Melanie Klein, “The Importance of Symbol-Formation in the Development of the Ego” (1930) (VIII)
From Anxiety to Defense (III)
In the last entries, I tried to clarify some of the general connections between aggression, anxiety and defense, emphasizing especially those qualities of the infant’s mental functioning that may be called “primitive.” During the sadistic phase that preoccupies Klein in “Symbol-Formation,” both (a) the defense and (b) the impulse upon which this defense operates are “violent.” By contrast, neither the later defense of repression, nor the libidinal impulse against which it operates, really qualify for this descriptor.
In this entry, I’d like to examine more concretely the particular defense Klein introduces in the opening sections of this essay. There are two aspects to this “earliest defense,” namely, “expulsion” and “destruction”:
“This defense, in conformity with the degree of the sadism, is of a violent character and differs fundamentally from the later mechanism of repression. In relation to the subject's own sadism the defense implies expulsion, whereas in relation to the object it implies destruction.” (220, my italics)
Klein’s way of phrasing things here (“This defense…”) suggests that expulsion and destruction are, not two discrete defenses, but rather two ways of conceiving a single defense. So while I will consider these two items in sequence, we should keep their inseparability in mind. I will return to this issue presently.
Expulsion
My first hunch is that “expulsion” here must mean something like projection or ascription. (The infant imagines: ‘This sadism — this drive to attack and persecute — belongs, not to me, but to the object.’) Moreover, in "Notes on Some Schizoid Mechanisms,” Klein describes this early defensive or “deflecting” strategy explicitly as one of “projection”: “The destructive impulse is partly projected outwards (deflection of the death instinct) and, I think, attaches itself to the first external object, the mother’s breast” (4-5). It would therefore seem unlikely that expulsion should additionally mean simple discharge — for surely the latter signifies, not a defense against one’s sadism, but a fairly unmediated expression of it.
Yet this might be precisely what Klein has in mind, which introduces a conceptual difficulty that is worth contemplating. Under what conditions is the infant’s effort to discharge its sadism — to phantastically attack the object of frustration — simultaneously a defense against that very sadism?
It seems that expulsion can denote the simple discharge of sadism, but only if by “sadism” Klein understands an item felt by the infant as in the first instance turned against itself — such that venting it upon the object is itself a defensive alternative for the infant. Yet this means that all those behaviors which a layperson might perceive as self-evident expressions of sadism are for Klein, paradoxically, ways for the infant to manage and defend against its internally self-directed sadism.
In one respect, of course, this conclusion should not surprise Klein’s readers. She certainly will depict the death drive along these lines. In Envy and Gratitude, for instance, she writes that the infant’s “primordial anxiety” is “engendered by the threat of the death instinct within” (215-16). And again later: “destructive impulses, the expression of the death instinct, are first of all felt to be directed against the ego” (223).
Nevertheless, the following considerations make this claim counter-intuitive:
In “Symbol-Formation” itself, Klein does not claim explicitly that the infant’s aggression is in the first instance self-directed, hence in need of defensive discharge outwardly. Only in other texts is this more clear.
Indeed, given the opportunity in the passage under review to explain why “the subject's own sadism” is felt by the infant as “a source of danger,” Klein simply introduces the (tautological-sounding) suggestion that this sadism “offers an occasion for the liberation of anxiety.” (This was the position I took in a previous entry, in which I considered these lines in detail.)
Surely the language of “sadism” suggests to our minds something inherently other- or object-directed. When self-directed, we’d naturally prefer to call this aggression — derivative of the death drive — masochistic.
Destruction
The second side of this early defense — or, perhaps more precisely, the second way of viewing this defense — is designated destruction. Together with our last considerations, this seems like conclusive evidence that expulsion does mean (in addition to ascription) something like discharge. For what is it, after all, to discharge one’s inner aggression (sadism) upon an object? For Klein, it is to shape this drive into phantasies of attacking and destroying that object. Hence, to put a point on our interpretation, “expulsion” (of the sadistic impulse) and “destruction” (of the object) are simply two names for one and the same action.
