Your child is a different person at four than at fourteen
Every divorce guide on the internet is written for the adult going through it, and the child appears as a variable in somebody else’s arithmetic. These pages are about the child — six age bands, what actually changes between them, and what the sources do and do not support.
This page is written for the parent, about the child. It is not written to be read by the child it describes, and we have kept the two kinds of page apart on purpose. Pages written directly to children start at age ten and are labeled as theirs.
The short version5 points · tap one to jump to it
- Pick the age they are now — The same divorce is a different event at two, at eleven and at seventeen — not because it hurts more or less, but because a child of two cannot be told, a child of eleven is measuring, and a young person of seventeen can be handed things they should not carry.
- The four things true at every age — The court guide states it flatly: parents must not use the child as a messenger to communicate with the other parent.
- When the court wants to hear from your child — A child may be interviewed by a guardian ad litem, a custody evaluator, a family-court counselor, or occasionally the judge in chambers, and what each of those titles means varies by state — see who speaks for your child for who each of them is and who pays.
- Where this material comes from, and why that is a problem
- What these pages are not — Pages written directly to young people are a separate set, they start at age ten, and they are labeled as theirs — because a twelve-year-old reading advice about how to manage them is a bad experience and we are not going to blend the two.
In this subject 5
Pick the age they are now
The same divorce is a different event at two, at eleven and at seventeen — not because it hurts more or less, but because a child of two cannot be told, a child of eleven is measuring, and a young person of seventeen can be handed things they should not carry. Each page below is written about one band.
Under 3
No memory of before. Routine is the explanation, frequent contact beats long blocks, and the handover is the whole ballgame.
3 to 5
Magical thinking. They can privately believe they caused it and will not say so, which is why it has to be said before they ask.
6 to 8
The reunion-fantasy age. They think it is fixable and they will try, and the trying looks like sweetness.
9 to 12
Keeping score, and often assigning blame. They want a say — and a say is not the same as a choice.
13 to 15
Two separations at once. Not every slammed door this year is about the marriage, and reading them all that way costs something.
16 to 18
Logistics get easier; the confidant problem arrives. The lasting harm at this age is usually being told too much.
Grown-up children
However long ago it was. Two Thanksgivings, forever. Written adult to adult, with no developmental framing at all.
Looking for the evidence rather than the ages?
The band pages above deliberately print no outcome figures. The research — what the studies actually found, where they disagree, which programs have trials behind them and which are custom — lives on its own page, and that is the one to read if you want the numbers with their caveats attached.
The four things true at every age
Conflict does more of the work than the divorce
This is the most consistent finding in the literature and it is the only one that is also actionable. How much of your conflict a child is asked to witness, carry or referee is the variable you control, at every age on this page. We print no percentage, because the figures in this field move enormously with how the question was asked.
No child is a messenger
The court guide states it flatly: parents must not use the child as a messenger to communicate with the other parent. It is the rule that gets broken most casually — usually as ask your mother whether — and the habit it teaches is the thing grown-up children of divorce describe most often decades later.
A say is not a choice
From ten upward the guide repeats one formula almost word for word in every band: let them express their views, while making clear that the parents make the final decisions. Both halves are the point. Hearing them protects them; handing them the decision hands them a choice between two parents.
Resistance is not evidence
Distress at a handover, reluctance to go, a sudden preference — the guide says at several ages that these do not necessarily mean the other parent is bad or that the child does not want to be there, and that children resist more when the adults are tense at the exchange. The most common misreading in the whole subject is treating a child’s difficulty as a fact about the other house.
When the court wants to hear from your child
Every band page above carries the same section, because it is the part of the process parents most want a script for and the part where a script does the most damage. A child may be interviewed by a guardian ad litem, a custody evaluator, a family-court counselor, or occasionally the judge in chambers, and what each of those titles means varies by state — see who speaks for your child for who each of them is and who pays.
We describe what happens. We will not tell you how to prepare a child to say something, on any of these pages. That is coaching, it is usually obvious to the person conducting the interview, and it asks a child to perform for an adult about their own family. Whether a stated preference carries weight, and at what age, is a question of state law rather than child development — see at what age a child can choose.
Where this material comes from, and why that is a problem
We went looking for the best age-banded public source on how children of different ages experience a separation. What we found is worth reporting on its own.
AACAP’s Facts for Families #01, the sheet everybody cites, is not age-banded at all. It is a good document and it says young children, older children, teenagers and adults — nothing finer. It cannot support a claim about nine-year-olds specifically, and we do not use it for one.
The best age-banded source is a court publication from 2009. Planning for Parenting Time: Arizona’s Guide for Parents Living Apart, written by a statewide committee of judicial officers and mental health providers, bands children birth–2, 2–3, 3–5, 6–9, 10–12, and teenagers 13–18 split into early and late. It is careful, it is specific, and it is seventeen years old. We could not find it on azcourts.gov on the day we read it and used a public mirror.
That is the finding: the most useful age-banded public document on children and divorce is a 2009 state court guide that is hard to find. Every band page here says which source band it draws on and where our bands differ from the source’s — three match exactly, three do not. We would rather print the mismatch than smooth it over.
What these pages are not
They are written for parents, about children. Pages written directly to young people are a separate set, they start at age ten, and they are labeled as theirs — because a twelve-year-old reading advice about how to manage them is a bad experience and we are not going to blend the two. They are at the pages written for them, and every one of them carries a safety exit above the content.
They are not a weapon. There is no checklist here for establishing that the other parent has turned your child against you. That framing is scientifically contested and is routinely used as ammunition in custody litigation, and a page that helps one parent build a case about the other has picked a side against the child.
They collect nothing. No quiz, no ages entered, no “tell us about your child.” Nothing on these pages builds a record about a real minor, and nothing ever will.
They are not therapy and they do not diagnose anybody. Where a child is struggling in a way that worries you, the person to talk to is a person.
What we do not print. AACAP’s Facts for Families #01 (March 2011) is the usual citation here and it is not age-banded at all — it says young children, older children, teenagers, and nothing finer. We do not use it for a band-specific claim. On outcomes, the direction of the literature is that conflict between parents does more of the work than the divorce itself (Kelly, J. B. & Emery, R. E. (2003), ‘Children’s adjustment following divorce: risk and resilience perspectives’, Family Relations 52(4), 352–362), and we deliberately print no percentage, because the figures in this field move enormously with how the question was asked. Nothing on these pages collects any information about a child. Corrections go to thecusp.app/corrections with the date, what changed, and how long the error was live.