Nine clauses, annotated.
Six that tend to hold up, three that tend to cause trouble, and the reasoning behind each. These are illustrations for a conversation with a lawyer — deliberately not a template.
These are illustrations, not a template. We are deliberately not giving you fill-in-the-blank language, and the reason is the whole point of this section: you cannot draft your way out of a bad process. A perfectly worded agreement signed two days before the wedding with no disclosure is worth less than a clumsy one signed six months out with a full schedule and two lawyers.
What clause language is good for is having a specific conversation. Read these, decide which shapes match your situation, and take that to your lawyer. Six of these are examples of drafting that tends to hold. Three are examples of drafting that tends to cause trouble — those are marked.
Separate property, defined by tracing rather than by list
The core clause in most agreements, and the one people get subtly wrong.
Why this is drafted this way
The tracing language is doing the work. A clause that simply lists what each person owns today protects nothing in twenty years, because by then the listed asset has been sold, rolled over, reinvested and renamed. Following the value rather than the item is what makes the clause survive time.
The proviso at the end is the honest part, and it is there for you as much as against you. A clause claiming that money can be run through the joint checking account for a decade and still be untouchable invites a judge to find the whole arrangement was never really observed. Building in a realistic commingling rule makes the rest more defensible — and it tells you what you actually have to do after the wedding.
A sunset that steps down rather than switching off
Not required anywhere. Disproportionately effective at surviving a challenge.
Why this is drafted this way
This is the answer to the hardest question in prenup law: the agreement that was defensible on the wedding day and looks brutal after a twenty-five-year marriage in which one person left the workforce. In a state that only tests fairness at signing, a court may have no power to fix that. A step-down fixes it in advance, on terms both people chose while they still liked each other.
The carve-out matters as much as the step-down. If the real purpose was protecting a family business or a prior marriage’s children, say so and protect that specifically — then let everything else soften. An agreement that protects one identified thing forever and releases the rest reads as purposeful. One that protects everything forever reads as a grab.
Disclosure acknowledgment that references a real schedule
A recital only helps if it is true. Here is what makes it true.
Why this is drafted this way
Boilerplate acknowledgments are weak evidence on their own; this one is strong because every limb of it points at a document that either exists or does not. A schedule with statements attached, a named lawyer, a dated waiver. Those are checkable years later. “Each party acknowledges full and fair disclosure” standing alone is checkable by nobody.
The thirty days and the separate seven-day waiver are borrowed from the two states that legislate timing. California’s rule runs seven days from presentation of the final agreement, and requires a separate written waiver if you go without counsel. Writing the tighter version into an agreement anywhere costs nothing and answers the duress argument before it is made. Virginia is a useful caution here: its § 20-151(B) makes the agreement’s own recitals prima facie correct — which is exactly why they need to be accurate.
Spousal support, written to survive your own state
The clause with four different legal answers depending on where you live.
Why this is drafted this way
9.4 is two safety valves in one paragraph. The first mirrors the public-assistance floor that most uniform-act states impose anyway — writing it in yourself costs nothing and makes the waiver look considered rather than reckless. The second is a severability provision aimed squarely at states where the whole waiver is void: Iowa is the clearest, where § 596.5(2) protects “the right of a spouse or child to support” and the permitted-subject list omits alimony entirely.
The independent counsel recital is not decoration in California. Cal. Fam. Code § 1612(c) makes a spousal support provision unenforceable outright if the party it is being used against was not represented by their own lawyer at signing. And note the direction of travel — counsel alone does not save the clause there either; it still cannot be unconscionable when enforcement is sought.
The one place this clause needs a different conversation: South Dakota. It dropped both spousal support from the permitted list and the public-assistance floor. There is no statutory floor there to fall back on.
Choice of law, written knowing you might move
Underrated. People sign in one state and divorce in another far more often than they expect.
Why this is drafted this way
Most premarital statutes expressly let you choose the governing law — it appears in the permitted-subjects list in nearly every act we read. What they do not do is guarantee the forum court will honor the choice. Utah is explicit about the limit: § 81-3-203(1)(f) lets a court apply the law of either party’s domicile instead “if it is fair and equitable.”
The last sentence is the practical one. If your choice is disregarded, you want the agreement construed as broadly as the new state allows rather than collapsing entirely. This clause matters most for a couple who sign in a state with a permissive rule and later move somewhere protective — or the reverse.