I have already written at some length about these destructive phantasies, and won’t spend much more time on them here. Indeed, the better part of Klein’s descriptions in the essay’s opening paragraphs elucidate and illustrate what she means.
However, I would like to emphasize— lest our emphasis on the action of “discharge” eclipses our initial reaction to Klein’s words — that “ascription,” too, must play a significant role in the infant’s expulsion of aggressive impulses (hence the destructive phantasy that this expulsion entails). What I mean is this: it appears to belong to Klein’s view of this sadistic phase that the infant can discharge and vent its aggression upon the object, phantasically attack and annihilate it — thereby managing its anxiety — only on the condition of ascribing this aggression to the object. The infant attacks the object because, from its own standpoint, the object is attacking it.
Of course, this defense is ultimately self-sabotaging, for it ensnares the infant in a vicious spiral of an “expulsive destruction” that promotes anxiety, which itself promotes more aggression, and so on. (This is not a critique of Klein; it is an anticipation of her own argument.) But for the moment this may suffice as a clarification of the “earliest defense” Klein attributes to the infant.
In the next entry, I will begin to discuss Klein’s understanding of symbolism, in light of the sadistic phase we’ve just reconstructed.
Melanie Klein, “The Importance of Symbol-Formation in the Development of the Ego” (1930) (VII)
From Anxiety to Defense (II)
After considering the paths by which the infant’s aggression causes it anxiety, we began in the last entry to examine the mechanisms — the “earliest defense” — by means of which the infant can diminish it. I made several observations regarding Klein’s description:
This defense, while having essentially to do with the infant’s intolerable anxiety, only works directly upon its aggression. Evidently the infant instinctively construes this aggression — discerned both in itself and (ascriptively) in its “persecuting” object — as the cause of its anxiety.
This early defense is “violent,” a predicate that matches the “violence” of the aggression or “excess of sadism” it confronts. Indeed, the defense is perhaps itself a manifestation of this same sadism. I would add that this sort of reflexive turn is familiar from Freuds structural model, which likewise involves an agency, the super-ego, split off from the id yet channeling id-aggression and, what is more, cruelly punishing the self for harboring that self-same aggression.
Klein is knowingly filling in a blank spot left behind — just as knowingly — by Freud. She quotes from the latter’s Inhibitions, Symptoms and Anxiety (1926) to this effect: “The excess of sadism gives rise to anxiety and sets in motion the ego's earliest modes of defense. Freud writes: ‘It may well be that, before ego and id have become sharply differentiated and before a super-ego has been developed, the mental apparatus employs different modes of defense from those which it practices after these levels of organization have been reached’” (220). Yet we might like to enter a caveat: though Freud indeed hypothesizes the existence of primitive defenses that precede “repression” — so opening the door to Klein’s “earliest defense” — he does not suggest here that such defenses would address aggression, rather than libidinal impulses.
In summary, then: both the object of defense and the defense itself have “primitive” qualities that distinguish Klein’s sadistic phase from subsequent phases.
So, on the one hand, Klein tells us that the direct or immediate object of defense is at the earliest stages, not libido, the life-drive, but the death drive, aggression. On the other hand, Klein insists that this more fundamental object of defense demands and activates more fundamental, primitive means of defense.
Excursus on Anxiety
I want to reiterate at greater length something I have already mentioned a number of times: the underlying reason for defense is, not aggression in itself, but rather the anxiety this aggression stirs up. Occasionally, Klein’s emphasis on aggression obscures this fact that it is not aggression per se which causes trouble to the infant, but only aggression as an engine of anxiety. The latter is clearer in “Notes on Some Schizoid Mechanisms” (1946), where Klein writes:
“We are, I think, justified in assuming that some of the functions which we know from the later ego are there at the beginning. Prominent amongst these functions is that of dealing with anxiety. The vital need to deal with anxiety forces the early ego to develop fundamental mechanisms and defenses.” (4-5)
Even more so, Envy and Gratitude (1957) contains what is presumably Klein’s considered statement on this question. Indeed, in the following passage, the ego is approximately synonymous with its anxiety-reducing function, hence verging on something like Harry Stack Sullivan’s “self-system”:
“[F]rom the very beginning, anxieties cannot be encountered without the defenses against them…[T]he first and foremost function of the ego is to deal with anxiety. I even think it is likely that the primordial anxiety, engendered by the threat of the death instinct within, might be the explanation why the ego is brought into activity from birth onwards. The ego is constantly protecting itself against the pain and tension to which anxiety gives rise, and therefore makes use of defenses from the beginning of post-natal life” (215-16)
Again, though, at this earliest stage the infant’s defenses directly manage internal aggression as the major source of its anxiety. Only later in its development must the infant similarly master its own libido — particularly by means of repression — and once again owing to the anxiety this libido now excites.
