8

Treatment of Borderline States

The borderline concept has, in recent years, achieved enormous popu-

larity within psychoanalytic and psychotherapeutic circles. Despite
this rise to stardom, vast differences of opinion and numerous un-
resolved questions continue to exist concerning just what, if anything,
the term "borderline" describes. We shall not attempt to cover the vo-
luminous literature on this subject here (see Sugarman and Lerner,
1980, for an excellent review). Instead, we offer a critique of the cur-
rently prevalent view that the term "borderline" refers to a discrete
pathological character structure, rooted in specific pathognomonic in-
stinctual conflicts and primitive defenses. An alternative understand-
ing of borderline phenomena emerges when they are viewed from an
intersubjective perspective. Our focus will be on the intersubjective
contexts in which borderline symptomatology takes form, both in
early development and in the psychoanalytic situation.
The term "borderline" is generally used to refer to a distinct charac-
ter structure that predisposes to faulty object relations, in which the
fundamental difficulties are ordinarily attributed to the patient's path-
ological ego functioning. Typically the borderline personality organi-
zation is pictured as a direct structural consequence of the patient's use
of certain primitive defenses-splitting, projective identification, ideal-
ization, and grandiosity- to ward off intense conflicts over depend-
ency and excessive pregenital aggression (which dependency presum-
ably mobilizes). But what is the clinical evidence that supposedly
demonstrates the operation of these primitive defenses? And what is
the meaning of the excessive aggression to which primary etiological
significance is ascribed in the genesis of borderline psychopathology?

106
 TREATMENT OF BORDERLINE STATES 107

THE QUESTION OF SPLITTING

The experience of external objects as "all,good" or "all,bad" is gener,

ally regarded as a clear manifestation of splitting, resulting in sudden
and total reversals of feeling whereby the view of the object is shifted
from one extreme to the other. Oscillation between extreme and con,
tradictory self concepts is similarly seen as evidence of splitting. This
fluid and rapid alternation of contradictory perceptions of the self or
others is seen as the result of an active defensive process whereby im,
ages with opposing affective valences are forcibly kept apart in order to
prevent intense ambivalence. But is this assumption warranted clinic,
ally? Splitting as a defense actively employed to ward off ambivalence
conflicts can come into play only after a minimum of integration of dis,
crepant self and object experiences has been achieved through devel,
opment (Stolorow and Lachmann, 1980). A defensive split into parts
presupposes a prior integration of a whole. It is our contention that
such a presupposition is not warranted when treating patients who are
ordinarily diagnosed ''borderline." Their fragmentary perceptions do
not result primarily from defensive activity, but rather from an arrest
in development, which impairs their ability reliably to synthesize affec,
tively discrepant experiences of self and other. Their rapidly fluctuat,
ing views of the therapist, for example, do not primarily serve to pre,
vent ambivalence toward him. They are, in part, manifestations of a
need for the therapist to serve as an archaic containing or holding ob,
ject whose consistently empathic comprehension and acceptance of
these patients' contradictory affective states function as a facilitating
medium through which their varying perceptions and feelings can
eventually become better integrated (Winnicott, 1965; Modell, 1976;
Stolorow and Lachmann, 1980).
It is our view that the lack of synthesis of self and object experiences
characteristic of so,called borderline states is neither defensive in na,
ture nor central in the genesis of these disorders. In our experience, the
intense, contradictory affective states that these patients experience
within the transference, and in particular their violent negative reac,
tions, are indicative of specific structural weaknesses and vulnerabil,
ities rooted in specific developmental interferences. Archaic mir,
roring, idealizing, and other selfobject needs are revived in analytic
108 CHAPTER EIGHT

transferences, together with hopes for a resumption of development.

When these needs are responded to, or understood and interpreted
empathically, intense positive reactions occur. Similarly, when these
needs are not recognized, responded to, or interpreted empathically,
violent negative reactions may ensue. If these angry reactions are pre,
sumed to represent a defensive dissociation of good and bad aspects of
objects, this in effect constitutes a covert demand that the patient ig,
nore his own subjective experiences and appreciate the "goodness" of
the analyst and his interpretations. It precludes analysis of the patient's
subjective experience in depth, the elements that go to make it up, and
their special hierarchy of meanings for the patient. In contrast, when
we have held such preconceptions in abeyance, we have found that the
intensity of the angry reactions stems from the way they encoded and
encapsulated memories of specific traumatic childhood experiences.

The Case of Jeff

A clinical vignette illustrates our idea of a specific vulnerability. When

Jeff, a young man of23, entered treatment, he was in a state of marked
overstimulation. He could not sit still for more than a few minutes at a
time; his eyes darted from object to object; and he spoke under con,
stant pressure. Although enrolled in college, he had not been able to
attend classes or concentrate on his work. Increasingly frightened
when alone at night, he had recently begun to take to the streets.
There he had been approached for homosexual purposes several times,
and this made him more fearful of his own unrecognized wishes and
heightened his agitation. In the sessions he gave the impression of
wanting desperately to cling to something around which he might be,
gin to reorganize and restructure himself. Consequently, during the
first months of treatment it was very difficult to bring any session to a
close. His initial resistances centered on fears of being used to fulfill the
analyst's needs. When these were interpreted, an early idealizing trans,
ference developed. This enabled Jeff to confront the area of primary
defect-a failure to have attained a cohesive self and a vulnerability to
recurrent states of protracted disorganization. The analysis thus re,
sumed a developmental process that had been stalled.
Jeff's relationship with his father had always presented difficulty for
him. The father reacted to any weakness or shortcoming in his son
 TREATMENT OF BORDERLINE STATES 109

with impatience and contempt. This situation directly entered the

analysis because Jeff's father had assumed financial responsibility for
the treatment. The arrangement became a source of greater and
greater tension between the two, for the father resented the burden of
payment, as well as what he saw as evidence of his son's weakness and
simultaneously a source of shame for himself. The difficulties in this
area increased whenever Jeff made it clear that the analysis was not
leading in the direction of making Jeff the son his father had always
wished for, but was instead increasing Jeff's determination to develop
in his own way.
Although the analyst realized the complications that might ensue,
after two and a half years he notified Jeff that he was raising his fees
generally. He wanted to discuss the matter with Jeff to see if and how it
might be worked out. The request came at a time when Jeff's relations
with his father were already strained, though it did not appear likely
that this would change within any foreseeable period of time. Jeff's ini-
tial response was one of some anger about the unfortunate timing, fol-
lowed by a remark to the effect that of course he knew how the analyst
felt because everything was going up in price. Recognizing Jeff's fre-
quent tendency to substitute an understanding of someone else's posi-
tion for an expression of his own, the analyst interpreted this, together
with Jeff's fear of the analyst's reaction to his expressing his own feeling.
(We would emphasize that in our experience such genuine emotional
expression is always obstructed, and with it an essential aspect of an
authentic relationship, when a patient's affective states are incorrectly
interpreted as defensive transference distortions.)
Gradually, over the course of the next few sessions, Jeff was able to
come out with his feelings-feelings of hurt, disappointment, and
violent anger. The hurt seemed to center on the analyst's failure to ever
(Jeff's words) consider him first, and the extent to which this experi-
ence revived feelings of always having been a burden, a supplicant,
someone standing in the way of other people's plans or enjoyment. Jeff
was a twin, and he recounted a welter of experiences in which his twin
had preempted his parents' attention by being exactly the child they
wanted and one who caused them no difficulty.
Jeff's anger at the analyst was related mostly to the poor timing and
what that meant to him. He spoke of the bind the analyst's request put
him in. Things were already going badly between him and his father.
110 CHAPTER EIGHT

