Intrusiveness and Closeness Caregiving R
Intrusiveness and Closeness Caregiving R
Intrusiveness and Closeness Caregiving R
"SYMBIOSIS"
In the early years of family therapy, several groups of clinical researchers reported distinctive characteristics in the family
relationships of persons labelled "schizophrenic." The term "symbiosis," which connoted a kind of parasitic relationship
between parent and child, originally was used to describe such relationships (for example, Boszormenyi-Nagy, 1965;
Stierlin, 1959). Old diagnostic labels such as folie à deux and folie en famille also were applied when both spouses or all
family members shared a fixed delusional system.
These terms referred to processes viewed as deeply pathological, a kind of pernicious connectedness filled with
emotional turmoil and mutually smothering restrictiveness. Symbiosis appeared to be maintained by implicit or explicit
threats of punishment or abandonment. Any moves toward separateness or divergent thinking among family members
seemed to elicit terror. Such extreme anxiety was presumed to derive from the psychological survival value of the
relationship and the participants' fears of dying (symbolically) if the quality of relating were to change. Thus, the early
concept of symbiosis was associated with a very extreme and almost horrific interpersonal process in which two or more
family members were so deeply engulfing of one another that even minor acts of autonomous functioning were experienced
as treachery and induced panic in the participants.
To capture the nuances of such processes, early theorists introduced a host of new concepts that ultimately replaced the
single term "symbiosis": the double bind (Bateson, Jackson, Haley, & Weakland, 1956); pseudomutuality (Wynne,
Ryckoff, Day, & Hirsch, 1958); family myths (Ferreira, 1963); mystification (Laing, 1965; Laing & Esterson, 1964);
intersubjective merger (Boszormenyi-Nagy, 1965); parental egocentricity (Lidz, 1973); marital schism and skew; (Lidz,
Fleck, & Cornelison, 1965); projective transference distortions (Framo, 1970); adaptation to the stronger person's reality
(Stierlin, 1959); id / ego / superego binding (Stierlin, 1973); the undifferentiated family ego mass, the family projection
process, and emotional fusion (Bowen, 1966); scapegoating (Ackerman, 1958; Vogel & Bell, 1968); and
consensus-sensitive style of relating (Reiss, 1971a,b).
1
_____________________________________________________________________________________________________________
These notions had in common an underlying dimension of intense connection, but they targeted very specific patterns
including:
(a) High levels of psychologically coercive control, based on a combination of behavioral dominance, restrictiveness,
character-focused criticism, threats of abandonment, and sometimes physical aggression or threats of aggression;
(b) Extreme separation anxiety;
(c) Possessiveness/jealousy, presumably as a result of insecure attachment and fears of rejection;
(d) Rapid and extreme emotional reactivity to other family members' affective states, including taking too much
responsibility for solving one another's emotional upsets;
(e) Interpersonal projection, manifested as a failure to distinguish one's own perceptions, motivations, opinions,
feelings, needs, and values from those of other family members;
(f) Mystification of experience, which involves imposing one's definitions of reality on other family members (Laing,
1965; Wynne, 1971);
(g) Confusing communications (including amorphous and fragmented attempts to share meaning, and nonverbal or
verbal disqualifications of messages); and
(h) Atypical views of the world and the family (peculiar or illogical interpretations of external events, and/or family
self-images that were at marked odds with other information about the family).
In the remainder of this article, we will use the term Intrusiveness to refer collectively to the first six processes described
immediately above (items a through f). These six components pertain to psychological coercion, excessive dependency,
and lack of tolerance for self/other differentiation, all of which are associated with a stifling of individuality. However,
because dimensions g and h above pertain to cognitive aspects of family functioning (that is, coherent communication and
the logical basis of beliefs), we believe they are best conceptualized as separate from intrusiveness despite possible
co-occurrence in the families of schizophrenics. For example, psychological coercion, excessive dependency, and lack of
tolerance for self/other differentiation seem to be very high between violent husbands and battered wives; but
communication in such couples ordinarily does not contain the extreme levels of fragmentation and ambiguities found in the
families of schizophrenics (Singer, Wynne, & Toohey, 1978).
Lastly, the early theorists proposed that dyadic processes of intrusive over-involvement or emotional distance in couples
often were accompanied by simultaneous blurring of generational boundaries and the formation of cross-generational
coalitions or other pathological triangles involving parents and children (Bowen, 1966; Haley, 1967; Lidz et al., 1965). In
these triads, a child might be pressured to take sides between two warring parents; or one parent and child might be
embroiled in an enduring coalition that excluded the other, more peripheral parent. Such parent-child overinvolvement was
viewed as both an effect and a cause of the family difficulties.
As an anchor for the rest of this article and to help readers visualize the possible intensity of these interpersonal
processes, we present brief vignettes of two intake interviews.
2
_____________________________________________________________________________________________________________
occasions when his wife aggressively insisted on the odor's presence, he thought he "might" have smelled it. Although he
had doubts about whether the odor problem existed, he never shared these doubts with his son for fear of upsetting his
wife's tenuous psychological equilibrium or incurring her fury.
In her individual session, Terri (a junior in high school) revealed she did not think Derek had an odor problem. Rather,
she thought the other family members were acting strangely. She coped by avoiding all discussions of the problem and
spending as much time as possible with friends and away from the family.
In his individual session, Derek expressed great ambivalence about the odor problem, stating that he never noticed any
unusual smell, but he assumed that other family members could detect the odor because they were not around it all the time.
He was frightened by his mother's attribution that he was "ill." Although he had been lonely, awkward, and shy for all of his
life, he thought he had been making some social progress recently. His mother's phone calls to the dormitory annoyed and
embarrassed him, yet he did not want to anger her or hurt her feelings by rejecting the calls.
