiled from our connection to our (theoretically) familial “mother tongue.”

As Wittgenstein says, “you learned the concept ‘pain’ by learning language” (2009)—learned in other words how to recognize and express pain to a (hopefully) empathic other: the primitive language game of emotionality (as when a caregiver learns to “read” their baby’s cries.) If a caregiver is withholding, the infant may need to “get inside” the caregiver’s thoughts to “find” him or herself, confirm existence, since our first relational home is after all in the “reflective” recognition of the mother/ caregiver’s face and presence (Winnicott.)***

For those of us grounded in the relational home of a favored theory, the primal disconnection of language can be disorienting or threatening, as if we too are drifting in space, isolated, far from home.

As with “pain,” we learned “projective identification” when we learned Kleinian language, serving in my recasting as a portal into our own subjectivity, illuminating attachments to our theoretical language-worlds amidst complex clinical dynamics.

I admire Klein for attempting what is very difficult for psychology, in understanding the absence of affect, implied or enacted albeit unformulated dialogically—the mutually unthought known (Bollas.) We speak of dissociation though I wonder how we know or propose to know what is dissociated by patients? What moves us beyond informed speculation, or is that all there is to it, as said affectivity “exists” in absentia.

Much analytic theory has this there/not there quality, such as “phantasy”; do we say a phantasy “exists” in the baby/patient’s mind, as mediated by our own? There is also “unconscious meaning,” “organizing principles,” or even “intersubjectivity,” all frames rather than windows into essential meaning. We imagine our way into dialogue as well, via the ideas and metaphors entertained here. In any theory, our own imaginations and associations are aspects of absorption and their clinical “application”; perhaps “translation” is a more accurate term for such actions, themselves then interpreted.

Perhaps the underlining point of all of this, for this analyst, is the ways Klein’s darkly magnetic descriptions are redolent of alcoholic family systems, darkly self-serving and chaotic—noir-like to the core, and deeply familiar to this author. Such mutely turbulent systems are marked by covert or outright attempts at mind control, demands that children attune, adapt and faithfully mimic the thinking and talking of parents: a source of narcissistic support, without the emotional honesty unstable adults cannot tolerate from children, who dare not articulate any of what they plainly perceive. For children, the blurring of self/other thinking, the twisting de-contextualization of their words amidst dysphoric volatility, outbursts of rage, and retaliatory attacks (and the denial of the all of the above), draw me to Klein’s atmospheres: a launchpad (and not an endpoint) of investigation, as in my treatment of Kevin, described next time.

###

ENDNOTES

**I should also mention that I am seeing these states (individual or intersubjective) as being so raw and primitive as to often resist or be tone-deaf to transference interpretations. Analyst Richard Tuch states in his 2011 paper that addressing transference is futile with concretely-organized patients (often seen in addiction, even in sobriety, as we shall see in the coming vignette): a deeply unfamiliar language game.

***I discussed the psychic imprint of the mirror-role of the mother (Winnicott, 1971) in chapter 4 of my book on addiction and psychoanalysis.