Jeff had started a new job and had been forced to ask his father for
money for new clothes. Each encounter of that kind was humiliating
for Jeff. Now he would have to face a review of how long he had been in
treatment and how much longer it was to continue. How could the a n ~
alyst, knowing all this, choose to put Jeff through it!
Frequently, after expressing himself unabashedly, Jeff would huddle
up, as if in a corner, his arms protectively wrapped around himself. In
response to questions, he confirmed that he was terrified. He was cer~
tain that the analyst would be furious with him, call him selfish, and
berate him for his lack of appreciation for the analyst.
There now emerged a host of memories in which the timing of Jeffs
life (and, indeed, his life itself) had to conform to someone else's wishes.
He had to go to bed when his father told his mother he should. He had
to wait until his father was done with the evening news before speaking
to him, and then he could only talk about what his father was inter~
ested in. Monday night, football night, was especially sacrosanct- not
an occasion when a pleasurable interest might be shared, but one more
occasion when Dad was not to be disturbed.
Jeff's mother told him when, what, and how to eat. She chose his
clothes for him, where and how he was to sit or stand. He was not to sit
on the couch lest the cushions be messed up, nor on his bed for similar
reasons. He had to renounce his own inclinations and adopt her
wishes regarding what music he was to like. Always before the family
left on an auto trip, he was instructed to urinate, and his mother
checked to make sure he didn't put anything over on them. Otherwise
they might have to stop along the way. And Jeff recalled that whenever
he attempted to protest or assert himself, perhaps because something
was especially important to him, he was squelched, accused of selfish~
n e s ~ and a lack of consideration. He was told that his father wouldn't
warlt to come home at all if he kept this up.
For Jeff, the most significant aspect of these repeated experiences
was a feeling of absolute powerlessness. Once, when he could not stand
it anymore, he went to his room and packed an overnight bag. When
he appeared in front of his parents to declare he was running away, no
one said a word or made a move to stop him. He then realized that he
was stuck- that no one else would want him and that he had to give in.
These experiences formed the background of Jeff's reaction to the
analyst's request for an increase. Jeff retained, in its most imperative
form, the longing that someone would put his wishes first, and he was
 TREATMENT OF BORDERLINE STATES 111

highly sensitive to the specific configuration of others' needs being put

before his own. He therefore responded acutely and intensely to that
configuration when it entered the transference. This response was cov*
ered over by a more moderate reaction, in which he apparently at*
tempted defensively to "synthesize" good and bad object concepts.
What was crucial, however, was for Jeff to recognize the underlying in*
tensity of his hurt and the experiences behind it, rather than having
his reaction regarded as an instance of splitting or a lack of apprecia*
tion for the analyst. This recognition opened up an entire area of the
transference to analysis and ultimate resolution. Jeff and the analyst
came to see clearly the extent to which Jeff had found it necessary to
define himself around what was expected, what would please, and
what would not offend in order to maintain his object ties. They were
able to comprehend the threat constantly posed by any authentic ex*
perience of self-the threat of estrangement and isolation Jeff had en*
countered whenever he asserted himself or attempted to act on his
own behalf. The analysis, then, brought out into the open and allowed
Jeff to work through the enormous resentment such subjugation of self
had aroused.

THE QUESTION OF PROJECTIVE IDENTIFICATION

Considerations similar to those we have discussed for splitting apply

to the view of projective identification as a primitive defense, charac*
teristic of borderline patients. In projective identification there is a
blurring of the distinction between the self and the object in the area of
the projected content. Such states of self*object confusion are pre*
sumed to be the product of an active defensive effort to externalize all*
bad, aggressive self* and object images. Once again, we question
whether this assumption is clinically justified.
Projection as a defense actively employed to ward off conflict can
come into play only after a minimum of self*object differentiation has
been reliably achieved (Stolorow and Lachmann, 1980). Defensive
translocation of mental content across self*object boundaries presup*
poses that those boundaries have been for the most part consolidated.
Our experience contradicts such a presupposition for patients diag*
nosed "borderline." Their states of self*object confusion arise primarily
from a developmentally determined inability to maintain the distinc*
112 CHAPTER EIGHT

tion between self and object. In the treatment context it is not useful to
view such states as examples of either defensive projection or general
ego weakness. Instead, these partially undifferentiated states are best
understood as manifestations of revivals with the therapist of a specific
need for immersion in a nexus of archaic relatedness, from within
which formerly thwarted developmental processes of self~articulation
and s e l f ~ d e m a r c a t i o n can be revitalized and once again resumed
(Stolorow and Lachmann, 1980).
Frequently we have encountered in the literature a second, and to
our minds even more questionable, use of the term "projective identifi~
cation." There is presumed to be not only a projective distortion of the
patient's subjective experience of the object, but also a purposefully in~
duced alteration in the external object's actual attitude and behavior
toward the patient. The patient is said to put split~off, disavowed parts
of himself inside the external object. This formulation is based on the
observation that intense reactions frequently occur in analysts who
are treating borderline patients. Because such reactions are experi~
enced similarly by most "reasonably w e l l ~ a d j u s t e d therapists," the rea~
soning goes, "countertransference reactions in these cases reflect the
patient's problems much more than any specific problems of the a n a ~
lyst's past" (Kernberg, 1975, p. 54). It is also suggested that if the a n a ~
lyst is reacting intensely to the patient, such countertransference is a
clue to the patient's hidden intention. Kernberg (1975), for example,
writes:

If the patient systematically rejects all the analyst's interpreta~

tions over a long period of time, the analyst may recognize his
own resultant feelings of impotence and point out to the patient
that he is treating the analyst as if he wished to make him feel de~
feated and impotent. Or when antisocial behavior in the patient
makes the analyst, rather than the patient, worry about the c o n ~
sequences, the analyst may point out that the patient seems to try
to let the analyst feel the concern over his behavior because the
patient himself cannot tolerate such a feeling [p. 247].