From the clinical team's perspective, Derek did not show any signs of severe mental illness (such as schizophrenia or
major affective disorder), nor did he have any noticeable odor, although he clearly had some long-standing interpersonal
problems.
3
_____________________________________________________________________________________________________________
Although the relationships between Derek and his mother or between the Joulets may have the superficial form of
closeness (for example, higher than expected levels of contact), this type of closeness seems to arise from processes of
coercion, collusion, and anxious attachment rather than from authentic mutuality (Wynne et al., 1958; Weingarten, 1991).
Like a close embrace held too long or too hard, the inability to free each other tends to be experienced as an act of
insecurity or control rather than an act of affection or support. Labeling such relationships "extremely close" or "cohesive"
masks their other qualities and potentially invalidates the experience of the participants.
Early writings in our field were replete with extraordinary case examples of symbiotic processes like those in the
Zelinsky and Joulet families (see, for example, Boszormenyi-Nagy & Framo, 1965; the classic articles reprinted in Green
& Framo, 1981; Jackson, 1968; Laing & Esterson, 1964; and Lidz et al., 1965). However, in the mid-1970s, the field's
attention turned away from schizophrenia and toward studies of adolescent delinquency and psychosomatic disorders.
Coinciding with this shift, some of the key elements of intrusiveness all but vanished from the literature. Under the global
umbrella of "enmeshment" offered by structural family therapists, distinctions among categories such as psychologically
coercive control, possessiveness/jealousy, and mystification seemed to fade from view.
"ENMESHMENT"
During the period 1974-78, Minuchin and colleagues condensed and systematized much of the prior theorizing about
family dysfunction and called this new framework "Structural Family Therapy" (Minuchin, 1974; Minuchin, Rosman,
& Baker, 1978). In order to achieve this distillation of family therapy concepts, Minuchin (1974) developed the central
organizing metaphor of family structure consisting of interpersonal and subsystem boundaries. Family structure was
defined as "the invisible set of functional demands that organizes the ways in which family members interact" (p. 51).
Family systems were viewed as being organized into subsystems (the spouse subsystem, the parental subsystem, and the
sibling subsystem). Boundaries were described as "the rules defining who participates and how," especially in terms of
participating in family subsystem functioning (p. 53).
Minuchin (1974) formulated boundaries as existing on a bipolar linear continuum ranging from diffuse boundaries
(enmeshed), to clear boundaries (normal range), to inappropriately rigid boundaries (disengaged). Clear boundaries
promoted functional relating and adaptation to changing intrafamilial and extrafamilial demands. Diffuse boundaries
(enmeshment) and rigid boundaries (disengagement) were put at opposite ends of the continuum, and both were viewed as
generally dysfunctional. Enmeshment was defined as involving excessively rapid and intense emotional reactivity that
blocked individual family members' autonomy. Disengagement was defined as involving excessively delayed and low
emotional reactivity, too much autonomy, lack of feelings of loyalty and belonging, and loss of the capacity for
interdependence. Minuchin rarely referred to enmeshment and disengagement as if these were properties of whole family
systems. Rather, he generally took pains to emphasize that some subsystems in a family may be enmeshed while others in
the same family may be disengaged.
Clearly, the above descriptions of enmeshment and diffuse boundaries incorporated some elements from the earlier
theories of family dysfunction. However, three important differences between enmeshment and symbiotic processes seem to
have escaped careful scrutiny: (a) In structural theory, enmeshment and disengagement were posed as logical opposites
rather than as separate or as sometimes co-occurring dysfunctional processes; (b) the theory emphasized "clarity" of
generational boundaries (weak versus strong hierarchy) rather than processes of psychological coercion, per se; and (c) the
theory's central metaphor was spatial (emphasizing closeness-distance) and lumped together presumably functional and
dysfunctional patterns on a single continuum, rather than encouraging separate assessments of health-related and
pathology-related domains. We will discuss each of these issues below.
4
_____________________________________________________________________________________________________________
"enmeshment."
Meaning #1: Enmeshment is a lack of self/other differentiation. At first, Minuchin et al (1978) define enmeshment and
disengagement as "two extremes" of the boundary clarity continuum. Enmeshment is described as entailing a dysfunctional
lack of self/other differentiation, the logical opposite of which (in our opinion) would seem to be individuation:
The enmeshed family is a system which has turned upon itself, developing its own microcosm [such that]
boundaries are blurred, and differentiation is diffused. Such a system may lack the resources necessary to adapt and
change under stressful circumstances. [pp. 56-57]
Meaning #2: Enmeshment is a style of high closeness and caregiving. In the next paragraph, Minuchin et al. (1978)
imply a very different definition, now placing enmeshment and disengagement within the realm of normal development. In
this second usage, enmeshment is presented as a transactional style of high closeness and caregiving, the logical opposite of
which would seem to be "disengagement" (emotional distance and lack of caregiving):
Most families have enmeshed and disengaged subsystems, varying according to function and developmental level.
Mother-children subsystems are often enmeshed while the children are small, even to the point of excluding the
father.... The parental subsystem moves from an enmeshed to a disengaged style as the children grow and begin to
separate from the family. [p. 57]
The implication in these passages is that "enmeshment" and "disengagement" (supposedly dysfunctional extremes) can
also be functional processes at times. We believe that Minuchin et al. (1978) cloud the underlying conceptual issues by
suggesting that "enmeshment" is desirable or normative developmentally for young children and their parents. It would be
more intelligible to state that parental closeness-caregiving (Meaning #2) should be higher when children are younger.
However, "extreme and dysfunctional diffusion in personal boundaries" (Meaning #1) would be inappropriate regardless of
a child's age, given that parents should support age-appropriate striving toward individuation in all children.
