GeistHaus

Personality / Styles

Dependent Personality Style

A dependent personality is organized around connection. People with this style feel most like themselves, and most secure, when an important relationship is going well. The PDM-3 puts the theme in a single imagined sentence: “I am fine as long as everything is OK with my husband.” Their sense of well-being rests on being connected, and the price, when the pattern is strong, is paid in personal agency and growth (McWilliams & Shedler, 2026, pp. 672–673).

That theme has an obvious upside. People with a dependent style tend to be attentive to what others need; the PDM-3 notes how readily they try to “read” and meet the needs of people who matter to them (McWilliams & Shedler, 2026, p. 673). The psychologist Robert Bornstein, whose research the PDM-3 describes as the only comprehensive empirical examination of problematic dependency its authors know of (McWilliams & Shedler, 2026, p. 673), reports that dependency is associated with “cooperativeness and conscientiousness” in patients, who tend to keep appointments and complete treatment more often than others. He adds that recent research suggests that expressing dependency in a flexible, situation-appropriate way “can strengthen interpersonal ties and facilitate adaptation” (Bornstein, 2007).

It also helps to remember that dependency itself is not a flaw. Nancy McWilliams writes that, “Western myths about human independence notwithstanding, we all need each other in both emotional and practical ways throughout the lifespan” (McWilliams, 1999, p. 24). What marks the dependent style is not the need, which everyone has, but how central it is, and how much of the self it crowds out.

Everyone has a personality style, and most people have a blend. This page describes a pattern, not a verdict on anyone.

Style, Disorder, and the Space Between

The PDM-3 finds people who are characterologically overdependent at every level of organization (McWilliams & Shedler, 2026, p. 673). Bornstein describes a continuum that runs from submissive overdependence, through healthy interdependence, to inflexible independence, which in some people masks powerful dependent longings (McWilliams & Shedler, 2026, p. 674). The style can show up anywhere along that line:

Where does a style become a disorder? Summarizing DSM-IV-TR, Bornstein describes dependent personality disorder as “long-standing, inflexible dependency that creates difficulties in social, sexual, and occupational functioning” (Bornstein, 2007). The useful markers are the usual ones. Is the reliance on others rigid, or can it flex? Does it show up in every setting, or only in some? Is the person suffering, or limiting their life, because of it? Can they step back and notice the pattern? Bornstein’s review of the research found that, “contrary to clinical lore,” high trait dependency and the disorder are linked to elevated risk of physical illness, functional impairment, higher health care costs, and harm within close relationships (Bornstein, 2012b).

Two cautions apply. The PDM-3 reminds clinicians that cultures differ in how much dependency they consider “too much,” so any diagnosis should take cultural context into account (McWilliams & Shedler, 2026, p. 673). And in later life, real losses and illness can make anyone more dependent. That is not a personality disorder, and the PDM-3 asks clinicians to look for evidence of the pattern earlier in life before calling it one (Del Corno & Plotkin, 2026, pp. 932–933).

In the official manuals, the American Psychiatric Association still lists dependent personality disorder among the ten personality disorders of DSM-5-TR (American Psychiatric Association, 2024), where it belongs to Cluster C along with avoidant and obsessive–compulsive personality disorders (Lampe & Malhi, 2018). The DSM’s alternative, dimensional model leaves it out of the six disorders it keeps (Lingiardi & McWilliams, 2026a, p. 7). ICD-11 has dropped the named personality types: clinicians rate overall severity and then note trait qualifiers (plus an optional borderline pattern qualifier), and what used to be the ICD-10 dependent type maps mainly onto the trait of negative affectivity (Bach & First, 2018). The passive–aggressive variant described below was once an official disorder but was moved to the DSM-IV appendix in 1994 (Hopwood et al., 2009). Counterdependency has never had a diagnosis.

Inner World

The central tension is between security and self-direction. Keeping a protective relationship feels essential, and acting independently can feel like risking it. Bornstein’s model breaks dependency into four parts: a strong need for guidance, support, and approval; a view of oneself as “powerless and ineffectual”; anxiety when one has to function autonomously; and a range of strategies for strengthening ties to potential caregivers (Bornstein, 2012a).