These formulations fail to take into account that when the analyst,
in his interpretations, insists that the patient's difficulties arise from vi~
cissitudes of aggressive~drive processing, the only alternatives open to
the patient are to agree with the premises being put forward or to find
 TREATMENT OF BORDERLINE STATES 113

himself in the position of inadvertently making the analyst feel defeated

and impotent. To us, this state of affairs seems to reflect the extent to
which the analyst's self-esteem depends on the patient's acceptance of
the correctness of his theoretical position, rather than necessarily re-
flecting any unconscious hostile intention on the part of the patient.
Similarly, the analyst's concerns about a patient's antisocial behavior
seem to us to reflect the analyst's difficulties in sufficiently demarcating
himself from the patient so as to be able to devote himself to the inves-
tigation of the meaning of the actions in question.
A description of a typical clinical application of the concept of
projective identification is contained in Kernberg's (1975) reference to
Ingmar Bergman's movie Persona:

A recent motion picture ... illustrates the breakdown of an im-

mature but basically decent young woman, a nurse, charged with
the care of a psychologically severely ill woman presenting what
we would describe as a typical narcissistic personality. In the face
of the cold, unscrupulous exploitation to which the young nurse
is subjected she gradually breaks down. She cannot face the fact
that the other sick woman returns only hatred for love and is
completely unable to acknowledge any loving or human feeling
toward her. The sick woman seems able to live only if and when
she can destroy what is valuable in other persons, although in the
process she ends up by destroying herself as a human being. In a
dramatic development the nurse develops an intense hatred for
the sick woman and mistreats her cruelly at one point. It is as if all
the hatred within the sick woman had been transferred into the
helpful one, destroying the helping person from the inside
[pp. 245-246].

We hold that conclusions such as this are unjustified and that the un-
derlying assumptions are unwarranted and antitherapeutic. In the first
place, there is no evidence that the sick woman is "able to live only if
and when she can destroy what is valuable in other persons"; there are
only indications that the sick woman does not respond in a way that
the nurse-therapist wants or needs. We are familiar in our own prac-
tices with many cases in which patients who have recently experienced
traumatic loss and disintegration resolutely protect themselves against
any involvement until some spontaneous recovery has set in. Second,
114 CHAPTER EIGHT

there is no evidence that "the hatred within the sick woman has been
transferred into the helpful one, destroying the helping person from
the inside." There is, instead, every indication that the patient's re~
sponsiveness was required in order for the nurse to maintain her own
self~esteem and to regulate her own psychological functioning. When
frustrated, the nurse demonstrated her own narcissistic vulnerability
and propensity for rage reactions. We have observed such factors at
work in ourselves and regard them as to some degree universal in ther~
apeutic relationships. In our view, their near universality does not w a r ~
rant their being ignored as originating in the personality structure of
the therapist. Nor does it warrant the assumption that these responses
are an indication of pathological projective mechanisms on the part of
the patient. We have found that the assumption that the patient
wishes the therapist to feel impotent or infuriated is much more often
than not directly contradicted in our own work. Such wishes, we sug~
gest, occur only when the patient's disagreements, assertions, and pri~
mary wishes to have his own subjective experiences empathically
understood have been consistently unresponded to. Far more often,
the patient's fear of the analyst's narcissistic vulnerability and of being
held responsible for the analyst's feeling of frustration constitutes a se~
vere resistance to free association and is a prominent motive for
defense.
The concept of projective identification is used extensively by a n a ~
lysts to explain any fear that is not readily intelligible as a response to a
real danger. It is consistently invoked to explain why patients are so
regularly afraid of their analysts. We have found, however, that the a n ~
alyst's insistence that negative reactions in analysis are to be explained
by the patient's innate aggression or envy, or by his projection of ag~
gressively distorted internal objects, can be damaging to the patient, to
the unfolding selfobject transference, and to the analysis (Brandchaft,
1983).
The application of the theory of projective identification carries
with it the real danger of depriving patients of a means of defending
themselves when they feel that the analyst is cruel, distant, control~
ling, or demeaning. This danger is increased if the analyst, for w h a t ~
ever reason, is unable or unwilling to become aware of his actual effect
on the patient, or if he minimizes that effect because of a conviction
that he has the ultimate best interests of the patient at heart. Fre~
quently, this conviction in the analyst takes the form of a conception
 TREATMENT OF BORDERLINE STATES 115

of a "more normal dependent" part of the patient, which is being d o m i ~

nated and excluded by the aggressive part. Such unwarranted, if
reassuring, concepts notwithstanding, the tendency to fall back on in~
terpretations of projection to the detriment of the subjective experi~
ence of the patient, even where such mechanisms exist, can in practice
be shown to foster a dependence on the analyst's perceptions at the ex~
pense of the patient's. These interpretations encourage, indeed re~
quire, a pro forma belief in the analyst's "goodness" and correctness at
the expense of the self. They impair the patient's sense of his own self
and belief in himself, and they encourage an agreement that necessary
and understandable efforts to protect a vulnerable self are indicative of
severe pathology and should be given up.

FURTHER MISCONCEPTUALIZATIONS

Closely allied with the developmental disturbances discussed so far

are the idealizations and grandiosity that often pervade the treatment
of patients who are called "borderline." Such perceptions of the self or
others are regularly interpreted as being defensive against dependency
and the attendant s u b j e c t ~ c e n t e r e d or o b j e c t ~ c e n t e r e d aggression. Our
experiences indicate that most often the idealizations and grandiosity
are manifestations of selfobject transferences (Kohut, 1971, 1977).
They are not pathological defenses, but rather revivals with the t h e r a ~
pist of the archaic idealizing and mirroring ties that were traumatically
and p h a s e ~ i n a p p r o p r i a t e l y ruptured during the patient's formative
years and on which he now comes to rely for the restoration and m a i n ~
tenance of his sense of self and for the resumption and completion of
his arrested psychological growth.
Having argued that much of the clinical evidence cited for the o p e r ~
ation of primitive defenses is actually evidence of needs for specific ar~
chaic selfobject ties, and of disturbances in those ties, how shall we
understand the "excessive pregenital aggression" that many authors
believe is the etiological bedrock of borderline pathology? We contend
that pervasive primitive aggression is an inevitable, unwitting,
iatrogenic consequence of a therapeutic approach that presupposes
that the psychological configurations we have been discussing are in
their essence pathological defenses against dependency and primitive
aggression. A patient revives an arrested archaic state or need, or at~
116 CHAPTER EIGHT

tempts a previously aborted developmental step within the therapeutic

relationship, and the therapist interprets this developmental necessity
as if it were a pathological defense. The patient then experiences this
misinterpretation as a gross failure of attunement, a severe breach of
trust, a traumatic narcissistic wound (Stolorow and Lachmann, 1980).
When vital developmental requirements reexperienced in relation to
the therapist once again meet with traumatically unempathic re,
sponses, is it surprising that such misunderstandings often bring in,
tense rage and destructiveness in their wake? We are contending, in
other words, that the pervasive aggression is not etiological, but rather
a secondary reaction to the therapist's inability to comprehend the de,
velopmental meaning of the patient's archaic states and of the archaic
bond that the patient needs to establish with him (Kohut, 1972, 1977;
Stolorow, 1984a).