In short, from our perspective, Minuchin and colleagues blended two conceptually distinct notions, which might be best
termed closeness-caregiving and intrusiveness. We think that assessing these two dimensions separately, rather than
lumping them together, will yield more valuable information for designing therapeutic interventions and testing research
hypotheses.
For the purposes of discussion only, we depict in Figure 1 the extreme combinations of these dimensions. We hasten to
add, however, that we do not intend Figure 1 as a grid for assigning all families to quadrants in a typology. Rather, we think
of closeness-caregiving and intrusiveness as two linear continua, and we hypothesize that interpersonal behaviors along
these dimensions have separate and joint effects on psychological well-being. Figure 1 is presented only to help readers
grasp the conceptual issues, and we would expect that only a small percentage of family dyads (with extreme behaviors on
both dimensions) would fit unconditionally into the hypothetical quadrants depicted in Figure 1.
5
_____________________________________________________________________________________________________________
Figure 1.
Four extreme combinations of closeness-caregiving and intrusiveness.
Although there may be sociocultural, developmental, and situational (for example, health-related) variations in what
constitutes high and low levels of these dimensions, we expect that family relationships with relatively high
closeness-caregiving and low intrusiveness (quadrant I) would be optimal in terms of family members' psychological
well-being. Compared to other families in that culture, at that developmental stage, and in that situation, such family
relations (quadrant I) would be highly nurturing and supportive but also highly respectful of individuation. On the other
hand, we would expect the combinations in the other quadrants to be associated with family members' global psychological
distress and perhaps with specific clinical problems such as anxiety disorders (quadrant III). If high levels of parental
communication deviances also were present, relationships in quadrants II, III, or IV might be associated with different
subtypes of schizophrenia spectrum disorders.
Figure 1 also underscores some dilemmas inherent in the structural family model. Would structuralists consider only
relationships that are high in both closeness-caregiving and intrusiveness (quadrant III) "enmeshed"? Or would
structuralists also view being high in either closeness-caregiving (quadrant I) or intrusiveness (quadrant IV) "enmeshed"?
Would they consider only relationships that are low in both closeness-caregiving and intrusiveness (quadrant II)
"disengaged"? Or would they also consider relationships that are low in either closeness-caregiving (quadrant IV) or
intrusiveness (quadrant I) "disengaged"?
According to the structural model, it is logically impossible for a family relationship to be both disengaged and enmeshed
simultaneously. However, we believe that such a "low closeness-caregiving/high intrusiveness" pattern is quite common
clinically (quadrant IV). For example, in our clinical experience, married partners who exhibit much
possessiveness/jealousy and projective mystification (components of high intrusiveness) often tend to become quite
withdrawn and distant emotionally from one another over time (low in warmth, nurturance, and emotional consistency).
6
_____________________________________________________________________________________________________________
Such distance, in turn, often seems to exacerbate mistrust, insecurity about one another's commitment, jealousy, and
projective attributional processes in the relationship.1
Also regarding quadrant IV, if a parent is highly emotionally reactive, clinging, and intrusive, an adolescent may become
increasingly cold, distant, and inaccessible in response. Thus, in reciprocal fashion, intrusiveness may "invite" low
close-ness-caregiving low closeness-caregiving from the other, and low closeness-caregiving may "invite" even greater
intrusiveness, in a vicious spiral or oscillating pattern. Conversely, in other families, quadrant I patterns may prevail
because high closeness-caregiving reduces or eliminates the kind of insecurity that propels intrusive processes. In this
sense, mutual closeness-caregiving may decrease mutual intrusiveness, which in turn may facilitate mutual individuation.
As a final comment on Figure 1, we believe that our two family dimensions seem most pertinent for characterizing dyads
(person to person relations) rather than the family as a whole. This view is consistent with the structural idea that different
subsystems within a family may differ in levels of enmeshment and disengagement. Patterns in family triads or larger family
groupings would seem best portrayed in terms of the configuration of, and comparisons among, dyadic relations in the
family. Thus, in a given case, the father may be close and intrusive with mother but distant and intrusive toward child; or
intrusively overinvolved with son but not with daughter; or each parent may be closer to the child of the same sex than to
the spouse or to the child of the other sex.
Proximity/Intrusiveness
Wood's (1985) analysis also revealed a second meaning of "boundary" in Minuch-in's structural family theory. In this
usage, "boundary" refers to interpersonal proximity. Thus a boundary is a semipermeable barrier to ingress or to the
exchange of material, energy, and information across person-to-person territories. For this analysis of proximity, Wood
(1985) intentionally borrowed concepts of interpersonal relatedness from the social sciences that were "neutral" with regard
to health versus pathology. She cautioned that earlier structural concepts such as "blurred" and "rigid" boundaries promoted
the attribution of pathology to potentially healthy relations (Wood & Talmon, 1983).
In this temporal-spatial, health/pathology neutral sense, Wood suggested that the relative sharing of territory could be
assessed along six dimensions of interpersonal proximity: (a) contact time (time together); (b) personal space (physical
7
_____________________________________________________________________________________________________________
nearness, touching); (c) emotional space (reactivity to, and sharing of, one another's affects); (d) information space (facts
known about each other, including feelings, thoughts, biographical, behavioral); (e) conversation space (shared private
conversations that are separate from conversations with others); and (f) decision space (extent to which various
decision-making processes are localized within certain individuals, dyads, subsystems, or the family-as-a-whole).