The PDM-3 describes people with this style as tending to feel inadequate when left to themselves and to see others as powerful and effective. Common worries include fear of performing badly, of being criticized, and of being abandoned. They may feel weak, find it hard to assert themselves, and be easily swayed. Anger is especially difficult: many struggle to express it or to negotiate a compromise that also serves their own interests (McWilliams & Shedler, 2026, p. 673).

Temperament probably plays a part. In adolescents with emerging dependent–victimized patterns, the PDM-3 notes low novelty seeking together with high harm avoidance and reward dependence (Tanzilli & Sharp, 2026, p. 495). Dependent people are not always passive, though. Bornstein found that some of their strategies for strengthening ties to potential caregivers are active, assertive, and even aggressive, including ingratiation, exemplification, self-promotion, and intimidation (Bornstein, 2007; McWilliams & Shedler, 2026, p. 673).

Self-esteem depends heavily on approval and closeness. When important others are pleased and present, things feel fine; when they withdraw, the person can feel worthless. In the more troubled forms, a devalued image of the self can sit beside an unrealistically positive, even all-powerful, image of others (Tanzilli & Sharp, 2026, p. 495). A pathogenic belief, in the language of case formulation, might be: I cannot manage on my own, and if I show anger or need too much, I will be left.

Not all dependency is conscious. Bornstein distinguishes dependency needs a person openly acknowledges from implicit ones that shape behavior without awareness. In one study, dependent and histrionic personality traits were both linked to high implicit dependency, but only dependent traits were linked to high acknowledged dependency (Bornstein, 1998).

Defenses

The defenses of a dependent style mostly protect relationships, often at the expense of the person’s own wishes.

On the healthy side, the same orientation supports affiliation, the mature capacity to turn to others for help and support. For many people with this style, the developmental task is to add self-assertion to it.

Relationships

People with this style tend to organize their lives around nurturing relationships in which they take the submissive role. They define themselves largely through others and look to relationships for both security and satisfaction. When such relationships are in place they can be content. When they are missing or threatened, distress can be acute (McWilliams & Shedler, 2026, pp. 672–673). Symptoms often appear when something goes wrong in a primary attachment (McWilliams & Shedler, 2026, p. 673).

In the dependent–victimized pattern, the person is drawn to relationships in which they are mistreated, becomes attached to people who are emotionally unavailable, and often takes the role of caring for or rescuing the other person. Unspoken resentment may leak out in indirect ways that can provoke anger or mistreatment from others (McWilliams & Shedler, 2026, pp. 673, 675). The passive–aggressive variant defines itself against others rather than through them, and gets caught in a cycle in which indirect hostility draws retaliation, which confirms the conviction that anger must never be shown directly (McWilliams & Shedler, 2026, p. 674).

How do others feel? Clinicians describe their own feelings as typically benign at first, then increasingly irritated and burdened (McWilliams & Shedler, 2026, p. 673). The PDM-3 describes similar strain in caregivers of dependent older adults, whose repeated requests for attention can lead to fatigue and burnout. In later life the dependency can also take on a grumpy, oppositional tone, with angry isolation or refusal of help, and older people who have always fended off their dependency may deny their impairments and refuse help with daily care (Del Corno & Plotkin, 2026, p. 934). With counterdependent people the problem is the reverse: partners may feel “starved for genuine emotional intimacy” (McWilliams & Shedler, 2026, p. 675).

As for origins, the theories are about influences, not blame. Bornstein’s research points to overprotective or authoritarian parenting, gender-role socialization, and cultural attitudes about achievement versus relatedness, in any combination (McWilliams & Shedler, 2026, p. 673). The PDM-3’s adolescent chapter adds families that teach constant attention to others’ needs, families that discourage individuation and reward obedience, and family climates associated with anxious–ambivalent or disorganized attachment (Tanzilli & Sharp, 2026, p. 495). In Bartholomew and Horowitz’s adult attachment model, which combines a positive or negative image of the self with a positive or negative image of others (Bartholomew & Horowitz, 1991), the “preoccupied” pattern pairs a negative model of the self with a positive model of others (Forsythe et al., 2012). That combination, in this site’s reading, closely resembles the dependent picture.