AN INTERSUBJECTIVE VIEWPOINT

At this point we are in a position to formulate our central thesis re,

garding the borderline concept. The psychological essence of what we
call "borderline" is not that it is a pathological condition located solely
in the patient. Rather, it refers to phenomena arising in an intersubjective
field -a field consisting of a precarious, vulnerable self in a failing, archaic
selfobject bond. In order to elaborate this thesis further, we must clarify
the nature of the self disorder that contributes to the emergence ofbor,
derline phenomena.
We view the various disorders of the self as arbitrary points along a
continuum (see Adler, 1981) rather than as discrete diagnostic entities.
The points along this continuum are defined by the degree of impair,
ment and vulnerability of the sense of self, the acuteness of the threat
of its disintegration, and the motivational urgency of self,reparative ef,
forts in various pathological states. The degree of severity of self disor,
der may be evaluated with reference to three essential features of the
sense of self- its structural cohesion, temporal stability, and affective
coloration (Stolorow and Lachmann, 1980).
In certain patients, the sense of self is negatively colored (feelings of
low self,esteem) but is for the most part temporally stable and struc,
turally cohesive. One might refer to such cases as mild self disorders. In
other patients, the sense of self is negatively colored and its organiza,
 TREATMENT OF BORDERLINE STATES 117

tion is temporally unstable (experiences of identity confusion) but,

notwithstanding fleeting fragmentations, it largely retains its struc-
tural cohesion. These cases might be called moderately severe self dis-
orders. In a third group of patients, the sense of self is negatively
colored, temporally unstable, and lacking in cohesion and thus subject
to protracted structural fragmentation and disintegration. Such cases
can be termed very severe self disorders. Roughly speaking, patients
who are called "borderline" fall within the moderate to severe range of
self disorders.
Our concept of self disorder as a continuum or dimension of psycho-
pathology is somewhat at variance with Kohut's (1971) early view of
"borderline" as a discrete diagnostic entity that is sharply distinguisha-
ble from the narcissistic personality disorders. The borderline person-
ality, according to this view, is chronically threatened with the
possibility of an irreversible disintegration of the self-a psychological
catastrophe that is more or less successfully averted by the various pro-
tective operations characteristic of borderline functioning. This vul-
nerability to a permanent breakup of the self is the product of a
traumatically crushing or depriving developmental history that has
precluded even minimal consolidation of the archaic grandiose self
and the idealized parent imago. Consequently, unlike the narcissistic
personality, the borderline patient is unable to form a stable mirroring
or idealizing selfobject transference and is therefore unanalyzable by
the classical method.
In contrast with Kohut's conceptualization, our observations are
consistent with those of other analysts who have reported analyses of
borderline personalities in which the therapist was eventually able to
help the patient form a more or less stable and analyzable selfobject
transference (Adler, 1980, 1981; Tolpin, 1980). It is true that the
selfobject ties formed by those patients who are called "borderline"
tend initially to be far more primitive and intense, more labile and vul-
nerable to disruption, and therefore more taxing of the therapist's em-
pathy and tolerance (Adler, 1980, 1981; Tolpin, 1980) than those
described by Kohut as being characteristic of narcissistic personalities.
Furthermore, when the selfobject ties of a patient with a moderate to
severe self disorder are obstructed or ruptured by misunderstandings
or separations, the patient's reactions may be much more catastrophic
and disturbed, for what is threatened is the patient's central self-
regulatory capacity-the basic structural integrity and stability of the
118 CHAPTER EIGHT

sense of self, not merely its affective tone (Adler, 1980, 1981; Stolorow
and Lachmann, 1980). Nevertheless, when their archaic states and
needs are sufficiently understood, these patients can be helped to form
more or less stable selfobject transferences, and, when this is achieved,
their so--called borderline features recede and even disappear. As long
as the selfobject tie to the therapist remains intact, their treatment will
bear a close similarity to Kohut's descriptions of analyses of narcissistic
personality disorders (Adler, 1980, 1981). 1 When the selfobject tie to
the therapist becomes significantly disrupted, on the other hand, the
patient may once again present borderline features. What we wish to
stress is that whether or not a stable selfobject bond can develop and
be maintained (which in turn shapes both the apparent diagnostic pic,
ture and the assessment of analyzability) does not depend only on the
patient's nuclear self pathology. It will be codetermined by the extent
of the therapist's ability to comprehend the nature of the patient's ar,
chaic subjective universe (Tolpin, 1980) as it begins to structure the mi,
crocosm of the therapeutic transference.

The Case of Caroline

Our conception of borderline as phenomena arising and receding

within an intersubjective field is exemplified by the case of Caroline.
The "borderline" symptoms that led Caroline to enter analysis were
immediately precipitated by severe disturbances in her relationship
with her husband. In other words, they arose within a specific inter,
subjective field- that of a precarious, vulnerable self in a failing, ar,
chaic selfobject tie. The analyst, however, did not sufficiently recog,
nize this when the treatment began, and his lack of understanding
complicated and prolonged the treatment. We have since observed
that most often patients enter treatment when there is a breakdown in
an archaic selfobject bond, which has hitherto served to maintain,
however precariously and at whatever cost, the structural cohesion
and stability of the self and the patient's central self,regulatory
capability.