This analysis highlights the importance of distinguishing between healthy and unhealthy aspects of proximity. With the
exception of extreme "emotional reactivity," Wood's (1985) components of proximity do not seem attuned to some of the
qualities of intrusiveness (such as possessiveness/jealousy and projective mystification) that were described under the
rubric "symbiosis." However, Wood's categories do provide outstanding anchors for thinking about aspects of
"closeness-caregiving"a dimension germane to the assessment of "healthy" functioning. Hence, we define
"closeness-caregiving" primarily along temporal-spatial lines, consistent with Wood's categories of contact time, personal
space, and conversation space. In our view, the most important elements of closeness-caregiving include: warmth,
nurturance, time together, physical intimacy (affection), and consistency. We would expect that higher levels of these
qualities would contribute to the psychological well-being of family members.
The subcomponent of interpersonal proximity that Wood (1985) labels "information space" does not necessarily seem
related to our notion of closeness-caregiving. For example, partners in highly traditional, role-oriented, routine-oriented
marriages often are quite warm, nurturing, physically affectionate, and consistent with one another, but not particularly
self-revealing or intimate in self-expressionone aspect of "information space" (Wynne, 1984). For this reason, we believe
that "information space" pertains to a dimension of family functioning best conceptualized as separate from either
intrusiveness or closeness-caregiving. We have termed this third major construct "openness of communication." The latter
involves such behaviors as self-disclosure and forthrightnessas opposed to self-concealment and evasivenessincluding
the extent to which family members openly reveal or avoid discussing differences.2
In terms of choosing family variables that are "neutral" with respect to health/pathology, we agree with Wood (1985) that
family theory and research methods ought to be able to encompass the entire spectrum of functional and dysfunctional
relating. However, rather than trying to assess both sets of variables on purportedly neutral dimensions or on bipolar
dimensions (as implied by structural theory's single boundary clarity continuum), we believe this enterprise will require
separately assessing pathology-related aspects of connectedness (such as intrusiveness) and health-related aspects of
connectedness (such as closeness-caregiving and openness of communication). For instance, assessments of
pathology-related variables such as family communication deviances and negative expressed emotion have proven equally
fruitful in studies of ordinary and high-achieving elementary school children, learning disabled youths, and young adults
with schizophrenia (Ditton, Green, & Singer, 1987; Green, 1995; Hagmann, 1993; Hahlweg & Goldstein, 1987;
Rasku-Puttonen, Lyytinen, Poikkeus, et al., 1994; Ratnam, 1994; Singer et al., 1978). Based on these findings, we think
that a health/pathology neutral approach does not require family researchers to select only measures that bear an uncertain
relationship to psychopathology. Rather, regardless of whether the focal population is clinical or nonclinical, assessment
should try to capture both positive and negative aspects of family relations by appraising them separately for each family.
8
_____________________________________________________________________________________________________________
bipolar Flexibility continuum (rigid versus chaotic) were hypothesized as generally problematic, whereas the middle levels
(structured or flexible) were hypothesized as more functional.
However, in explaining how families alter their relational patterns to adapt to changing circumstances, Olson
(1993)like Minuchin previouslyintermingles the concepts of enmeshment and closeness. For instance, in a recent case
description of a normal couple's development, Olson states: "After marriage, the newlywed couple could best be described
as structurally enmeshed.... Being in love and enjoying spending maximum time together, they are still enmeshed" (pp.
117-118). This passage implies that the notion of "enmeshment" (a generally dysfunctional extreme) also refers to being "in
love" and "enjoying" each other.
Similarly, in describing a case of a normal family's response to a 53-year-old father's massive heart attack, Olson (1993)
concluded: "Once the heart attack occurred, ... the family quickly shifted to becoming more chaotically enmeshed.... Very
high levels of closeness, characterized by enmeshment, occurred because the illness brought the family closer together
emotionally" (p. 119).
We believe that the above passages merge definitions of two distinct relational dimensions"closeness-caregiving" and
"intrusiveness." Whereas well-functioning newlyweds and families of heart attack patients might be unusually close and
caregiving relative to their mode of operating at other times (such as spending more time together, being more nurturing,
having more physical contact), one would not expect ordinary newlywed couples or families of heart attack patients to show
higher levels of interpersonal projection, jealousy, coercion, or other forms of poor self/other differentiation.
With respect to the Family Adaptability and Cohesion Evaluation Scales (FACES), much has been written and debated
about the conceptual coherence and validity of the Circumplex Model and its assumptions that Cohesion and Flexibility
bear a curvilinear relationship to family pathology. These controversies fall into three broad areas:
(a) Unsuitable Labels: It is confusing conceptually to define very high "Cohesion" and very high "Flexibility" as
problematic, given that in common discourse these two words generally connote positive attributes and imply that
higher amounts are better.
(b) Enmeshment and Disengagement Are Not Opposites: With reference to our previous discussion, the extreme
end-points on the Circumplex Model's Cohesion dimension (enmeshed versus disengaged) are not necessarily
antithetical kinds of family interaction. The converse of "enmeshed" is more appropriately defined as
"individuated/not enmeshed" (rather than "disengaged," the opposite of which is more appropriately defined as "high
closeness and caregiving").
(c) Lack of Empirical Support for the Curvilinear Hypothesis: Research on the most popular measure (FACES)
designed to assess these two dimensions has shown that the low end of the Cohesion dimension (disengaged) and the
low end of the Flexibility dimension (rigid) are associated with clinical problems as predicted by the model.
However, contrary to the Circumplex Model's predictions, higher scores on Cohesion and Flexibility have proven to
be associated with better functioning (Cluff, Hicks, & Madsen, 1994). These findings imply that the associations
between psychological well-being and FACES dimensions are linear (higher scores signifying that relationships are
more functional) rather than curvilinear (in which moderate scores would signify more functionality than either high
or low scores). As Olson (1993) suggests, these results render the labels "enmeshed" and "chaotic" inappropriate for
high scores on FACES cohesion and flexibility.