Twenty years after “Symbol-Formation,” Klein cites the same conjecture of Freud’s quoted earlier, once more in Envy and Gratitude:
“In the earliest stages splitting and other defense mechanisms are always paramount. Already, in Inhibition, Symptoms and Anxiety, Freud had suggested that there may be methods of defense earlier than repression.” (231)
Again Klein suggests that only later on, and in conjunction with the anxiety surrounding libidinal impulses, does the infant make use of the relatively sophisticated defense of “repression” — and only if that infant is fortunate, and its development has not been derailed by an “excessive” reliance on more primitive defenses. Hence what in Freud generally appears as the ur-defense, repression, upon whose foundation others are subsequently established, is for Klein a relatively late marker of psychological maturity. (I should probably soften this contrast somewhat. For not only, as Klein reminds us, does the Freud of Inhibitions clear space for such a hypothetical possibility. My understanding is that in his last works — “Fetishism,” the unpublished “Splitting of the Ego in the Process of Defense” — Freud puts names to such defenses: “disavowal,” “denial.”)
In any event, the passages quoted in this entry underscore our view, to belabor the point a moment longer, that for Klein it is from anxiety that the infant essentially requires protection. If the defense(s) Klein itemizes — expulsion, destruction, splitting, denial, omnipotence — all lay hold of impulses (initially aggressive impulses, later on libidinal ones), nevertheless this action is intelligible only as part of an overarching strategy of anxiety-reduction. Not the impulse, then, but the impulse’s byproduct, anxiety, puts the ego to work.
Melanie Klein, “The Importance of Symbol-Formation in the Development of the Ego” (1930) (VI)
From Anxiety to Defense (I)
In the last entries we attempted to clarify the paths which for Klein lead from the infant’s aggression to its experiences of anxiety. We can now return to the passage we’ve been examining and consider those lines that concern defense. The passage, again:
“The excess of sadism gives rise to anxiety and sets in motion the ego's earliest modes of defense. Freud writes: ‘It may well be that, before ego and id have become sharply differentiated and before a super-ego has been developed, the mental apparatus employs different modes of defense from those which it practices after these levels of organization have been reached.’ According to what I have found in analysis the earliest defense set up by the ego has reference to two sources of danger: the subject's own sadism and the object which is attacked. This defense, in conformity with the degree of the sadism, is of a violent character and differs fundamentally from the later mechanism of repression. In relation to the subject's own sadism the defense implies expulsion, whereas in relation to the object it implies destruction” (220)
I would like in this entry to make a few observations.
First, we will notice a feature of this “earliest defense set up by the ego” that, while obvious, doesn’t receive any explicit comment from Klein. At their core, this defense helps the infant manage anxiety — the infant’s aggression is distressing to it only because of the anxiety it occasions. (If per impossible the infant could feel and express its sadism without incurring anxiety, it would presumably have no need to defend against it.) It is noteworthy, then, that while this early defense, and defenses more generally, have essentially to do with diminishing anxiety, they do not manage or influence this anxiety directly. Instead, according to Klein’s account, this early defense directly addresses either aggression or the object of aggression, and precisely because both of these stir anxiety. Again: “the earliest defense set up by the ego has reference to two sources of danger: the subject's own sadism and the object which is attacked.” (By contrast, one imagines that a mechanism like dissociation might directly defend against anxiety, by simply expelling the latter from awareness. Such a defense, though, does not appear in Klein’s thinking — at least in this essay.)