1ln a personal communication (1981), Kohut stated that he had long held views

compatible with those developed here. He wrote: "Insofar as the therapist is able to
build an empathic bridge to the patient, the patient has in a way ceased to be a border·
line case •.. and has become a case of (severe) narcissistic personality disorder."
 TREATMENT OF BORDERLINE STATES 119

Caroline's two previous attempts at treatment had not materially af~

fected the underlying defect in her self~structure. When she entered
the analysis described here, she was 42 years old. Her last analysis had
ended about three years earlier when her analyst told her he didn't feel
he could do any more for her. Since that time she had thrown herself
into various pursuits. She had returned to school to finish her e d u c a ~
tion, which had been interrupted many years before when she m a r ~
ried. In addition, she had involved herself in some charitable and
social activities in an attempt "to feel useful" and to keep herself
occupied.
Caroline spoke with a Southern accent, which became more p r o ~
nounced when she was tense. She was somewhat overweight and at~
tempted to cover this with loose~fitting clothes, which only made it
stand out more. For some time she had been in a state of more or less
constant anxiety, at times hyperactive and at other times withdrawn,
apathetic, and unable to get moving. Early in her treatment, she dis~
played a frightened, little~girl look, expressing her evident discomfort
and not infrequently her terror. She avoided the analyst's eyes almost
completely. In the first weeks, she openly voiced her disbelief that a n y ~
one could help her and said she saw no way out of her difficulties.
Gradually it was reconstructed that her present intractable state
dated from about 10 years earlier and had followed a deterioration in
her relationship with her husband (to whom she had then been m a r ~
ried for about a dozen years). Although Caroline had been a r e a s o n a ~
bly attractive young woman, her shyness and lack of confidence, in
concert with a puritanical upbringing, had constricted her social and
sexual development. Thus, her husband was the first man with whom
she had had a serious relationship. She had been an outstanding
student-her remarkable intelligence was to become clearer as the
treatment progressed-but she left college when she married, in order
to support and further the career of her husband, then in law school.
Subsequently, when he set up practice, she kept house for him, assisted
him in many ways, reared their child, and operated a small business so
that they could prosper financially. In spite of this, their relationship
became more strained and conflicted, as her husband became ever
more displeased with and critical of her-of her accent, her weight,
her anxiety and depression. This culminated in a "borderline" state,
with progressive lethargy, hypochondriacal symptoms, feelings of
deadness that began in her extremities and threatened to engulf her
120 CHAPTER EIGHT

whole body, and frightening delusions about her husband harming,

poisoning, or killing her.
Caroline recovered from this early episode in a matter of weeks, but
many of the symptoms recurred (though not the delusions) and other
symptoms took hold. She began to eat compulsively, and there were
periodic withdrawals during which she remained preoccupied with
puzzles or needlework for long periods of time. In the early months of
treatment, Caroline appeared so distraught and disorganized that the
analyst believed that only by seeing her six times a week could he avert
a prolonged hospitalization or suicide (to which she made several
references).
Whatever the content of the sessions, Caroline reacted to their end,
ing with enormous anxiety and clung to the analyst as the hour drew
to a close, speeding up her associations so that he could not interrupt
her. When he succeeded in calling the session to a halt, she either
continued the conversation until he closed the door behind her or, en,
raged by his interrupting her, walked out in a sullen pout. Weekends
and more prolonged separations produced severe regressive states and
numerous dreams filled with disaster-flooding and drowning, houses
perched precariously on a cliff edge, supports crumbling, black men
pursuing her, and imagery involving a variety of mutilations.
In the first dream that Caroline reported in the analysis, she de,
scribed her husband and her analyst sitting in the living room. She
went to the freezer and took something out. It was the trunk of a frozen
corpse with no limbs. She showed this to the men, but they began to
have sport with it- tossing it around and laughing.
The early sessions were marked by an almost uninterrupted stream
of associations. The analyst found it hard to think, let alone formulate
a coherent understanding of any underlying meaning. As this contin,
ued for some time, it was difficult for the analyst to escape the convic,
tion that she was projecting her anxiety and helplessness into him in
an attempt to rid herself of these feelings.
Gradually, however, it became clear that she was terrified of the an,
alyst and the treatment- terrified that she would be treated cruelly,
driven mad, or abandoned as a hopeless case. These fears were inter,
preted to her as indications of a lack of trust and reluctance to depend
on the analyst. Such interpretations seemed for a time to calm her, and
they evoked memories of her early experiences.
Caroline was the first child of her mother and father. They had mar,
 TREATMENT OF BORDERLINE STATES 121

ried when her mother was approaching 40. Her father, four years older
and a widower with two t e e n ~ a g e d sons, was a h a r d ~ w o r k i n g account-
ant who needed someone to take responsibility for their upbringing.
As a young woman, Caroline's mother had wanted desperately to es-
cape from the drudgery of her small town life, and her love of music
seemed to offer her the opportunity. But she realized rather late that
her hopes of becoming an opera singer or the coach of an operatic
prodigy were destined to disappointment. By that time her chances for
a good marriage had passed her by, and she settled on Caroline's fa-
ther, more with resignation than ardor, a bird in no gilded cage.
Caroline was born two years later, after what she was repeatedly
told was an extremely difficult labor. Three years after her birth a
brother was born. This birth was even more difficult and resulted in se-
vere damage to the mother's pelvic tissues. Afterward the mother took
to her bed in a depression that lasted for many months during which
time she was preoccupied with an assortment of hypochondriacal and
somatic symptoms. When she recovered, she treated Caroline as if the
little girl were an extension of her own defective, diseased self. She re~
acted to every sneeze as if it were a harbinger of death, took Caroline
from doctor to doctor, and kept her out of school for two years. As
Caroline and her health became her mother's sole preoccupation, in-
tense conflicts arose. These centered on what foods Caroline was to
eat, how much and at what intervals she was to sleep, and especially
her bowel habits.
As the treatment progressed, the analyst noted that Caroline was
somewhat better as each week proceeded, but then regressed toward its
end. Weekends remained disasters, with the patient unable to think or
function except at a minimal level. The analyst thought that the mate-
rial indicated Caroline's inability to retain any image of a good object
built up during the sessions-she and it underwent a nearly complete
deterioration during separations. When she returned to analysis, it
was in a state of helplessness. Repeatedly, she then complained that
the analysis was not helping her, and frequently, apparently forgetting
her condition when she entered treatment, she angrily asserted that
the analyst was responsible for her pain and lack of progress.
It was easy for the analyst to conclude that the archaic states of con-
fusion and disintegration into which Caroline lapsed came about b e ~
cause of persistent splitting, that her good internal objects were being
kept widely apart from the bad, that synthesis was being actively pre~
122 CHAPTER EIGHT