Business interest with a stated purpose
A recital explaining why the clause exists does real work.
Why this is drafted this way
Recitals are not operative and courts know it — but they frame everything that follows. A judge reading a lopsided-looking business clause with this recital in front of it sees a specific, legitimate, verifiable purpose. Reading the same clause with no explanation, the judge sees someone protecting a large asset from a spouse.
The final phrase is doing something specific. By pointing at the sections that do address the overall economic relationship, it tells the court that the parties thought about fairness somewhere in the document, and that this clause was not the place they did it. That is a much better story than silence.
Three that cause trouble
These are drawn from the shapes that recur in agreements people bring to a lawyer. They are marked because they are examples of what not to do.
The infidelity penalty
Extremely popular in conversation. Considerably less popular in courtrooms.
Why this fails
Two independent problems. The first is subject matter: premarital statutes permit you to contract about property, its management, disposition on separation or death, wills and trusts, life insurance, choice of law, and then “any other matter, including their personal rights and obligations, not in violation of public policy.” A penalty for conduct is not obviously financial and sits squarely on that public-policy line.
The second is more concrete and more dangerous: a clause under which one spouse profits financially from the other’s misconduct looks a great deal like a term that encourages or rewards the end of a marriage. Ohio codified exactly that limit in 2023 — terms must “not promote or encourage divorce or profiteering by divorce” — and Colorado and North Dakota reach it too.
Being straight with you about what we do not know: we could not confirm a single appellate holding striking a specific infidelity clause from an official court source, so we are not going to claim courts have done so. What we can say is that the statutory framework is hostile to it, and that the practical risk is not just losing the clause — it is that a punitive-looking term colors a judge’s reading of the sections you actually need.
The lifestyle clause
Weight, chores, in-laws, frequency. The genre that gets written about and rarely litigated.
Why this fails
Set aside enforceability for a moment and consider the mechanism. What is the remedy? A court is not going to order a weigh-in. Damages for a missed dinner are not calculable. A term with no available remedy is not really a contract term.
Now the risk that actually matters. Most agreements contain a severability clause, and most statutes let a court strike an unenforceable term and enforce the rest — Iowa says so expressly at § 596.8(2). So the likely outcome is that these clauses are simply ignored. The real cost is different and worse: a judge reading a weight requirement forms a view about the relationship the agreement came out of, and carries that view into the sections about the house and the retirement accounts.
If a non-financial commitment genuinely matters to you, it belongs in a conversation, in counseling, or in a letter you write each other. Not in the document a stranger will read aloud in a courtroom in 2049.
The all-in-one disclosure waiver
The clause that looks like it saves you work and instead removes your protection.
Why this fails
Waiving further disclosure is genuinely permitted almost everywhere — the uniform act contemplates a “voluntary and express” waiver “in writing,” and North Dakota requires it be in a separate signed record. So this is not void on its face. The problem is that this version waives disclosure that was never actually made.
The clause asserts full and fair disclosure and then waives the mechanism by which anyone could have obtained it. There is no schedule referenced. There is nothing attached. If the agreement is challenged in fifteen years, the party relying on it has a recital and no evidence, against a spouse testifying they never saw a balance sheet. In a state where inadequate disclosure is a standalone ground — Colorado, North Dakota, Iowa, Connecticut — that is close to fatal.
The version that works is boring: disclose properly, attach the schedule, and if you then want to waive further inquiry, do it in a separate signed document that says what was already provided.
One more time, because it is the thing worth remembering. Every clause above is a conversation starter for your own lawyer, in your own state, with your own facts. None of it is language to copy. The agreements that hold up are not the ones with the cleverest drafting — they are the ones where both people had a lawyer, both people saw the numbers, and nobody was signing in a hurry.
Find someone to draft it
Family law attorneys in all 50 states, credentials checked against official state records. Nobody paid to be there.
Read this part
The Cusp is not a law firm and nothing on this page is legal advice. This is an explanation of how these agreements work and where they fall apart, written so you can have a better conversation with a lawyer — not so you can skip having one. Prenups are one of the few documents where doing it yourself reliably costs more than it saves, because the thing that voids them is almost never the wording. It is the process.
Every rule below is cited to a statute and linked to the state’s own text. Where we could not confirm a state, we say so instead of filling it in.