A second observation: the defense is “violent”: “This defense, in conformity with the degree of the sadism, is of a violent character…In relation to the subject's own sadism the defense implies expulsion, whereas in relation to the object it implies destruction.” With the descriptor “violent,” Klein seems to have in mind, not only the considerable strength, crudeness, or efficacy of this defense, but beyond this some unique quality in the defense itself. This again suggests something self-evident about Klein’s position that does not receive special attention, namely: this primitive defense does not merely guard against aggression; it is itself a mode or expression of this aggression. To put this another way: “expulsion” and “destruction” are both essentially “violent” ways of handling one’s own “violence” — including that portion of the infant’s violence it ascribes to its object.
A third observation, perhaps less self-evident, follows from these reflections. Klein will suggest it again, more clearly, in one of the essay’s concluding statements:
"It is only in the later stages of the Oedipus conflict that the defense against the libidinal impulses makes its appearance; in the earlier stages it is against the accompanying destructive impulses that the defense is directed. The earliest defense set up by the ego is directed against the subject's own sadism and the object attacked, both of these being regarded as sources of danger. This defense is of a violent character, different from the mechanism of repression” (232)
Now it is worth recalling that when Freud first addressed himself to the question of mental defense, he had not even solidified his first concept and classification of drives, let alone that second concept and classification to which he was later impelled (one embracing the life- and death-drives). Indeed, the earlier definition of drive (i.e. inner excitations pressing for discharge) and classification (self-preserving drives, like hunger, over against libidinal drives), left no room for such a primary drive as Thanatos, whose “end” is dissolution and which is manifested psychologically as aggression.
Hence within that early Freudian context, at least, Klein’s conjecture that the first “defenses” put into operation are directed against the infant’s aggression, and only later against libidinal impulses, would have made no sense at all. For at that point in Freud’s development, not only were libidinal drives sui generis in demanding defense, and particularly the defense of repression. (As far as I can tell, the “self-preserving” drives lodged no such demand.) There is no death drive at all, the manifestations of which could precede “libidinal impulse.”
In the next entry, I will continue to examine the connection between anxiety and defense in Klein’s thinking.
Melanie Klein, “The Importance of Symbol-Formation in the Development of the Ego” (1930) (V)
From Sadism to Anxiety (II)
We have been considering the connection, for Klein, between aggression and anxiety. In particular, we have attempted to grasp her claim that the infant’s inborn aggression necessarily generates anxiety in it.
As we have seen, this aggression — more specifically an “excess of sadism” — causes the infant anxiety in several ways. On the one hand — [1] — the infant (putatively) experiences its own sadism as dangerous, and in two respects:
[1a] As I interpreted Klein, the infant is made anxious by its own aggressive impulses once these are linked in its mind to an “object” that will retaliate for that aggression.
[1b] The infant experiences the instruments or “weapons” of its sadism — its bodily excreta — as dangerous. Perhaps they are liable to be mishandled and thus injure their handler. Or perhaps, beyond this, the retaliatory “object” may appropriate these weapons and redirect them against the sadistic infant.
Yet on the other hand, as I indicated in the last entry, it is the second “source of danger” or catalyst for anxiety — [2] — that is more immediately comprehensible and, beyond this, more closely associated with Kleinian thinking in general. This is the persecutory object and anxiety that suffuse the “paranoid schizoid position,” a term Klein hadn’t yet introduced when “Symbol-Formation” was published in 1930. In “Notes on Some Schizoid Mechanisms” (1946), though, this is plainly the central, if not the sole “form” that anxiety assumes for the infant, regardless of its original catalyst. In this later piece, Klein writes:
“I hold that anxiety arises from the operation of the death instinct within the organism, is felt as fear of annihilation (death) and takes the form of fear of persecution. The fear of the destructive impulse seems to attach itself at once to an object — or rather it is experienced as the fear of an uncontrollable overpowering object. Other important sources of primary anxiety are the trauma of birth (separation anxiety) and frustration of bodily needs; and these experiences too are from the beginning felt as being caused by objects. Even if these objects are felt to be external, they become through introjection internal persecutors and thus reinforce the fear of the destructive impulse within.” (4-5)
It appears that, for Klein, at least initially, nothing (negative) emerges “internally” for the infant that isn’t perforce ejected outward and, moreover, isn’t attributed as a quality (property, motivation, attitude, action) to that primary “object” through which all its experience is invariably filtered. In fact, every one of the infant’s feelings — positive or negative — bears some necessary “reference” to this object and is “intelligible” to the infant only in terms of the latter. This, as I understand it, is one leitmotif of the object-relations tradition — perhaps its core — and it is reflected in Klein’s words from Envy and Gratitude: “From the beginning, all emotions attach themselves to the first object” (234).