vented from occurring, and that she could not simultaneously accept
the analyst's goodness and his separateness. She reacted to his u n ~
availability on weekends and to what he believed were thoughtful and
helpful interpretations as if they were purposely meant to make her suf~
fer. Attacking him in that way, she anticipated being attacked in re~
turn. And she experienced every attempt on his part to explain this
situation to her, no matter how cautiously, tactfully, and empathically
phrased, as a renewed attack on her.
Another "symptom" appeared in Caroline's treatment. One day, in
striking contrast to her usual outfit of jeans and tennis shoes, she a p ~
peared in a lovely skirt and jacket, a pretty blouse, and fashionable
shoes and purse. Greatly embarrassed, she revealed that she had gone
on a spree, bought three outfits, several pairs of shoes, and an assort~
ment of matching accessories. She confided that she did this every
once in a while, in spite of herself. She knew that when she went home
she would have to hide all the things she had bought and might never
be able to wear them, for her husband would be furious with her. He
would be frightened and horrified by her excesses. He maintained a b ~
solute control over the family finances and regarded her buying binges
as symptoms of insanity or as inconsiderate breaches of contract.
Moreover, now he would have further grounds for his understandable
concern over her treatment.
The analyst felt that if her purpose was to project into him her a n x i ~
ety over behavior for which she wished to escape responsibility, she
could not have devised a more effective means. He was also struck by
the excess, the suddenness, and the lack of control, and he tried, w i t h ~
out success, to investigate the spree from that perspective. He was to
learn later that Caroline did not buy another stitch of clothing for
three years.
Caroline's fears of the analyst and the analysis kept recurring. Her
dreams were filled with scalding suns, Chinese tortures, and m o n ~
strously cruel people. Such images were generally interpreted as trans~
ference projections. And gradually some small progress seemed to
occur. Her anger subsided somewhat, her anxiety assumed more m a n ~
ageable proportions, and she was able to read and to socialize to a
greater extent. Yet whenever her old symptoms returned, she thrashed
herself mercilessly. Repeated working through of these themes seemed
to the analyst to leave no alternative to the explanation that s o m e ~
thing in her was opposing success, making it impossible for her to b e n e ~
 TREATMENT OF BORDERLINE STATES 123

fit further from treatment, her marriage, and, indeed, her life. She
made many starts in many directions, but invariably her enthusiasm
disappeared, to be mourned and to become the focus of renewed disap-
pointment and anger with herself. It seemed that continued treatment
would only confirm an omnipotent fantasy that somehow some expe-
rience would magically solve her difficulties without her having to
change.
The analysis, then, appeared to have reached a stalemate. Although
basic problems had not been solved, the prospect of termination
loomed unmistakably, for it seemed to the analyst that more analysis
would only serve to keep Caroline from utilizing the considerable in-
sights she had attained. Rationalizations appeared like weeds after a
rain. After all, her background had left her with a considerable toll.
The difficulties of her attachment to or detachment from her husband,
especially at her age, were all but insurmountable. Her gains, looked at
in a certain light, were not negligible, and it seemed certain that she
was no longer so vulnerable to the threat of collapse that had brought
her into treatment.
In the fourth year of treatment, with many of Caroline's borderline
features still intact, the analyst decided to take one last look. It had
long been apparent that Caroline was disappointed and felt herself to
be a failure, but it was now also becoming clear that she felt that the an-
alyst was disappointed in her and that he considered her and himself
failures. This factor-Caroline's responsiveness to cues of the analyst's
feeling about her-had been grossly underestimated. In fact, as was
later understood, her imperative need to be liked and approved of and
the devastating effect on her of the analyst's disapproval, which she
sensed, had been crucial in structuring the first phase of treatment.
Her depression, her attacks on herself, and her lack of sustaining mo-
tivation all became understandable from this perspective. The analyst
could not continue to maintain that her perceptions of him were all
projection, for he began to recognize in himself what she had been re-
sponding to. This dawning awareness ushered in the second phase of
the analysis.
In a subsequent session, in response to Caroline's expression of wea-
riness and thoughts about terminating, the analyst commented that
he realized that the process was becoming wearing. But could they take
one more good look at what had been occurring before deciding to ter-
minate? Perhaps there was something he had not understood, some-
124 CHAPTER EIGHT

thing that might prove helpful. Perhaps he had conveyed an increasing

disappointment in her and in himself, especially around her continu,
ing symptoms, and perhaps that had contributed in an important way
to her dejection and disparagement of herself. Caroline responded en,
thusiastically. Yes, she exclaimed, she had felt awful about the ana,
lyst's disappointment, which she had sensed. By this time she should
be able to feel better and to control her diet, for she had learned so
much. She had attacked herself mercilessly for not having tried hard
enough. She was weak and self,indulgent, she said, and must want to
spite both her husband and the analyst as she had always defied her
mother. When she was on her diets, she could somehow kill her crav,
ing for food and not be hungry. But something always happened and
she again felt the urge to eat. Then she felt she was a failure and tried
harder and harder. When she was finally unable to stick to her diet,
she hated herself, for she had let the analyst and her husband down.
Once that point had been reached she was absolutely unable to restrain
herself-the more alone she felt, the more she hated herself and the more she
felt compelled to eat.
The analyst was now able to glimpse the transference configuration
that had actually determined the course of Caroline's analysis. To,
gether they began to look at what happened to her when she was
alone, paying increasing attention now to her subjective experiences
and trying to understand them in a different way. There seemed to be a
complex and thoroughgoing alteration of her state of mind- a slipping
away of self,esteem, feelings of accelerating disorganization and dis,
connectedness, an inability to concentrate, and increasing feelings of
deadness, involving coldness and loss of sensation in her limbs, so that
they no longer seemed to belong to her. All these symptoms the ana,
lyst came to recognize as signs of a fragmenting process and of an un,
derlying defect in her self,structure. It became apparent how much
Caroline had looked to the analyst to maintain her sense of self,
needing from him what had not been acquired in her childhood.
When the analyst had interpreted her archaic states and transference
needs as expressions of pathological splitting and projection, she had
become intensely ashamed and self,hating. In their impact on Caro,
line, the interpretations of pathological defenses had repeated the
fragmentation,producing effects of her mother's view of her as defec,
tive and diseased.
It was especially important to Caroline that the analyst be pleased
 TREATMENT OF BORDERLINE STATES 125

with her. She had tried valiantly to get this across to him early in the
analysis, but he had regarded this as defensive. He had not recognized
as primary her specific need to establish him as a selfobject who would
be a source of the mirroring, affirming responsiveness that her self,
absorbed, depressed, and hypochondriacal mother had been unable to
supply during her early formative years. Behind this specific need lay
the vulnerability to fragmentation that had pervaded Caroline's ana,
lytic experiences. When the selfobject tie to the analyst was disrupted
by a failure of the analyst to understand her subjective experience in its
essence or by a loss of connectedness during weekends or vacations,
she could not maintain the cohesion, stability, and affective tone of
her precarious self. She fell apart, eating compulsively in an effort to
strengthen herself and to fill the defect in her sense of self- trying to re,
cover through oral self,stimulation the feeling that she existed at all.
As the structural weakness was being worked through, Caroline re,
alized at one point that she was becoming addicted to television and ra,
clio. When she thought about the vague, apprehensive restlessness she
felt in the absence of sensory stimulation, she realized that "empty" did
not really describe her feeling. Rather, she recognized "a feeling of defi,
ciency, a lack of some very specific supporting structure which would
prevent everything from falling in- some essential piece of myself miss,
ing." When the analyst had taken her symptoms as a disparagement of
his efforts, as a defensive aggrandizement of herself, or as an indication
of greed, she had felt even worse. Feeling blamed, she had relentlessly
blamed herself.
As the disturbance in the transference tie was seen and analyzed in
this new way, with focus on the fragmented states and the underlying
structural deficit, Caroline became more alive, friendlier, much more
enthusiastic, and increasingly capable. Her desire to understand her
states of mind grew in direct proportion to her sense of the analyst's de,
sire to help her acquire this understanding. She expressed appreciation
that the analyst now recognized her vulnerability and the legitimacy of
her fears. "The first thing I had to get across to you," she explained
when she was certain that he would understand her, "was how impor,
tant what you thought of me was. Until that happened nothing else
could happen. I couldn't disagree with you because I was afraid of
worse consequences. So I tried to see and use and apply what you said,
even when it made me hate myself. I tried to think you were opening
up a new world for me, a new way of seeing things that would work out
126 CHAPTER EIGHT