It is beyond the scope of this theoretical article to review each of the research studies that have generated the preceding
doubts about the validity of FACES; however, Cluff et al. (1994) and Olson (1993) have covered this ground
comprehensively. From the standpoint of theory, we view the difficulties that Olson and colleagues encountered in
operationalizing FACES to be the natural outgrowth of confusions that existed in the field prior to their initial 1979 article.
Following Minuchin and colleagues after 1974, the authors of the Circumplex Model continued to blend the concepts
of"closeness-caregiving" and "intrusiveness" into one concept, which Minuchin (1974) called "Boundary Clarity," and
which Olson and colleagues (1979) subsequently named "Cohesion."
As a result of this conceptual jumble, virtually all of the content in the FACES Cohesion items pertains only to what we
have termed "closeness-caregiving." The actual content of FACES III Cohesion items makes this point plain. The
questionnaire asks respondents to answer items using a 5-point scale ranging from "almost never" to "almost always." The
10 Cohesion questions include content pertaining to intrafamilial help-seeking, approval of family members' friends, liking
joint activities, feeling closer to family than to nonfamily members, liking leisure time together, feeling close, sharing
avocations, making decisions together, ease of generating ideas for joint activities, and valuing "togetherness"highly. In
terms of content validity, from our perspective, these items measure a positive form of closeness-caregiving.
By contrast, the FACES Cohesion questions do not seem attuned to measuring the intertwining of psychesthe lack of
self/other differentiationthat underlies the key dimensions of intrusiveness (psychologically coercive control, separation
anxiety, possessiveness/jealousy, emotional reactivity, and projective mystification). Hence, the FACES items are not able
to capture the levels of intrusiveness that occur in some families seen in clinical practice, such as the Zelinskys and the
9
_____________________________________________________________________________________________________________
Joulets (presented at the outset of this article); or "gruesome-twosomes"couples such as Martha and George as depicted in
the play Who's Afraid Of Virginia Woolf; or battered-wife couples such as Ike and Tina Turner as depicted in the film
What's Love Got To Do With It. To paraphrase Tina Turner's rhetorical lyric from the film's theme song: "What's love
[closeness-caregiving] got to do with intrusiveness?"
It seems fair to conclude that only closeness-caregiving (not intrusiveness) is measured by FACES cohesion. In fact, we
think FACES is one of our field's most concise and strongly validated instruments for measuring global closeness (versus
distance) and deserves continued use for this purpose. From a heuristic standpoint, the quandary in interpreting FACES
results (Cluff et al., 1994; Olson, 1993; Zacks, Green, & Marrow, 1988) ultimately alerted us to the conceptual issues
being discussed here and spurred us to attempt a new operationalization of the "enmeshment" construct.
CLINICAL IMPLICATIONS
Given the field's general confusion over the enmeshment construct, it would seem likely that family therapists sometimes
misinterpret closeness-caregiving in relationships, pathologizing such competence-enhancing processes as though they
constituted forms of negative enmeshment. In particular, ethnic minority families and women's relational styleswhich in
many cases are more oriented toward caregiving, closeness, cooperation, and expressivenessmay be particularly
vulnerable to such mislabelling (Boyd-Franklin, 1989; Goldner, 1985; McGoldrick, Anderson, & Walsh, 1989;
McGoldrick, Pearce, & Giordano, 1982; Walters, Carter, Papp, & Silverstein, 1988; Weingarten, 1991). Women in all
10
_____________________________________________________________________________________________________________
cultural groups may bear the brunt of this bias (being inappropriately tagged "overinvolved mothers," "co-dependent
personalities," or "pursuers" in relationships). One might even speculate that the field of family therapy's tendency to equate
closeness-caregiving and enmeshment was fueled by androcentric, European/American, middle-class ethnocentric models
of mental health, which place comparatively lower value on closeness and caregiving (and higher value on individuals'
separateness and self-sufficiency) in family relations (Surrey, 1985).
A particularly striking example of such apparently sexist and heterosexist bias appeared in early clinical articles on
lesbian couples. These articles described lesbian couples as generally "suffering" from an intense and problematic level of
"fusion" or "merger" (Burch, 1982; Krestan & Bepko, 1980). However, as a function of both partners being socialized
normatively as women, one would expect lesbian couples to be more oriented toward caregiving and more cooperative in
their decision making than heterosexual couples. To clinicians trained in an androcentric and heterocentric society, this
mutually caregiving, highly relationship-oriented style might make lesbian couples seem "too close" (Mencher, 1990).
Research on community (nonpatient) samples of lesbian couples reveals that they show dramatically higher levels of
Cohesion and Flexibility on FACES III than do heterosexual or gay male couples. However, consistent with the
closeness-caregiving content of FACES items, the lesbian couples also reported significantly more satisfaction with their
couple relationships (Green, Bettinger, & Zacks, 1996; Zacks et al., 1988). Other research has shown that lesbian couples
are exceptionally egalitarian in decision making and in allocating household and childrearing tasks (Laird & Green, 1996;
Peplau, 1991). Based on the general finding that FACES Cohesion scores are linearly related to well-being, and given
lesbian couples' higher satisfaction scores, it now appears that lesbian couples' extreme scores on FACES III (which would
be labeled "enmeshed") actually represent very high levels of closeness-caregiving, as would be expected for two partners
socialized as women.
More generally, our field's preoccupation with enmeshment and fusion-related concepts may act as a self-fulfilling
prophecy, predisposing family therapists to look for enmeshment and to misinterpret closeness-caregiving. Evidence
consistent with such a self-fulfilling prophecy comes from a study of therapists' perceptions of normal family functioning.
