You have a model of your partner in your head. It is a compressed representation of who they are, what they like, how they will react, what they think about politics or money or in-laws, what they want from you, what they cannot stand. The model is necessary; you cannot run a relationship without one. It is also, almost certainly, out of date. The person you are married to today is not exactly the person your model predicts. Some of the divergence is small. Some of it is large enough that you have been having arguments for years that are arguments with a ghost — with the partner you used to have, not the one in the room.
This article is about updating the model. It is a Law 5 article — Revise — applied to the inner representation each partner carries of the other. The case for the update is simple: the partner has been changing while you have been busy. Their values have shifted, their preferences have evolved, their fears have moved, their hopes have rearranged themselves. Most of this has happened slowly, without announcement. You have probably not been tracking it carefully because tracking it carefully requires asking, and asking requires admitting that you do not already know.
There is a particular cognitive trap in long relationships. The trap is that the partner becomes one of the most familiar objects in your environment, and familiar objects get represented heuristically rather than freshly. You stop looking. You operate on the cached model. The cached model was accurate in 2009, less accurate in 2016, badly out of date now. Meanwhile your partner has been doing the same to you, and both of you are reacting to outdated simulations of each other across the kitchen table.
Adam Phillips writes that the people we love are partly imagined — we relate to a version of them constructed inside ourselves, and the construction is always at some distance from the real other. This is unavoidable. The question is how big the distance is and whether it is growing or shrinking. In a healthy relationship, the distance shrinks over time as the model gets refined. In a stagnant relationship, the distance grows as both partners drift while the models stay frozen. You can tell which kind of relationship you are in by noticing how often you are surprised, in good faith, by your partner. Frequent surprise is a sign the model is being updated. Zero surprise — total predictability — is a sign the model has ossified, and possibly that you have stopped looking.
The update is not a single conversation. It is a posture: an ongoing curiosity about the actual person across from you, including a willingness to be wrong about them. The curiosity has to be visible to them. People who are loved by curious partners tend to keep growing and to share that growth. People who are loved by partners with frozen models tend to perform the frozen version, because there is no audience for the new one. The relationship can keep someone small or invite them to be large; the difference is largely in whether the model in the partner's head has room for largeness.
Karen Fingerman's research on long-term close relationships consistently finds that satisfaction tracks with whether partners feel known and feel responded to in their current state. Feeling known by a model that is fifteen years old is not feeling known; it is feeling categorized. The work of being known requires being asked. The work of knowing requires asking. Most long-married couples report a fall-off in mutual asking after about the first decade; the model freezes around then for many couples, with predictable consequences.
There is a particular danger of negative ossification. Some partners' models of each other are not just out of date but unfairly negative — fixed on a flaw, a tendency, a failure from years ago, treated as permanent character. Every new instance is read through the old frame. The partner senses the unfair frame and either gives up on changing it or stops being open with the person who is reading them through it. The relationship hardens. The fix is to consciously notice when you are about to interpret a partner's behavior through an old frame, and to ask whether the frame is still earning its place.
There is a parallel danger of positive ossification — clinging to an idealized model of who the partner used to be, or who you needed them to be, while the actual partner has evolved away from it. This is sometimes called "I miss the man I married" or "she's not the person she used to be." Sometimes this is a real loss to grieve. More often it is a refusal to update — an insistence on a model that has expired. The partner cannot win against an idealized predecessor; the relationship cannot grow if one of the parties is being compared to a static reference.
Esther Perel writes about the eroticism of the partner remaining partly unknown — the way desire requires some space, some sense of mystery. The mystery is not preserved by not knowing your partner. It is preserved by knowing them currently rather than historically, with the recognition that the current them will keep changing. Frozen knowledge is the opposite of mystery; it is the wax figure of a relationship. Live knowledge — provisional, updating, curious — keeps the partner real and partly new.
The update is, in the end, the form love takes over a long period: a continued willingness to find out who this person is, today, including the parts that have changed since yesterday. This is not exotic. It is the basic discipline of staying in actual relationship with an actual other, rather than with a memory of them.