better in the end. And when it wasn't working out that way, I blamed
myself."
With the working through of her fragmented rstates in relation to
their triggering experiences within a disrupted self6bject tie, Caroline's
borderline symptomatology and paranoidlike fears dropped away, to-
gether with what had previously been regarded as splitting, projection,
and a failure to internalize a good object. She and the analyst could
now better understand her dream of the frozen torso and her expecta-
tions of being laughed at. She had often been terrified as a little girl,
but her fears had always been mocked. She could not, for example, let
her mother bathe her or wash her hair, and her mother would be furi-
ous with her. No one understood why she was afraid of her mother-
indeed, afraid of almost everything. She was teased mercilessly by her
brothers for being so afraid. "Girls can't do anything," they would say.
As Caroline,s vulnerability decreased, there were increasing signs
that she was turning once more to the analyst to help her understand
her early relationship with her mother, its effect on her, and how cru-
cial elements were being replicated with her husband and the analyst.
The analyst could now understand the symbolism of an earlier
turning, which he had missed. Her buying binge had contained both
her fear and her intense need to be noticed. As a girl, she had turned to
her father to be noticed, for it was only through connecting herself to
him that she felt she might be able to extricate herself from the
traumatogenic enmeshment with her mother. "But he was remote and
embarrassed by emotion-even by mothees emotion, and even
though he loved mother," she remarked. "When feelings were ex-
pressed, he would look away. Then, after a point, he would introduce
another subject, as if what had taken place before did not exist." Caro-
line remembered wanting her father to pick her up, but he never did,
except as part of a game. She didn't play right, she felt, so she couldn't
be held. And she so wanted him to want to be close to her. She realized
now that when the analyst spoke to her gently and smiled when he
greeted her, she felt real and warm, not frozen. If she had been feeling
bad and hating herself, that made her feel all right.
Caroline had blamed herself when her father hadn't noticed her or
loved her. In particular, she had blamed her anger. The anger evoked
by her father's unresponsiveness had been enormously threatening to
her because of her desperate need for him. Thus, she exonerated him
and blamed her reactive anger for his faulty responsiveness. A similar
 TREATMENT OF BORDERLINE STATES 127

sequence could be observed in reaction to unattuned responses from

her husband and the analyst. Her idealizations were not primarily a
defense against her anger. Rather, she preserved the vitally needed ide*
alizations at the expense of her anger and of her ability to assert herself
when her interests were disregarded.
Caroline had turned to her father not primarily as an oedipal love
object, but as an idealized selfobject whose responsive interest in her
might open a compensatory path along which her thwarted develop*
ment could resume. When this developmental thrust was revived in
the transference, her associations led her back to her fourth and fifth
years. Her memories clearly showed that what she most needed her fa*
ther to notice and understand was what she was going through with
her mother. In the analysis she realized that she had to return to that
time because something had happened then that had made her life
thereafter almost unbearable. She remembered herself before this time
as a well,dressed little girl; afterward she felt like a ragamuffin.
When Caroline was four her mother, then recovering from a pro*
longed depression, had resumed her involvement with the church as
an organist and choral leader. The church and the little girl largely
made up the boundaries of the mother's restricted world. Even then
her mother would often go to bed for the day, saying, "I know I can't
get out today." Caroline remembered that during this period she had
wanted to learn to play the piano. Taking affront that Caroline might
want anyone else to teach her, her mother undertook the task. Caro*
line recalled that as with everything else, her mother insisted on a strict
routine-first, months of finger exercises away from the piano, and
only then the real thing. Her mother was an overwhelming teacher.
When Caroline tried and pleaded, "I can't," her mother flew into a
rage. Later, Caroline came to understand that the rage was toward her
mother's own recalcitrant self, indistinguishable from that of her
daughter. The mother desperately wished that her daughter would not
give up, as she herself had done, that Caroline would not become a no,
body doing the things in the kitchen no one else wanted to do. She in,
sisted that Caroline did not care about her, did not value her. Caroline
could see that her mother believed this, and it scared her. But then she
told herself perhaps her mother was right, perhaps she would never be
able to care for anyone (as she was also told) if she couldn't care for her
mother. It was so frightening to think that her mother didn't under,
stand her that she found it a relief to believe that she herself was bad.
128 CHAPTER EIGHT

Why couldn't she practice, her mother would ask. It was just a m a t ~
ter of moving her fingers. Her mother would demonstrate and then
take Caroline's fingers and show her. It could only be rebelliousness,
Caroline was always so stubborn. Then her mother would get out the
whip as the little girl froze and cowered. It was a black, braided leather
affair with a number of thongs, perfect equipment for not spoiling the
child. Although it was only used three or four times, Caroline would
remember her fear and humiliation for the rest of her life. That ended
her career in music.
One of the most terrifying aspects of these childhood experiences
was that something was glaringly wrong, but nobody seemed to know
it or do anything about it. When Caroline went to her father, he would
change the subject. When she went to the maid, she was told how it
was to be an orphan as the maid had been. Caroline had to find some
way to live with her mother, so she made herself responsible, telling
herself that if she were better her mother would love her. "It is terri~
fying to be in the power of another person," she observed. The feeling
that something was wrong and nobody seemed to know or do a n y ~
thing about it was replicated in the analysis when the analyst failed to
respond to Caroline's assertions of the threat to herself posed by many
of his interpretations.
There was something even worse than whipping, Caroline realized
one day. One of the major methods by which her mother controlled
her was by continuously threatening to leave her. That was always,
and still remained, the ultimate whip, both with her husband and in
the transference. She realized that the threat may have been completely
false objectively, but it was very real to her. Even now, anyone she needed
could reduce her to submission by threatening to leave her. Her
mother had simply walked away from her when the little girl had " m i s ~
behaved" or acted cranky. "It is almost as if you have a choice of ex~
isting or your mother existing, but not both," Caroline explained. The
meaning of a remark at the beginning of the analysis was now more u n ~
derstandable: "I have had to be able to hate my mother in order to stay
alive!"
Caroline recalled that the family had a small house near the ocean,
at the mouth of a river. Her mother was afraid Caroline would drown
and so insisted on teaching her to swim- not in the small river but in
the ocean. Yet her mother herself could barely swim. Caroline r e m e m ~
bered her terror when her mother approached her. She couldn't let her
 TREATMENT OF BORDERLINE STATES 129

