Why do I keep ending up in relationships that never become real relationships?

Can an Undefined Relationship Be a Healthy Choice for an Adult?

Yes, and for a good number of adults it plainly is one. The question sitting underneath that one is what most people actually arrive here carrying, so it is worth answering in the same breath. Is anything actually wrong with you?

Probably not, and this needs saying early rather than in a closing paragraph nobody reaches.

Two adults spending months in something they have not formally categorized is an entirely ordinary thing for human beings to do. It has been ordinary for a very long time. What is new is the word, the expectation that everything gets categorized quickly, and a set of platforms whose business depends on you believing that your arrangement is a problem with a product-shaped solution.

The Armstrong interviews are the useful corrective here, and they are the reason this page refuses to open with a diagnosis. Fifty-three women in their thirties, forties and fifties, outside stable partnerships, described ongoing involvements that sat deliberately between casual and committed. A majority had them. What the interviews found was that many were choosing the shape on purpose, because it preserved autonomy and a single identity they had lost inside more conventional arrangements, and because the alternatives on offer looked worse to them. The paper is honest about the drawbacks too, and it does not claim these arrangements suit everybody. It is interview work with a modest sample in one country, so it establishes that this exists and is deliberate rather than establishing how common it is.

That finding should change how you read the rest of this page. If the thing you have is one you would keep, at its current size, if you were told today that it would never grow, then you are in the first root below and there is nothing here for you to fix.

There is one honest exception worth flagging in advance. The reader worth worrying about is rarely the one in obvious distress about it. It is the one who has stopped noticing the count, has quietly organized eighteen months around something they describe with a shrug, and has not asked in a long time whether they would choose this if it were offered to them plainly.

What Makes a Conversation About This a Turning Point Rather Than a Formality?

Because a conversation about status is an event, and events change relationships in both directions.

This is worth establishing carefully, because it is the piece that makes your reluctance reasonable rather than cowardly. Communication researchers have studied what happens when something occurs that raises a person's uncertainty about a relationship. Planalp and Honeycutt asked people to recall such events and describe what followed. Two findings from that work are the useful ones. First, the effects tended to persist rather than pass, and a substantial share of the relationships were reported as having deteriorated afterward. Second, the most common response people described was doing nothing about it, and quietly reducing closeness instead.

Hold both halves. Raising the question is a genuine event with a genuine chance of a worse outcome. And saying nothing is also a move, with its own reliable result, which is a slow reduction in what the thing is.

The evidence has real limits and they should be stated. That work is retrospective recall, gathered from people describing events after the fact, in modest samples, in the 1980s. Retrospective accounts are shaped by how things turned out. It establishes a shape rather than a probability, and nobody can give you the odds for your specific Tuesday.

But the shape is enough to retire two pieces of standard advice.

Just have the conversation describes one available move while presenting its outcome as though it were known, which is what disqualifies it as advice. Whether it is the right move depends entirely on what the arrangement is doing for you, which is the subject of the next five sections.

Give it time is also not advice, for the opposite reason. Time does not accumulate into clarity. It accumulates into constraint, which is a different substance, and it makes every version of the conversation harder rather than easier, because a question that would have been ordinary in month two is an accusation in month nineteen.

Who Is Actually Keeping This Undefined?

Almost always both of you, and this is the least comfortable section on the page.

The popular version of this story has one person withholding and one person waiting. That version exists. It is well described, it has its own vocabulary, and the page on keeping somebody interested with no intention of choosing them handles it from the withholding chair, including the specific line that neither person could describe where they stand and one of them arranged it that way deliberately.

That is not the common case. The common case is two people, each finding the arrangement tolerable enough to postpone, each waiting for the other to raise it, and each performing a level of casualness they do not privately feel.

Your half is the only half you can do anything with, so here is what your half usually consists of. Read it as a list rather than as an accusation, because every item on it is small and reasonable in isolation.