Kazak, McCannell, Adkins, et al., (1989) administered FACES II and the Family Environment Scale (FES; Moos & Moos,
1986) to four groups: families with young children, college students, grandmothers, and therapists (who had an average of
6.2 years doing family therapy). All respondents were instructed to base their answers on "the extent to which you think
each statement is true of a normal family" (Kazak et al., p. 282). In this way, the researchers examined the four groups'
"normality beliefs" about Cohesion in families.
Each group's "normality beliefs" about Cohesion on the FACES II and FES scales were compared to the three other
groups' "normality beliefs," as well as to the actual family Cohesion norms published for each measure. Analyses revealed
that, compared to therapists, each of the other three groups thought that significantly higher amounts of family cohesion
were normal. In addition, the published norms for actual cohesion in families were significantly higher than the amounts
of cohesion believed to be normal by therapists!
There are numerous alternative interpretations for these results, including the possibilities that therapists, based on their
clinical or other life experiences, have relatively scaled-down, more realistic, or more pessimistic expectations for the
amount of cohesion in average families. However, another possible interpretation, suggested by one of the authors of
FACES (Russell, 1989), is that therapists, in contrast to ordinary people, may have been taught to devalue togetherness
and to believe that high levels of cohesion (closeness-caregiving) are somehow dangerous, equivalent to enmeshment,
and pathological rather than functional:
In particular, family therapists are generally introduced to such concepts as "enmeshment," "fusion," being too
"close", too finely tuned-in to other family members ... The general public is not trained to attend to the permeability
of boundaries and may well equate "closeness" and "togetherness" with voluntary intimacy, comfort, support, and an
absence of pathology. It may not occur to the general public that it is possible to be "too close." Those possibilities
may, however, be primary in the minds both of clinicians and of those researchers who construct self-report scales.
[p. 300]
From our point of view, the general public's assumption seems correct, in the context of FACES and FES cohesion items'
content. It is not possible to be "too close" in terms of the way cohesion is operationalized in these self-report
questionnaires.
If enmeshment is so "primary" in the minds of therapists that they value or expect less closeness and caregiving than do
ordinary persons, there is a danger that these professionals will:
(a) pathologize ordinary amounts of closeness-caregiving by labeling them "excessive or enmeshed";
(b) collude with one or more family members' lower-than-average caregiving behavior toward other family members;
and
(c) reinforce one or more family members' unnecessarily low expectations for closeness and caregiving from other
11
_____________________________________________________________________________________________________________
family members.
In view of these possibilities, we invite family therapists to consider the concepts presented in the Appendix. These
definitions should help distinguish intrusiveness, closeness-caregiving, and openness of communication in family relations
and in client narratives about self and others. In addition, certain of these dimensions may be useful for assessing
therapist/client interactions. Regardless of therapists' theoretical orientations, the 13 dimensions may help focus treatment
on specific areas for change and on client strengths.
REFERENCES
1. Ackerman, N. W., (1958). The psychodynamics of family life: Diagnosis and treatment of family relationships.
New York: Basic Books.
2. Bateson, G., Jackson, D. D., Haley, J. and Weakland, J. H., (1956) Toward a theory of schizophrenia. Behavioral
Science, 1, 251-264.
3. Beavers, W. R. and Hampson, R. B., (1993). Measuring family competence: The Beavers Systems Model (pp.
73-103). In F. Walsh (ed.), Normal family processes (2nd ed.). New York: Guilford Press.
4. Boszormenyi-Nagy, I., (1965). A theory of relationships: Experience and transaction (pp. 33-86). In I.
Boszormenyi-Nagy & J.L. Framo (eds.), Intensive family therapy. Hagerstown MD: Hoeber Medical Division,
Harper & Row.
5. Borzormenyi-Nagy, I. and Framo, J. L. (Eds.), (1965). Intensive family therapy: Theoretical and practical
aspects. Hagerstown MD: Hoeber Medical Division, Harper & Row.
6. Bowen, M., (1966) The use of family theory in clinical practice. Comprehensive Psychiatry, 7, 345-374.
7. Boyd-Franklin, N., (1989). Black families in therapy: A multisystems approach. New York: Guilford Press.
8. Burch, B., (1982) Psychological merger in lesbian couples: A joint ego psychological and systems approach.
Family Therapy, 9, 201-208.
9. Byng-Hall, J., (1995) Creating a secure family base: Some implications of attachment theory for family therapy.
Family Process, 34, 45-58.
10. Cluff, R. B., Hicks, M. W. and Madsen, C. H., Jr., (1994) Beyond the Circumplex Model: I. A moratorium on
curvilinearity. Family Process, 33, 455-470.
11. Ditton, P., Green, R.-J. and Singer, M. T., (1987) Communication deviances: A comparison between parents of
learning-disabled and normally achieving students. Family Process, 26, 75-87.
12. Dunn, J. and Plomin, R., (1991) Why are siblings so different? The significance of differences in sibling
experiences within the family. Family Process, 30, 271-283.
13. Ferreira, A. J., (1963) Family myth and homeostasis. Archives of General Psychiatry, 9, 457-463.
14. Framo, J. L., (1970). Symptoms from a family transactional viewpoint (pp. 125-171). In N.W. Ackerman, J. Lieb,
& J.K. Pearce (eds.), Family therapy in transition. Boston: Little, Brown & Co.
15. Goldner, V., (1985) Feminism and family therapy. Family Process, 24, 31-47.
16. Green, R.-J., (1995). High achievement, under-achievement, and learning disabilities: A family systems model
(pp. 207-249). In B.A. Ryan, G.R. Adams, T.P. Gullotta, R.P. Weissberg, & R.L. Hampton (eds.), The
family-school connection: Theory, research, and practice. Thousand Oaks CA: Sage Publications.