mother near her! She couldn't tolerate looking at her because she knew
that just the touch or the look would immediately cause her to lose h e r ~
self, not feel herself. Her mother frequently said, "If you could just see
yourself through somebody else's eyes." Caroline realized how much
she had needed for someone to see through her eyes. In the water she
would scream, "I'll do it myself; please let me do it myself!" Her mother
would stand over her, coldly retorting, "When are you going to do it?
When are you going to do it?"
Caroline often imagined running away from her mother's ruthless
training. One day, in the analysis, she spoke of this, remarking, "If I
had had a father to run to, I would have." It was when she saw all her
little friends playing and going places with their fathers that she began
to feel like a ragamuffin. She remembered so much wanting to run
away, but she was concerned about not having any food. She began to
think about packing food in small packages. She collected Tarzan
books, and she recalled being fascinated because he was able to survive
in the jungle with only a knife; he didn't have to depend on or submit
to ~nyone. Eventually, however, her daydreams of escaping from her
mother collapsed. She was too aware of reality and knew that she
would have to come back, so she made her peace.
At this stage of the analysis, Caroline remarked on a feeling of being
better integrated. The analyst had allowed her to revive in the trans~
ference the l o n g e d ~ f o r selfobject bond to an idealized father, who
would help her understand and separate from her pathological
enmeshment with her mother. Everything she thought about now
seemed more vivid, she commented. Her thoughts and feelings made
more sense to her. She felt more self~confidence, although she was still
worried that this would disappear and not return. Still, she felt she was
stronger, as she put it, than the threat to her was. Moreover, she noted
an increased ability to stick to her moderated diet. Slowly but noticea~
bly, she began to lose weight. There was much more to be done, she re~
alized, but she felt that a corner had been turned, as indeed it had.
To summarize this case: Caroline's adult "borderline" characteristics
and paranoidlike distrust had arisen in the intersubjective field of her
vulnerable, f r a g m e n t a t i o n ~ p r o n e self within a failing, archaic self~
object tie (with her husband). These borderline characteristics re~
mained and were periodically intensified in the new intersubjective
field of the psychoanalytic situation when the analyst's incorrect inter~
pretive stance and faulty responsiveness unwittingly triggered and ex~
130 CHAPTER EIGHT

acerbated her states of selUragmentation. The failures in her marital

relationship and in the first phase of the analysis replicated the specific,
traumatogenic selfobject failures of her early childhood years. C a r o ~
line had adapted to these failures by attempting to serve the archaic
selfobject needs of her mother and pushing herself even harder when
her mother found her wanting in that role. This was repeated with the
analyst. In contrast, in the second phase of the analysis, when the a n a ~
lyst became able to comprehend the actual meaning of Caroline's ar~
chaic subjective states and needs, thereby permitting her to revive and
establish with him the specific selfobject ties that she required, her so~
called borderline features dropped away.

CONCLUSION

We have criticized the view that the term ''borderline" designates a

distinct pathological character structure, rooted in pathognomonic in~
stinctual conflicts and primitive defenses. Instead, we propose an
alternative conceptualization of so~called borderline phenomena from
an intersubjective perspective. In particular, we believe that the d i n ~
ical evidence cited for the operation of primitive defenses against
pregenital aggression is better understood as an indication of needs for
specific archaic selfobject ties, and of disturbances in those ties. As the
case of Caroline suggests, the psychological essence of what is called
"borderline" does not rest in a pathological condition located solely in
the patient. Rather, it lies in phenomena arising in an intersubjective
field, consisting of a precarious, vulnerable self in a failing, archaic
selfobject bond.
We wish to clarify some potential sources of misunderstanding of
our point of view. Conceptualizing borderline phenomena as arising in
an intersubjective field is not equivalent to claiming that the term " b o r ~
derline" refers to an entirely iatrogenic illness. As seen in the case of
Caroline, the failing, archaic selfobject bond is not always with a ther~
apist or an analyst, although this will become increasingly more likely
as the patient's selfobject needs are engaged in the therapeutic transfer~
ence. More importantly, the claim of an entirely iatrogenic illness
would be markedly at variance with our concept of an intersubjective
field and would overlook the contribution of the patient's archaic
states, arrested needs, and f r a g m e n t a t i o n ~ p r o n e self to the formation
 TREATMENT OF BORDERLINE STATES 131

of that psychological field. If we view the therapeutic situation as an

intersubjective field, then we must see that the patient's manifest psy-
chopathology is always codetermined by the patient's self disorder and
the therapist's ability to understand it.
Our claim is not that borderline symptomatology is entirely
iatrogenic, but that the concept of a "borderline personality organiza-
tion" is largely, if not entirely, an iatrogenic myth. We believe that the
idea of a borderline character structure rooted in pathognomonic con-
flicts and defenses is symptomatic of the difficulty therapists have had
in comprehending the archaic intersubjective contexts in which bor-
derline pathology arises.
We wish to emphasize that selfobject failures are developmentally
codetermined subjective experiences of the patient and that therefore
their occurrence in treatment is not to be regarded as an objective in-
dex of the therapist's technical incompetence or inadequacy. They are
revivals in the transference of the patient's early history of develop-
mental deprivation and interference. Thus, the therapeutic task is not
to avert such experiences of selfobject failure but to analyze them from
within the unique perspective of the patient's subjective world.
From the standpoint of the archaic nature of the arrested needs re-
vived in the transference, it is inevitable that the therapist will"fail" the
patient, and that under such circumstances borderline symptoms may
appear. In our experience, it is only when the subjective validity and
meaning for the patient of these disjunctions and selfobject failures go
chronically unrecognized and unanalyzed (often because they
threaten the therapist's self-organization requirements), and the
reestablishment of the therapeutic bond is thereby prevented, that
borderline phenomena become encrusted into what has been de-
scribed as a "borderline personality organization." This formulation of
borderline symptomatology illustrates the general psychological prin-
ciple that psychopathology cannot be understood psychoanalytically
apart from the intersubjective contexts in which it arises and recedes.