• Not asking, on approximately two hundred separate occasions, each of which felt like the wrong moment for a specific reason.
• Saying you are keeping it light out loud, to a friend, in their hearing or in a way that gets back to them.
• Removing a question from a message before sending it, because a question invites an answer.
• Staying available at short notice, so the arrangement never has to be scheduled, because scheduling is the beginning of a structure.
• Never introducing them to anybody. Not once, in seven months, which is the single most effective way to prevent a name from forming.
• Matching their level of definition rather than setting your own, so that whatever they do not say, you also do not say.
• Answering a direct question from a friend with a shrug that you have practiced.

None of that is dishonest. All of it is protective. And all of it is yours.

There is a described model for the general version of this, and it is worth having. Sliding versus deciding distinguishes transitions people actively decide from transitions they slide into, and its central claim is that constraints accumulate faster than commitment does, so people end up progressively harder to leave without ever having chosen to stay. Stanley and Rhoades developed it mostly out of research on couples moving in together, where a lease and a shared sofa are visible constraints with dates on them.

The honesty note matters here. That model was built for a different arrangement, the cohabitation findings it grew out of have been substantially revised since, and nobody has tested it on undefined involvements. Treat it as a shape that fits rather than as a finding about you. What makes the shape fit is that your version has all of the constraint and none of the lease. Seven months of habit, a rhythm, a shorthand, a toothbrush, and a rearranged week are real constraints. They just have no paperwork, which means they accumulated without anybody being able to point at the day it happened.

Interview work on how couples enter shared arrangements finds much the same thing. Sassler's interviews with couples about how they came to live together found that a large share described no explicit conversation at all, and that both partners frequently reported waiting for the other one to bring it up. That is interview evidence about a different transition, gathered as recollection, so hold it loosely. It is worth having anyway, because it says the thing nobody wants to hear: two people, both waiting, is the normal case rather than the pathological one.

Which Part of This Was Drift, and Which Part Is Chosen?

The drift and the choice are different objects, and nearly all writing on this subject treats them as one thing.

The drift is not a decision and was never available to you as one. Nobody chose month four. There was no meeting at which the two of you agreed to leave this unnamed for the better part of a year. What happened is that a first month was reasonably undefined, and a second month resembled the first, and by the time the arrangement had a shape, the shape had arrived without anybody proposing it. That is how nearly all of these form, and treating it as a failure of character misdescribes what happened.

The choice is one question, and you are the only person it can be put to. Not the conversation with them. That is a separate matter, with real risk, and I am not going to prescribe it. The chosen part is this:

If you were told, reliably and today, that this will stay exactly as it is and will never become anything more, would you keep it?

Sit with that before answering, because the honest answer arrives about four seconds after the flattering one.

If the answer is yes, and it stays yes overnight, you have an arrangement rather than a problem, and the rest of this page is a curiosity for you.

If the answer is no, then what you have been doing for seven months is waiting, and the thing you have been calling keeping it light has been a strategy for waiting without having to admit you are waiting. That is worth knowing, and it does not yet tell you what to do, because what to do depends on which of the five roots below is producing it.

Notice what the question does not ask. It does not ask whether you love them, whether they are good for you, or whether they would say yes if you asked. Those are all unanswerable from where you are sitting. It asks about the only thing you have direct access to, which is what you would choose if the uncertainty were removed.

What If You Genuinely Wanted Something Light?

Most pages on this subject cannot accommodate this reader at all, and a good number of you are this reader.

In this version there is nothing underneath. You wanted an undefined arrangement, you got one, and it fits. Your week is full of other things you care about. You are not counting. When they are unavailable for two weeks, you notice and you are fine. The word situationship, applied to you, would be accurate and slightly insulting, in the way that any word carrying a verdict is when it lands on somebody who is not experiencing a problem.

The tells here are dull and reassuring. You could describe the arrangement to your closest friend in one sentence with no qualifier and no softening. You have declined nothing because of it. If somebody you were genuinely interested in appeared on Thursday, you would go, and you would not experience that as a betrayal of anything. The good evenings leave you steadier rather than more unsettled, which is the single most reliable indicator on this page.