Moreover, in the passage above, Klein effectively distinguishes between the source, the object, and the outcome of anxiety. (Freud famously distinguishes the different aspects of a “drive.”) So, while the major source of anxiety, and certainly the one that preoccupies her in these essays, is the infant’s own, endogenous death drive, this does not exclude “other important sources of primary anxiety.” Yet no matter the source, which is variable, Klein’s words here suggest that both the “object” and the “outcome” are unitary: the catastrophe this anxiety ultimately intimates is “annihilation (death),” a fate ascribed to “an uncontrollable overpowering object.” Whatever its source, the infant’s anxiety is experienced as “fear of persecution.” So, while Klein does recognize distinct sources of anxiety that are not reducible to the “operation of the death instinct” — for instance, “the trauma of birth (separation anxiety) and the frustration of bodily needs” — these mental states, too, are “felt as being caused by objects,” that is, are experienced in the final analysis as persecutory.
For the purposes of our commentary on “Symbol-Formation,” this digression provides an interesting caveat to our account. While the death drive suffices all on its own to cause the infant’s anxiety (through the channels we are now examining), it might not be necessary —at least in the first instance — for doing so. Anxiety is also brought about from other quarters (separation, frustration), even if these, too, are immediately drawn into the experiential vortex of aggression and persecution by an “object.” But this means that, even where the death drive is not the proximate source of anxiety, it nevertheless inexorably overlays, contours, and defines that anxiety in its characteristic way.
But there is a complication in this account, which I’d like briefly to discuss. As I’ve suggested, Klein appears to conceive the infant as an instinctive anthropomorphizer. Particularly with regard to its aggression — though applying more broadly to all feeling-states — the infant can experience feelings only in connection with its object, where this latter is both (a) the necessary recipient of that feeling (“I am angry at the mother-breast for withholding its feeding from me”) and also (b) as the bearer and agent of that feeling, via projection or ascription (“the mother-breast is angry at me, is attacking and persecuting me”). Both of these, I think, are contained in the line from Envy and Gratitude quoted above: “From the beginning, all emotions attach themselves to the first object.”
Now in the last entry I observed a certain “intermingling” in Klein’s account of aggression, anxiety, and defense — and this conception seems to be a case in point. For while the rough order of explanation leads from (a) aggression, to (b) the anxiety it stimulates, to (c) the defenses required to manage that anxiety, these comments about projection scramble the sequence. The infant has no experience of either aggression or anxiety that is unmediated by its “object.” Yet the entirety of the infant’s contact with this object seems premised on the mechanism of projection — the “violent defense” which we suppose only emerges in response to aggression-induced anxiety. Again: “The fear of the destructive impulse seems to attach itself at once to an object — or rather it is experienced as the fear of an uncontrollable overpowering object.” In other words, the defense of projection antecedes any experience of that very anxiety which (ostensibly) calls it into action.
Melanie Klein, “The Importance of Symbol-Formation in the Development of the Ego” (1930) (IV)
From Sadism to Anxiety (I)
What exactly is the link between the infant’s aggression and its anxiety? I suggested in the last entry that, for Klein, this link is neither arbitrary nor avoidable. In every imaginable scenario, it seems, the infant’s inborn aggression will generate some, and probably a great deal of anxiety.
Why does Klein take this position?
