A man who cancels his weekend plans to sit with his sick partner is devoted. A woman who does the same is needy, or codependent, or has attachment issues. This double standard isn't just a social annoyance; it shapes how people interpret their own experience, which means it shapes what they do about it. A woman who feels she cannot function without her partner may hear that as confirmation of a character defect rather than as a signal about what the relationship is actually providing or withholding. The framing matters because it determines whether someone examines the relationship or examines herself.

The term codependency is not a diagnosis in the DSM-5. It is a concept that has migrated from clinical observation through pop psychology into moral judgment, and along the way it has accumulated contradictory definitions. An integrative review by Gonzalez Molina, Taiwo, and Grey published in Clinical Psychology and Psychotherapy in 2026 examines the landscape of how codependency has been conceptualized and what mental health outcomes are actually associated with it. The fact that a peer-reviewed integrative review in 2026 is still sorting out what the term means and what it predicts is itself informative. It tells you that the field has not converged on a stable definition, which means any confident pop-psychology claim about what codependency is should be treated as one interpretation among several, not as settled science.

Attachment theory, which originated in the work of John Bowlby and Mary Ainsworth, offers a more precise framework for understanding why some people seek more reassurance than others. Inconsistent caregiving in early development can produce what is now called an anxious attachment pattern, which in adulthood looks like heightened sensitivity to a partner's availability, a need for frequent confirmation of the relationship's stability, and difficulty tolerating ambiguity. This is not a moral failing. It is a regulatory strategy that served a function when you were three years old and your primary caregiver was unpredictable. The problem is not that the strategy exists; the problem is when it becomes the only tool available and the relationship becomes the sole source of emotional regulation.

The empirical work on this question is developing, and it is worth noting where it is pointed. Ercengiz, Aka, and Şata published a study in Current Psychology in 2026 that used latent profile analysis to examine codependency profiles in married women, looking at how those profiles relate to attachment styles, marital mindfulness, emotional reactivity, and psychological resilience. The methodological choice here is significant. Latent profile analysis does not ask whether a person is codependent or not; it asks what configurations of traits co-occur and how they relate to other variables. This is a more granular approach than the binary framing that dominates public discourse, and the fact that it is being applied specifically to married women suggests the field is taking the gendered dimension seriously rather than treating it as a footnote.

Here is the distinction I want to draw, and it is not a matter of degree. The difference between interdependence and dependency is not that one person needs too much and the other person needs just the right amount. That framing implies there is a correct quantity of need, and it sets up a competition in which the winner is the person who needs least. The actual distinction is structural, and it has three components. First: the ability to leave. Not the willingness, not the intention, but the actual capacity. Do you have the financial means, the social connections, the housing options, the psychological infrastructure to walk out of this relationship and survive? A person who needs their partner daily, calls them multiple times a day, and feels lost without them can still be in a healthy interdependent relationship if that door is genuinely open. A person who is perfectly self-sufficient on paper but cannot imagine leaving because the relationship has systematically eliminated every outside connection is in a different position entirely.

Second: the ability to disagree. Can you say I don't agree with you, or I want something different, without the relationship becoming an emergency? Can you hold a position that differs from your partner's without feeling that the relationship is in danger? In a genuinely interdependent relationship, disagreement is a normal feature of two separate people making a shared life. In a dependency structure, disagreement is a threat because the dependent person's sense of self has been organized around the relationship's stability, and any challenge to that stability reads as an existential threat. The difference is not in the emotion felt during disagreement but in the structural relationship between the self and the dyad.

Third: the maintenance of a separate self. Not a performance of independence, not a curated Instagram feed of solo adventures, but an actual interior life. Interests that are pursued for their own sake. Friendships that are maintained through effort, not just through the partner's social circle. Opinions that are formed independently and can differ from the partner's. Ambitions that exist outside the relationship's narrative. A person who has all of these and still chooses closeness daily is in a fundamentally different position than a person who has none of them and experiences the relationship as the entire content of their identity. Both may describe themselves as loving, committed, and devoted. The internal architecture is different.

I want to be fair to the counterargument, because it is a real one. Not all closeness-seeking is healthy, and not all independence is healthy. Some people's need for reassurance is so intense that it becomes coercive: constantly demanding proof of love, punishing a partner's normal autonomy with escalation, making the partner's separate life a source of conflict rather than a background fact. Some people do lose themselves in relationships, and the loss is real and damaging. The Ercengiz et al. study examines emotional reactivity and psychological resilience as variables alongside codependency profiles in married women, which points toward the idea that the relevant distinction is not just how much connection you seek but how you respond when that connection is challenged or temporarily unavailable. A person who can tolerate a partner's no without the relationship collapsing is in a different position than one for whom that no is indistinguishable from abandonment. The line between anxious attachment and dependency is not always clean, and pretending it is clean does nobody a service.

There is also a specific relational template worth naming because it produces a pattern that looks like neediness but has a different origin. Dariotis, Chen, Park, Nowak, French, and Codamon published a mixed methods systematic literature review in the International Journal of Environmental Research and Public Health in 2023 examining parentification vulnerability, reactivity, resilience, and thriving. Parentification is the experience of being made emotionally responsible for a parent or caregiver before you are developmentally capable of bearing that responsibility. In the clinical literature, adults who carry this template can come to mistake the management of a partner's emotional state for love, and the partner's emotional stability for their own security. The need is not a sign of weakness; it is a learned pattern in which a person's sense of worth becomes contingent on another person's emotional regulation. This is a specific, identifiable pattern with a specific origin, and it responds to different interventions than generic anxiety about closeness.

The goal is not independence. That is its own distortion, and the culture applies it to women with particular cruelty. The strong independent woman script pathologizes any need, any vulnerability, any dependence on another person for emotional support, and it produces a specific kind of loneliness that is harder to name than the loneliness of dependency because it is experienced as strength. The goal is interdependence with agency: a relationship in which you choose closeness daily, in which your needs are met, in which you can disagree and survive the disagreement, and in which the door is always open. You walk through it only if you choose to. That is the test. Not how much you need. Whether you can still walk.