The discomfort in this root is nearly always borrowed. It arrives from a friend who asks what you are doing with your life, from a parent counting, from a set of assumptions about what a person your age is supposed to be building. Somebody hands you a timetable you did not agree to and the failure to be on it registers as a personal problem. It is not one. Being asked what you are and having a true, cheerful, small answer is a perfectly good position to occupy for as long as it keeps being true.

One honest note attached to this root, because it is the one that goes wrong quietly. Being here today does not make you immune to the second root. Arrangements that started as genuine preferences convert into arrangements that are protecting something, without any announcement and without the behavior changing at all. The transition is easy to miss precisely because nothing on the surface moves. The private count is the first thing that appears. When you find yourself knowing the number of weeks, something has changed underneath.

What If You Do Not Actually Know What You Want?

This is more common than anybody writes about, and it explains the strangest feature of the whole pattern, which is that you cannot answer the question either.

Here is the test. Somebody asks what you want this to be. Not what they want. What you want. Most people in this root find that nothing specific arrives, and that what surfaces instead is a list of things they suppose they should probably want by now.

If there is no specific want, then leaving it open amounts to the only honest position available rather than to indecision, because deciding requires something to decide in favor of. The arrangement stands in for the fact that you have not established what you want rather than preventing you from getting it.

The tells are worth sitting with.

You can describe in considerable detail what you do not want a relationship to look like, and go completely blank on what you do want. You are relieved when they are unavailable and would not have been able to decline the evening yourself. Asked to picture a version of this a year from now that you would be pleased about, you produce nothing with any content in it, or you produce a picture assembled entirely out of other people's arrangements.

And there is the reversal that gives this root away. When they do move toward more definition, something in you pulls back, and you cannot explain why, and you conclude you must not want them. The more accurate reading is that you do not have a shape to compare their offer against, so any specific offer feels like a constraint rather than a gain.

What follows from this root is slow and nobody sells it. The work sits on developing a want rather than on resolving the arrangement, and a want gets developed by having specific experiences and noticing which ones you would repeat. It takes seasons rather than weeks. What it does not respond to is any amount of thinking about this particular person, because the missing piece was never located in them. If the thing you cannot do is name a want and say it, the page on not being able to say what you need covers the machinery of that properly, including why a want with no case attached almost never clears the bar to be spoken.

What If the Effort You Have Already Put In Is What Is Keeping You Here?

This root explains why leaving feels like wasting rather than like stopping, and it operates almost entirely outside your awareness.

Effort gets read backward as evidence of significance. The reasoning runs: I have put seven months into this, I have rearranged my weeks around it, I have thought about this person more than about anything else this year, therefore it must matter enormously. Turn that around and the direction of causation is at least as plausible in reverse. You have thought about it constantly because it is unresolved. Unresolved things demand attention regardless of their value. The intensity is reporting on the state of the question rather than on the size of the thing.

The tells are quite distinctive.

You defend the arrangement to friends with the amount of time as the argument. Seven months is a long time to be wrong about somebody. Notice that duration is the whole case, and duration is exactly what accumulates automatically.

You describe leaving in the vocabulary of loss rather than choice. You would be throwing away the time. Wasting a year. Giving up after everything.

And you can name a great deal that you have invested and comparatively little that you have received, which people in this root usually discover with some discomfort when they write both columns down.

The general machinery here has a long research history and it is thoroughly worked through on the page about how to know whether to leave a relationship, which handles the years-already-spent problem with the evidence properly attached. I am not going to duplicate it. The version specific to this page is worth one sentence: in a defined relationship, the thing you would be losing is at least visible and can be named. Here, the effort has gone into something that has no agreed name, which means the loss has no shape either, and a loss with no shape is unusually hard to walk away from.

Why Do I Keep Ending Up in Undefined Relationships With Different People?

Because the part that repeats is smaller and more specific than a type, and it lives in a single moment rather than in a category of person.

Every one of these arrangements contains a point at which a name could have formed and did not. Somebody nearly asked. An introduction nearly happened. A question was written into a message and then taken out of it again. That point arrives in every version, with every person, usually somewhere in the second or third month, and what you did at that point is what has stayed the same across all of them. It is a far narrower object than a preference for unavailable people, and it is the only part of the sequence you have any access to.