Let us return to the passage quoted in the last entries, this time excising Klein’s ideas about “defense” and preserving only those lines that concern the connection between sadism and anxiety:
“The excess of sadism gives rise to anxiety…According to what I have found in analysis…[for the infant there are] two sources of danger: [1] the subject's own sadism and [2] the object which is attacked…The sadism becomes a source of danger because it [1a] offers an occasion for the liberation of anxiety and also because [1b] the weapons employed to destroy the object are felt by the subject to be leveled at his own person also. [2] The object of the attack becomes a source of danger because the subject fears similar, retaliatory attacks from it” (220, bracketed numbers mine)
I have interpolated numbers that help us to clarify Klein’s view and its (asymmetrical) subdivisions. More precisely, Klein tells us here that the infant’s sadism elicits anxiety through two major routes, “two sources of danger,” [1] and [2]. Notably, one of these — [1] — itself resolves into two types, [1a] and [1b].
To take these in sequence: one generic “source of danger” is [1] “the subject's own sadism.” (This is, we might say, the way in which sadism causes anxiety directly, rather than indirectly.) And it does this, again, in two distinct ways:
“The sadism becomes a source of danger because it [1a] offers an occasion for the liberation of anxiety and also because [1b] the weapons employed to destroy the object are felt by the subject to be leveled at his own person also.”
What exactly does Klein have in mind with each of these? Unfortunately, as far as I can see, Klein’s assertion in [1a] — “sadism becomes a source of danger because it offers an occasion for the liberation of anxiety” — is either tautological, or else implicitly dependent on that “indirect” danger represented in [2].
On the one hand, Klein’s words at least have the sound of tautology, because “the liberation of anxiety” is surely synonymous with the apprehension of “danger.” Interpreted in this (uncharitable) way, Klein’s statement that sadism is “a source of danger” because it “offers an occasion for the liberation of anxiety” is rather like saying: sadism is a source of danger because it is a source of danger. But our puzzle is precisely why the infant experiences its own aggression as “dangerous,” or why it elicits “anxiety” at all.
On the other hand: the more promising interpretation of [1a] is that this claim ultimately depends on [2]. To briefly look ahead, Klein will momentarily tell us that, “The object of the attack becomes a source of danger because the subject fears similar, retaliatory attacks from it.” In other words, the infant’s sadism gives rise to persecutory anxiety. My more charitable reading of [1a], then, is something like: once the “object” is established as a retaliatory persecutor, then the infant’s own aggression will cause the infant anxiety. For at this point, as Klein puts it, sadism indeed “offers an occasion for the liberation of anxiety” — is experienced by the infant as dangerous — because it believes this sadism will incur a reciprocal attack by the “object” toward which it is directed.
In fact, I submit that only [1b] offers a plausible basis for sadism’s production of anxiety in the infant, though it has a speculative flavor to it. Again: another reason that the “subject’s own sadism” appears to it as a “source of danger” is because “the weapons employed to destroy the object are felt by the subject to be leveled at his own person also.” Here we will observe that, strictly speaking, it is not the sadism itself that is immediately apprehended as dangerous, but rather those “weapons” which serve the infant’s sadistic ends, or which have been invested with sadistic significance. The weapons are those “excreta” named in the previous paragraph: specifically the infant’s “wetting” and “faecal mass.” (The particular value of these objects as weapons — whether as something used to “stab” and “drown,” or as a “missile” or “poisonous substance” — is fixed by the infant’s psychosexual stage of development.)
Yet even here it may seem like Klein’s “explanation” begs the major question. I at least would like to know why these weapons “are felt by the subject to be leveled at his own self as well.” Let us assume that the infant does experience its excreta in the way Klein describes, that is, as dangerous. Must an infant experience any weapon — even one over which it has control, and one directed outwardly — as dangerous to itself? Or is the danger, hence the anxiety, a product of the infant’s sense that these weapons are not entirely under its control, that they might be mishandled and “backfire” from this mishandling? Indeed: does the infant have some dim apprehension, even at this early stage, that the object will wrest these weapons away and turn them against him or her?
These last reflections open the door, once again, to the suspicion that proposition [1] — this time [1b] — implicitly depends upon [2]. I will continue with this idea in the next entry.