17. Green, R.-J., Bettinger, M. and Zacks, E., (1996). Are lesbian couples fused and gay male couples disengaged?:
Questioning gender straight-jackets (pp. 185-230). In J. Laird & R.-J. Green (eds.), Lesbians and gays in couples
and families: A handbook for therapists. San Francisco: Jossey-Bass.
18. Green, R.-J. and Framo, J. L. (Eds.), (1981). Family therapy: Major contributions. New York: International
Universities Press.
19. Hagmann, J., (1993). Parent-child communication clarity, children's academic achievement, and psychosocial
functioning: A study of inner-city African-American families. Unpublished doctoral dissertation, California School
of Professional Psychology, Alameda.
20. Hahlweg, K. and Goldstein, M. J. (Eds.), (1987). Understanding major mental disorder: The contribution of
family interaction research. New York: Family Process Press.
21. Haley, J., (1967). Toward a theory of pathological systems (pp. 11-27). In G. Zuk & I. Boszormenyi-Nagy (eds.),
Family therapy and disturbed families. Palo Alto CA: Science & Behavior Books.
22. Jackson, D. D. (Ed.), (1968). Communication, family, and marriage: Human communication, Vol. 1. Palo Alto
CA: Science & Behavior Books.
23. Karpel, M., (1976) Individuation: From fusion to dialogue. Family Process, 15, 65-82.
24. Kazak, A. E., McCannell, K., Adkins, E., Himmelberg, P. and Grace, J., (1989) Perception of normality in
families: Four samples. Journal of Family Psychology, 2, 277-291.
12
_____________________________________________________________________________________________________________
25. Krestan, J.-A. and Bepko, C. S., (1980) The problem of fusion in the lesbian relationship. Family Process, 19,
277-289.
26. Laing, R. D., (1965). Mystification, confusion, and conflict (pp. 343-363). In I. Boszormenyi-Nagy & J.L. Framo,
(eds.), Intensive family therapy: Theoretical and practical aspects. Hagerstown MD: Hoeber Medical Division,
Harper & Row.
27. Laing, R. D. and Esterson, A., (1964). Sanity, madness, and the family. New York: Penguin.
28. Laird, J. and Green, R.-J. (Eds.), (1996). Lesbians and gays in couples and families: A handbook for therapists.
San Francisco: Jossey-Bass.
29. Lidz, T., (1973). The origin and treatment of schizophrenic disorders. New York: Basic Books.
30. Lidz, T., Fleck, S. and Cornelison, A., (1965). Schizophrenia and the family. New York: International
Universities Press.
31. McGoldrick, M., Anderson, C. M. and Walsh, F. (Eds.), (1989). Women in families: A framework for family
therapy. New York: W.W. Norton.
32. McGoldrick, M., Pearce, J. K. and Giordano, J. (Eds.), (1982). Ethnicity and family therapy. New York: Guilford
Press.
33. Mencher, J., (1990). Intimacy in lesbian relationships: A critical re-examination of fusion. Work in Progress,
No. 42. Wellesley MA: Stone Center Working Papers Series.
34. Minuchin, S., (1974). Families & family therapy. Cambridge: Harvard University Press.
35. Minuchin, S., Montalvo, B., Guerney, B. G., Jr., Rosman, B. L. and Schumer, F., (1967). Families of the slums:
An exploration of their structure and treatment. New York: Basic Books.
36. Minuchin, S., Rosman, B. L. and Baker, L., (1978). Psychosomatic families: Anorexia nervosa in context.
Cambridge: Harvard University Press.
37. Moos, R. H. and Moos, B. S., (1986). Family Environment Scale manual (2nd ed.). Palo Alto CA: Consulting
Psychologists Press.
38. Olson, D. H., (1993). Circumplex Model of Marital and Family Systems: Assessing family functioning (pp.
104-137). In F. Walsh (ed.), Normal family processes (2nd ed.). New York: Guilford Press.
39. Olson, D. H., Sprenkle, D. H. and Russell, C. S., (1979) Circumplex Model of Marital and Family Systems: I.
Cohesion and adaptability dimensions, family types, and clinical applications. Family Process, 18, 3-28.
40. Peplau, L. A., (1991). Lesbian and gay relationships (pp. 177-196). In J.C. Gonsiorek & J.D. Weinrich (eds.),
Homosexuality: Research implications for public policy. Newbury Park CA: Sage Publications.
41. Rasku-Puttonen, H., Lyytinen, P., Poikkeus, A.-M., Laakso, M.-L. and Ahonen, T., (1994) Communication
deviances and clarity among the mothers of normally achieving and learning-disabled boys. Family Process, 33,
71-80.
42. Ratnam, U., (1994). Whole family functioning and parental expressed emotion in inner-city African-American
families: Correlates of children's academic achievement and psycho-social functioning. Unpublished doctoral
dissertation, California School of Professional Psychology, Alameda.
43. Reiss, D., (1971a) Varieties of consensual experience: I. A theory for relating family interaction to individual
thinking. Family Process, 10, 1-28.
44. Reiss, D., (1971b) Varieties of consensual experience: II. Dimensions of a family's experience of its environment.
Family Process, 10, 28-35.
45. Russell, C. S., (1989) Multiple perspectives on normal family functioning. Journal of Family Psychology, 2,
299-302.
46. Singer, M. T., Wynne, L. C. and Toohey, M. L., (1978). Communication disorders and the families of
schizophrenics (pp. 499-511). In L.C. Wynne, R.L. Cromwell, & S. Matthysse (eds.), The nature of
schizophrenia: New approaches to research and treatment. New York: John Wiley & Sons.
47. Stierlin, H., (1959) The adaptation to the "stronger" person's reality: Some aspects of the symbiotic relationship of
the schizophrenic. Psychiatry, 22, 143-152.