This matters because the popular explanation runs the other way. You attract a certain kind of person. You have a type. Something about you signals availability to people who will not commit. Those explanations are comfortable in one specific respect, which is that they put the mechanism outside you, where you cannot work on it, and they are also the kind of account that cannot be wrong, since any run of arrangements can be described afterward as a type.

The honest state of the evidence here is thinner than anybody writing about it lets on. Nobody has followed a person across two undefined arrangements. There is no research on repetition in this pattern at all. The work described further up this page looks at single arrangements, at one moment, mostly by asking people about them, and none of it was built to answer the question you are actually asking. Anybody handing you a mechanism for why this keeps happening is reasoning from a shape rather than reporting a result, and that includes this page.

What can be said is what the second root above already implies. Protection that works never produces the failure that would make you abandon it, so a strategy that has kept you from being turned down for seven months arrives fully intact at the start of the next arrangement, unexamined and unchallenged by anything that happened. That is a mechanism rather than a finding, and it is offered as one.

There is a cheap way to check whether any of this describes you, and it does not involve the person you are seeing now. Take the previous two arrangements and write down, for each, the week at which it stopped moving and what you did in that week. Most people find the same two or three actions in both, and find themselves reaching for the same words to describe them. If the two look nothing alike, the repetition is more likely to be circumstance than pattern, which is a genuinely useful thing to establish before you spend a year working on something about yourself that may not be there.

Why Does the Ultimatum Fail, and Why Does Waiting Fail Too?

Both fail, in opposite directions, and understanding why is what keeps this page from ending in either piece of standard advice.

The ultimatum fails because it produces a rule. Somebody who agrees to a label under pressure has complied with a demand, and compliance leaves the underlying situation entirely as it was. You get the word. Some months later you notice the same distance, the same avoided subjects, the same sense of standing slightly outside something, all of it now wearing a title. The mechanism is worked out carefully on the page about how to know whether to leave a relationship, which distinguishes settling a rule from meeting a need and shows why the first reliably fails to become the second. I am borrowing it rather than rediscovering it.

There is a second problem with the ultimatum that belongs specifically to this page. It converts a question about what you want into a test of what they will do. That hands the whole thing back to them, which is the move this pattern already makes far too often.

Waiting fails because nothing in the arrangement generates the event that would change it. This is worth being precise about, because waiting feels active. It feels like giving something room, or like being patient, or like letting things develop naturally.

Nothing develops naturally here. The arrangement is stable by construction. Both people are avoiding the same conversation, both find the current state tolerable, and every additional month adds constraint without adding clarity. There is no internal mechanism by which month twenty-four differs in kind from month seven. What does change is that the question becomes progressively more expensive to ask, because a plain question in month two is a reasonable request and the same words in month twenty-four arrive carrying an accusation about the twenty-two months in between.

So this page does not tell you to raise it, and it does not tell you to leave.

I want to be plain about why. Both of those are decisions that depend on which root is yours, and I do not know which root is yours. Somebody in the first root should do neither. Somebody in the fourth root should re-run a calculation and possibly do nothing at all. Somebody in the second root who forces a conversation before understanding what the arrangement has been protecting will very often reproduce the identical structure with a different person inside a year, which is the outcome nobody writing about this seems to anticipate.

What this page can do is get you to the point where whichever you do is chosen rather than defaulted into. That is a smaller promise and it is the honest one.

What Does This Actually Cost, If Anything?

For some readers, close to nothing, and an honest page has to start there rather than manufacturing a cost to justify itself.

Where a cost does appear, it appears in specific places and it is countable.

The fourth and fifth rows are the ones people underestimate.

Maintaining three accounts of your own life is genuinely tiring in a way that does not present as tiredness. It presents as a general flatness, a reluctance to talk about yourself, and a slow narrowing of which friends you tell things to. Nobody notices it happening because each individual softening is small.