In Therapy

Transference is the term for the feelings and expectations from earlier relationships that a person brings to the therapist. Countertransference is the therapist’s own emotional response. Both can be used as information.

People with a dependent style often start therapy readily and can be “compliant to a fault.” They tend to idealize their therapists, ask for advice, and want reassurance that they are good patients. Some try to read and meet the therapist’s needs, and some keep offering favors or gifts even after boundaries have been explained (McWilliams & Shedler, 2026, p. 673).

Therapists’ feelings usually start out warm and gradually become more irritated and burdened. The PDM-3 suggests that patients may unconsciously test whether the therapist will support their first steps toward autonomy or will enjoy being cast as the expert. The recommended stance is to decline the expert role, encourage autonomous functioning, and help the person contain the anxiety that this stirs up. Therapists may be tempted to keep things pleasant, but inviting anger and other aggressive feelings into the room can help the person develop a sense of agency and pride (McWilliams & Shedler, 2026, pp. 673–674). McWilliams makes a related point about limits: “Most overtly clingy, dependent patients have an equally strong covert need to express anger and oppositionality.” Reasonable limits on availability, she argues, serve them better than an over-caretaking response that infantilizes them (McWilliams, 2004, p. 126). Bornstein likewise recommends firm limits on after-hours contact early in treatment, and names two common countertransference fantasies: that no amount of support will ever be enough, and that the patient will never leave therapy (Bornstein, 2007).

With adolescents, the PDM-3 describes a similar arc. Young people with dependent patterns tend at first to idealize the therapist as a powerful rescuer and to seek reassurance and approval; when their needs are not met, they may react with resentment or shame and experience the therapist as cold or rejecting. Therapists’ early compassion and protectiveness can give way to a sense of burden, and with dependent–victimized adolescents, to helplessness when every effort to help is defeated. Both psychodynamic and cognitive-behavioral treatments aimed at greater autonomy may help (Tanzilli & Sharp, 2026, pp. 496–497).

The goal is not to cure dependency. In McWilliams’s words, “Psychotherapy does not take dependent people and make them independent; rather, it makes them capable of handling their natural dependency in their best interests” (McWilliams, 1999, p. 24).

The variants call for adjustments. With passive–aggressive patients, the PDM-3 recommends a sense of humor, not colluding with the denial of aggression, side-stepping power struggles, and not seeming too invested in the patient’s progress, since that can trigger opposition. Such patients often need help naming their angry feelings, telling thoughts and feelings apart from actions, and seeing the price they pay for indirect hostility (McWilliams & Shedler, 2026, p. 674). Counterdependent people seldom seek help on their own and are often brought in by partners who feel starved for intimacy. They first need help accepting their dependent needs as a normal part of being human. Therapists who tolerate their protestations of independence long enough to build an alliance report a period of mourning for unmet early needs, followed by more genuine autonomy (McWilliams & Shedler, 2026, p. 675). Because becoming attached feels humiliating to them, counterdependent clients may flee once an attachment forms; McWilliams suggests that simply acknowledging how much courage it takes to stay can help them stay (McWilliams, 2011, p. 16). When a counterdependent client finally risks asking for something, that request may itself be a sign of growth (McWilliams, 2004, p. 106).

On medication, Bornstein reports that no class of drugs has been consistently more effective than placebo for the symptoms of dependent personality disorder, while psychotherapies of several kinds help modestly, and combined approaches help most (Bornstein, 2007). In older adults who are also depressed, the PDM-3 notes that recognizing the dependent pattern can prevent harm from treating the depression alone (Del Corno & Plotkin, 2026, p. 934), and that the task is to balance sensitivity to the person’s requests with protecting their realistic areas of autonomy (Del Corno & Plotkin, 2026, p. 935).

Variations

The PDC-3, the PDM-3’s rating form, lists the dependent style with two subtypes, passive–aggressive and counterdependent (Gordon & Bornstein, 2025). The PDM-3 text adds a third, research-derived form.