48. Stierlin, H., (1973) A family perspective on adolescent runaways. Archives of General Psychiatry, 29, 46-62.
49. Surrey, J., (1985). Self-in-relation: A theory of women's development. Work in Progress, No. 13. Wellesley MA:
Stone Center Working Papers Series.
50. Vogel, E. F. and Bell, N. W., (1968). The emotionally disturbed child as the family scapegoat (pp. 412-427). In
N.W. Bell & E.F. Vogel (eds.), A modern introduction to the family (rev. ed.). New York: Free Press.
51. Walters, M., Carter, B., Papp, P. and Silverstein, O., (1988). The invisible web: Gender patterns in family
relationships. New York: Guilford Press.
52. Weingarten, K., (1991) The discourses of intimacy: Adding a social constructionist and feminist view. Family
Process, 30, 285-305.
53. Werner, P. D. and Green, R.-J., (1993). Preliminary manual: California Inventory for Family Assessment.
13
_____________________________________________________________________________________________________________
Systems model of family functioning also describe families that seem to fit the "Low closeness-caregiving/high intursiveness"
pattern in quadrant IV.Bowen (1966) suggested that fusion could be manifested either by emotional distance, emotional reactivity,
or alternations between the two, essentially viewing both low closeness-caregiving and high intrusiveness as signs of one common
underlying process (Karpel, 1976). Beavers and Hampson (1993) have hypothesized that a low score on their health/competence
dimension (which seems to indicate low intrusiveness) and a centifugal style (which indicates low closeness) contribute together to
the development of sociopathy.
2For readers unfamiliar with different communication concepts, we wish to emphasize that openness of communication as used
here is different from the dimension communication clarity versus deviance mentioned elsewhere in this article. We mean
"openness" to refer to specific topics of communication (disclosure of intimate personal feelings, important facts about the self, and
directness in discussing conflicts). By contrast "communication clarity versus deviance" pertains to styles of sharing focal attention
and conveying a coherent meaning regardless of the topic (that is, the extent to which a given speech is understandable or
"followable" versus ambiguous, fragmented, or otherwise confusing to the listener). As an example, "giving someone instructions
about how to drive from New York to Philadelphia" does not involve openness, but one could be very clear or very unclear in
communicating the directions. See Singer et al. (1978) and Green (1995) for more about communication deviances.
3In earlier reports of the CIFA (Werner & Green, 1993), we used the labels "Cohesiveness," "Expressiveness," and
"Intrrusive-Enmeshment" to refer to the tree domains that CIFA assesses. We now prefer the labels "Closeness-Caregiving,"
"Openness of Communication," and "Intrusiveness" because these terms are less contaminated by prior, inconsistent usage in the
literature.
4From
P.D. Werner, & R.-J. Green (1993). Preliminary manual: California Inventory for Family Assessment (CIFA). Reprinted
by permission of the authors.
APPENDIX
California Inventory for Family Assessment (CIFA): Definitions of 13 Constructs and Their Primary
Grouping into Three Domains4
CLOSENESS-CAREGIVING
1. Warmth: The degree to which Person A shows warmth, kindness, acceptance, caring, friendliness, love, and positive
regard toward Person B.
2. Time Together: The degree to which Person A seems to enjoy, puts a high priority on, seeks to spend time with, and
gives attention to Person B.
3. Nurturance: The degree to which Person A extends emotional comfort, help, caretaking, and emotional support to
Person B.
4. Physical Intimacy: The degree to which Person A hugs, holds hands, cuddles with, kisses, seeks physical closeness to,
and enjoys touching and being affectionate with Person B.
14
_____________________________________________________________________________________________________________
5. Consistency: The degree to which Person A's behavior toward Person B is consistent, clear, predictable, and
emotionally constant from day to day (as opposed to emotionally labile, inconsistent, confusing, unpredictable, changeable,
ambivalent).
OPENNESS OF COMMUNICATION
6. Openness/Self-disclosure: The degree to which Person A is open, self-revealing, honest, direct, and forthright with
Person B (as opposed to keeping self hidden, guarded, closed, or secretive).
7. Conflict Avoidance: The degree to which Person A tries to avoid facing conflict, denies that there are differences of
opinion or needs, evades discussing sources of tension (for example, by changing the topic), pretends to agree or pretends
that everything is harmonious, and generally tries to "sweep difficulties under the rug" in their relationship.
INTRUSIVENESS
8. Separation Anxiety: The degree to which Person A acts uncomfortable, upset, anxious, worried, hurt, or left out when
Person B wants to spend time alone, have privacy, or spend time doing things independently.
9. Possessiveness/Jealousy: The degree to which Person A acts possessive, insecure, neglected, threatened, jealous, or
afraid of losing to others Person B's attention.
10. Emotional Inter-Reactivity: The degree to which Person A acts overidentified with, and overreacts emotionally to,
problems in Person B's life (for example, Person A may worry excessively or become more upset than Person B when B is
sad, anxious about something, going through a hard time, etc.).
11. Projective Mystification: The degree to which Person A erroneously assumes that he/she knows what is on Person
B's mind without checking it out first ("mindreading," projection of feelings or motives), especially the degree to which
Person A presumes that he/she knows better and interprets what B really needs, thinks, feels, and why B acts in various
ways.
12. Anger/Aggression: The degree to which Person A gets angry, is critical, gets annoyed, uses harsh words, says hurtful
things, raises his/her voice, and acts in an aggressive/hurtful way toward Person B.
13. Authority/Dominance: The degree to which Person A is dominant and more in charge of the relationship in terms of
having a final say over how things are done, taking the lead, getting his/her own way when disagreements arise, setting the
rules for Person B's behavior, taking the initiative, and making decisions for Person B or for what they will do together.
15