The fifth row is the one with the longest reach. Time spent in an undefined arrangement is time spent practicing a particular position, which is the position of somebody who is available and not chosen. Practice is practice. Two years of it produces genuine skill at tolerating that position and no skill whatsoever at occupying a different one, which is part of why the next arrangement forms so easily. The explanation has less to do with attracting a type than with being extremely well practiced at one specific shape, which you can then produce with almost anybody.

The last row deserves its own sentence. The arrangement itself is survivable and often pleasant. The conclusion you draw about yourself from a run of them does more damage than any individual one, and it is entirely an addition. Nothing in the evidence supports it.

Where Does This Stop and Something Else Begin?

Six neighboring patterns get confused with this one, and the wrong name leads to the wrong work.

You are the one withholding. If you are the person keeping somebody warm with no intention of choosing them, rationing contact, and replying just enough that it cannot be called nothing, that is the opposite chair and it has its own page. Keeping somebody interested when you have no intention of choosing them covers the sender's side, including the timing mechanism and the five reasons people do it. It is worth reading even from this chair, because a good number of people occupy both positions with different people in the same year.

Preoccupation with somebody you barely know. If the thinking is enormous and the actual contact is thin, if you are rebuilding four short conversations rather than an ongoing arrangement, and if you do not actually know whether they are interested at all, that is a different structure entirely. Not being able to stop thinking about somebody you barely know owns the involuntary preoccupation and the mechanism by which ambiguity about somebody's interest keeps a loop open. The difference from this page is contact and mutuality. Here they are plainly involved, several times a week, for months, and the question is about the category rather than about whether they like you.

A defined relationship you are deciding about. If there is a name, an agreed status, possibly a shared address, and the question is whether to stay, that is a different decision with a different structure. How to know whether to leave a relationship handles it, including the years-already-spent problem and why an ultimatum produces the wrong kind of answer. This page ends where a definition begins.

The scan that runs with no evidence. If what you recognize is a constant checking for signs that you are about to be replaced, running whether or not anything has happened, that is jealousy that fires with nothing behind it, which is about a supply of affection rather than about a status.

Small things inside a defined relationship. If you are in something named and are doing things you would not want your partner to see, each small enough to pass on its own, that is a scale problem rather than a status one, and the small things you would not want seen covers it, including why a decision made in four hundred installments never registers as a decision.

The missing channel between wanting and asking. If the thing you recognize hardest is that you cannot say what you want in any setting, that the words are gone before they reach your mouth, and that this happens with colleagues and family too, then the arrangement is a setting rather than a subject. Not being able to say what you need owns that, and it explains the specific reason a question about status almost never clears the internal bar to be spoken: it is a want with no case attached, and wants are far harder to voice than needs.

What this page owns, and what those deliberately leave alone, is the sustained absence of a defined status between two involved people, the reasons it forms, and the reader's own part in keeping it that way.

How Long Does This Take, Honestly?

It depends entirely on which of the five is yours, and confusing them is why people conclude that nothing works.

If you genuinely wanted something light, no time at all, because nothing needs to change. The only thing worth altering is what you concluded it says about you, and that can change in the time it takes to read a paragraph.

If ambiguity has been protecting you, expect months, and expect the movement to be indirect. What changes is your tolerance for a plain answer rather than your circumstances, and that gets built by collecting small clear answers in low-stakes places first rather than by attempting the largest one available. People commonly notice about a year later that they have stopped needing things to stay vague, and cannot identify when it happened.

If you do not know what you want, allow a season or more, and expect it to move through experience rather than through thinking. You do not decide a want. You find one out, by doing specific things and noticing which you would repeat.

If the calculation was genuinely favorable, this changes when the conditions change, which is out of your hands. What is in your hands is a check every few months to see whether the reasoning still describes your actual life.

If effort has been reading as evidence, this is the slowest of the five, because what has to happen is the accumulation of contrary experience rather than an insight. You have to discover several times that stopping something is different from wasting it, and that discovery has to happen experientially before it is believed.

One thing works immediately in every version, which is separating the drift from the one question you can put to yourself. That is available today and it does not depend on knowing which root is yours.