In older adults, dependency and a passive–aggressive counterdependency can alternate: worsening health can heighten the need for support, while conflicts with family or staff can bring out irritability and withdrawal (Del Corno & Plotkin, 2026, pp. 934–935).

Common blends include the depressive style (especially its “anaclitic” form), the masochistic style, the hysterical–histrionic style, and the anxious–avoidant style. One study reviewed by Lampe and Malhi found avoidant personality disorder associated with dependent, schizoid, and paranoid diagnoses (Lampe & Malhi, 2018).

How Different Thinkers See It

Karl Abraham: the oral character

In 1924 Abraham proposed that early experiences of feeding and sucking leave traces in character. Some people who had a deeply gratifying start, he wrote, carry an “imperturbable optimism”; but others “are dominated by the belief that there will always be some kind person—a representative of the mother, of course—to care for them,” a belief that “condemns them to inactivity.” He traced a different, more insistent pattern to early frustration (Abraham, 1927, pp. 399–401). Few clinicians now hold to the feeding theory. Discussing depressive personalities, McWilliams notes that the idea of oral fixation has lingered more for its intuitive appeal than for its theoretical standing (McWilliams, 2011, p. 238).

Erikson, Fairbairn, and the relational turn

Erik Erikson recast Freud’s oral stage as a stage of total dependency in which basic trust, or mistrust, is at stake (McWilliams, 2011, p. 24). Object relations theorists went further. McWilliams describes how, starting with Fairbairn’s argument that infants seek relationship rather than drive satisfaction, and then with Bowlby’s attachment research, analysts came to treat dependency as a basic motive in its own right, alongside sex and aggression. Self psychologists add that the need for people who mirror and validate us is lifelong (McWilliams, 1999, p. 24). In this view, the question is not whether someone is dependent but whether their dependency is infantile or mature (McWilliams, 1999, p. 24).

Karen Horney: moving toward people

As later reviewers summarize her work, Horney described three neurotic trends or interpersonal coping styles: moving toward others (compliance), moving against them (aggression), and moving away from them (detachment) (Coolidge & Segal, 1998; Coolidge et al., 2001). In this site’s reading, the compliant trend is the clearest ancestor of the dependent style. Coolidge and colleagues built a questionnaire around her typology and found that it correlated meaningfully with DSM personality disorder features (Coolidge et al., 2001).

Sidney Blatt: anaclitic and introjective

Blatt distinguished depressive experiences centered on “I’m empty, I’m hungry, I’m lonely, I need a connection” (anaclitic, from the Greek for “to lean on”) from those centered on self-criticism (introjective). He later renamed the two poles “self-in-relationship” and “self-definition,” and McWilliams adds that everyone has both needs and that some balance between them is part of mental health. She writes that in her book’s first edition she had implicitly treated the anaclitic version as a dependent personality style (McWilliams, 2011, p. 237). One way to read the dependent style in this framework is as relatedness developed at the expense of self-definition.

Robert Bornstein: dependency as adaptive and maladaptive

Bornstein’s research reframed dependency as something that can be healthy, harmful, or defensively denied. His cognitive/interactionist model pairs a view of the self as powerless and ineffectual with a marked need for guidance, support, and approval. Dependency can involve passive or quite active behavior, and in some settings, such as fostering compliance with medical and psychotherapeutic treatment, it is adaptive (Bornstein, 2012a). His work on gender found that women score higher on self-report dependency tests at every age, while adult men score slightly higher on projective tests (Bornstein, 1995). This supports the PDM-3’s point that women may simply be more willing to express dependent longings openly (McWilliams & Shedler, 2026, p. 673).

Attachment research

The PDM-3 links the dependent pattern to anxious–ambivalent attachment, in which caregivers respond inconsistently and confusingly to a baby’s bids for closeness (Tanzilli & Sharp, 2026, p. 495), and suggests that counterdependent adults would likely have been classified as avoidant in childhood (McWilliams & Shedler, 2026, p. 675). Read together, these suggest that dependency and counterdependency may be two different answers to uncertainty about whether care will be there.

Feminist critiques

The PDM-3 notes that feminist psychoanalytic scholars, including Benjamin, Chodorow, and Gilligan, have criticized Freudian theories of female sexuality for normalizing the devaluation of femininity and of a relational orientation, and argue that this bias reinforced gender stereotypes (McWilliams & Shedler, 2026, p. 673). McWilliams cites a related psychoanalytic observation, drawn from Chodorow and Dinnerstein: because primary caregivers are usually women, and girls identify with them while boys tend to “disidentify,” women may be more prone to difficulties of too much attachment (she names dependent personality among them) and men to difficulties of too little. She adds that, when feeling empty, men may be more likely to deny need and act counterdependently (McWilliams, 2011, pp. 202–203, 248). These are hypotheses about socialization, not facts about sex.

The DSM, ICD, and trait models

The manuals treat dependency as a cluster of observable criteria (DSM-5-TR) or as an aspect of the trait domain of negative affectivity (ICD-11), not as a pattern of meaning (American Psychiatric Association, 2024; Bach & First, 2018). The PDM-3 criticizes the DSM tradition for describing personality styles only in their most extreme forms, which has taught many clinicians to equate any reference to personality with pathology (McWilliams & Shedler, 2026, p. 653).

Examples

“Tell me what I ought to do” (fiction)

In Jane Austen’s Emma, the young Harriet Smith receives a proposal from the farmer Robert Martin, a man she likes. She brings the letter to her patron, Emma Woodhouse, and asks again and again what she should do: “Pray, dear Miss Woodhouse, tell me what I ought to do.” Emma insists she will not advise her, while steering her toward refusal, and Harriet refuses. The scene shows how a suggestible, approval-seeking person can hand a decision to someone she admires. Austen also gives Harriet a better ending: by the close of the novel she marries Robert Martin after all, because, as Emma finally admits, Harriet “had always liked” him.

Jane Austen, Emma (1815), Vol. I, Ch. 7; Vol. III, Ch. 19

The kind person who will provide

Karl Abraham described a type of person, which he traced to an especially gratifying infancy, who believes that someone, “a representative of the mother, of course,” will always take care of them. That optimism can help them reach their goals, but in its less favorable forms it leaves them waiting rather than acting. Abraham set this against a type marked by early frustration, who keeps asking for things, by modest request or aggressive demand, and dislikes being alone even briefly.

Karl Abraham, Selected Papers of Karl Abraham (1927), pp. 399–401

Molly

McWilliams describes the psychoanalytic therapy of Molly, a nurse with impressive ego strength whose personality was mainly masochistic but who also had “significant counterdependent tendencies.” She had been raised to put others first and had faced a stark choice between staying in her family and giving up autonomy, or cutting ties and sacrificing her need for others. McWilliams saw the main task as helping Molly accept her longings for closeness as normal rather than as defects of character. A month after therapy ended, Molly wrote that she had “learned to accept dependence on Steve” as an integral part of the relationship.

Nancy McWilliams, Psychoanalytic Psychotherapy (2004), pp. 197, 201, 204–206, 216

How It Differs From Look-Alikes

Depressive style (anaclitic form)
Both center on longing for connection. The PDM-3 notes that dependent adolescents cope with sad feelings by attaching themselves to others and seeking reassurance, while those with anaclitic depressive personalities keep their attachment longings private (Tanzilli & Sharp, 2026, pp. 483–484).
Masochistic style
McWilliams notes that masochistic people with intense anaclitic needs are sometimes called relational masochists, whose self-defeating actions come from efforts to keep an attachment at any cost (McWilliams, 2011, p. 270). The overlap with the dependent–victimized form is large; the PDM-3 sees features of relational masochism in dependent–victimized adolescents, tied to a belief that all relationships involve sacrifice and pain (Tanzilli & Sharp, 2026, p. 495). The difference is emphasis. In masochistic dynamics, suffering itself carries meaning: the PDM-3 describes a conviction that suffering is the best or only route to care (McWilliams & Shedler, 2026, p. 672), and McWilliams describes moral masochists who organize their self-esteem around their capacity to tolerate pain (McWilliams, 2011, p. 270). In dependent dynamics, the relationship itself is the point.
Hysterical–histrionic style
People with hysterical tendencies also lean in the anaclitic direction and move toward people rather than away (McWilliams, 2011, p. 313), and at more troubled levels they may become “clingily dependent” (McWilliams, 2011, p. 317). But beneath the surface they are preoccupied with gender, sexuality, and power, and they tend to seek power through seductiveness toward those they see as dominant (McWilliams & Shedler, 2026, pp. 685–686). Bornstein found that both styles involve strong unconscious dependency, but only the dependent style involves openly acknowledged dependency (Bornstein, 1998).
Anxious–avoidant style
Both fear criticism and rejection, and the PDM-3 notes that anxious people, too, try to elicit protection from those they see as powerful (McWilliams & Shedler, 2026, p. 677). The difference is in what organizes the personality: in the anxious style, fear and the avoidance of perceived danger (McWilliams & Shedler, 2026, p. 678); in the dependent style, keeping a nurturing relationship (McWilliams & Shedler, 2026, pp. 672–673).
Realistic dependency
Illness, disability, grief, and aging create genuine need. The PDM-3 cautions that regression into dependency under stress in older adults may be “a natural response” rather than a personality pattern, and that the question is whether dependency is long-standing or emerged later in life (Del Corno & Plotkin, 2026, pp. 932–933). Cultural norms about interdependence matter too (McWilliams & Shedler, 2026, p. 673).

History

The psychoanalytic history begins with Karl Abraham’s 1924 paper on the influence of oral erotism on character formation, which described character traits rooted in early experiences of being fed and cared for (Abraham, 1927, pp. 393–406). Erik Erikson later reframed the oral stage as a question of basic trust (McWilliams, 2011, p. 24), and object relations and attachment theorists made dependency a basic human motive (McWilliams, 1999, p. 24).

In psychiatry, the first DSM (1952) did not have a separate dependent personality. It placed a “passive-dependent type,” described as helpless, overly dependent, and indecisive, as a subtype of “passive-aggressive personality,” alongside passive–aggressive and aggressive subtypes (Coolidge & Segal, 1998). The term “passive aggressive” itself came from the American military, where it described soldiers who did not comply with superiors’ commands (Hopwood et al., 2009). DSM-II (1968) kept passive–aggressive personality but had no dependent category (Coolidge & Segal, 1998). Dependent personality disorder entered with DSM-III in 1980, as part of an “anxious or fearful” cluster (Coolidge & Segal, 1998). In DSM-IV (1994), passive–aggressive personality disorder was moved to the appendix for further study, with criteria reworked around Millon’s broader “negativistic” pattern (Hopwood et al., 2009); Coolidge and Segal (1998) cite poor reliability and questionable validity. Some researchers have since questioned that decision (Hopwood et al., 2009).

The dependent diagnosis is given more often to women (Bornstein, 2007), which raises the question of what that difference means. The PDM-3 points to gender-role socialization as one influence, and to Bornstein’s view that women may simply be more willing to express dependent longings openly (McWilliams & Shedler, 2026, p. 673). Bornstein’s meta-analysis found that the sex difference reverses on projective tests in adulthood (Bornstein, 1995). As DSM-5 was being prepared, Bornstein argued that the research on its costs, including physical illness, impairment, and harm in close relationships, meant it should be included (Bornstein, 2012b). DSM-5 kept it in its main classification (American Psychiatric Association, 2024) but left it out of the alternative dimensional model (Lingiardi & McWilliams, 2026a, p. 7), and ICD-11 replaced the named personality types with severity ratings and trait qualifiers (Bach & First, 2018).

Meanwhile, psychodynamic diagnosis moved in the other direction. McWilliams covered dependency only briefly in her 2011 book, referring readers who wanted more on dependent, counterdependent, and passive–aggressive people to the PDM (McWilliams, 2011, p. xiii). The PDM-3 now describes a dependent style with dependent–victimized, passive–aggressive, and counterdependent forms, drawing on both clinical tradition and SWAP research (McWilliams & Shedler, 2026, pp. 672–675), and gives developmental versions for adolescents and older adults (Tanzilli & Sharp, 2026, pp. 494–497; Del Corno & Plotkin, 2026, pp. 932–935).

Sources