This post is a companion to The Bloodletting of Family Law. That post diagnosed a system that mistakes itself for the cure. This one is about the cure the system already has and declines to use: the standard of proof, honestly applied.
Utah domestic relations cases run on the preponderance of the evidence standard for most questions, with clear and convincing evidence reserved for a few, such as contempt. Ask any judge or commissioner whether they apply these standards and they will say yes, of course. Watch the motion calendar for a month and you will see something different. In the moments that matter most—temporary orders, protective orders, motions to restrict parent-time—the standard of proof is not so much misapplied as bypassed. And nearly every pathology in modern family law practice grows in the space that bypass creates.
What the Standard Actually Commands
The burden of proof is not a formality recited in findings. It is a decision rule for uncertainty. It tells the court what to do when it does not know what happened—which, in cases built on private conduct and dueling credibility, is often.
The rule is simple and unforgiving: the party seeking relief must persuade the court that his or her version is more likely true than not. If the evidence balances—if the court hears both parties and cannot say it is persuaded either way—the movant loses. Not partially loses. Loses. Equipoise is not a tie to be split; it is a failure of proof, and failure of proof means denial.
Utah courts routinely violate this rule while believing they are honoring it, in two opposite directions.
The first violation is partial relief on unproven allegations. A parent alleges abuse. The evidence at the proffer hearing is a coin flip. The commissioner, unwilling to ignore the allegation but unable to find it proven, orders supervised parent-time out of an abundance of caution. This feels moderate. It is not. It awards relief to a party who failed to carry the burden, and it converts an unproven allegation into a “half-proven” one—a category our law does not recognize. Caution is a virtue in drafting orders. It is not a substitute for proof, and a restriction on a parent-child relationship entered without proof is not caution. It is error with good manners.
The second violation is the phantom corroboration requirement. Some factfinders treat uncorroborated testimony as presumptively insufficient—as though one witness’s account can never preponderate against a denial. That is not the law. No corroboration requirement exists in Utah civil practice; a single witness’s testimony, if the court credits it, satisfies preponderance, clear and convincing, and even proof beyond a reasonable doubt. What the standard requires is not corroboration but persuasion.
A court that hears an uncorroborated petitioner, finds her credible and the respondent not, and still denies relief “because it’s just her word” has invented a rule of evidence that does not exist. And a court that grants relief without ever making the credibility finding that alone could support it has committed the mirror-image error.[1]
Both violations share a root: the factfinder declines to decide. Splitting the baby avoids deciding. Demanding corroboration avoids deciding. The standard of proof exists precisely to make non-decision impossible—when the court cannot say it is persuaded, the standard decides for it, and the answer is denial. Holding fast to that rule is uncomfortable, because it means unproven allegations produce nothing, including allegations that may be true. But the alternative—relief calibrated to suspicion rather than proof—is a system in which accusation itself is punishment. We have written about where that road leads.
The Filer Who Already Knows
Honest application of the standard has a second-order effect that matters as much as the rulings themselves: it changes who files what.
Some litigants and some lawyers file motions they know, or strongly suspect, cannot meet the applicable standard of proof. They file anyway, because filing is cheap, odds of adverse consequences are slim, and the payoff structure rewards it. A protective order petition that fails still occupied the marital home for six weeks. An abuse allegation that fizzles still shaped the temporary orders hearing. A contempt motion abandoned on the eve of the evidentiary hearing still bled the other side’s retainer. These are not lawsuits; they are leverage instruments wearing captions.
The court cannot see a filer’s state of mind. It can see the proxies, and the proxies are reliable:
- The pleading-versus-proof gap. Dramatic, specific allegations in a verified motion, followed at the evidentiary hearing by no witnesses, no documents, and testimony that retreats into vagueness under oath. Honest losers usually have evidence that simply fails to preponderate. The nuisance filer’s signature is the allegation no one ever attempted to prove, because it was never meant to reach a hearing.
- Conduct inconsistent with the allegations. The petitioner who alleges terror and then initiates contact, proposes unsupervised exchanges, or invites the respondent to family gatherings has generated evidence of what she believed when she filed.
- Tactical timing. The protective order filed the week before the temporary orders hearing. The abuse allegation surfacing only after an adverse custody evaluation. One instance proves little; a pattern proves plenty.
- Withdrawal upon resistance. The claim discounted to nothing the moment it is set for an evidentiary hearing has told the court what it was for.
Here is the uncomfortable part: the tools to address this already exist, and courts do not use them. Rule 11 requires factual contentions to have evidentiary support after reasonable inquiry; the knowing filer violates it by definition. Utah Code § 78B-5-825 mandates—not permits—an award of attorney fees against a party who asserts a claim in bad faith without merit. Yet bad-faith findings are nearly extinct in domestic practice, because judges experience the finding as calling a litigant a liar and reserve it for cartoonish misconduct. A perfect anti-nuisance statute enacted tomorrow would die the same quiet death.
The reform target, then, is not new law but lower activation energy for the law we have. When a party’s verified allegations produce essentially no admissible support at the hearing—not weak support, none—the burden on fees should shift to that party to show the filing had an objectively reasonable basis. Declarations signed under penalty of perjury should carry an actual penalty when affirmatively contradicted under oath. And repeat behavior should compound: a party’s pattern of unsupported filings within a case should shift fee presumptions for the rest of it.
One line must be drawn carefully. Doubt is not the offense. Every honest lawyer has filed motions without certainty of prevailing; in credibility-dependent practice, certainty at filing is rarely available to anyone. The offense is filing for a purpose other than winning—leverage, harassment, delay, cost infliction. Purpose is the target; the proxies above are how courts see it.
The Hard Case: The True Victim Who Cannot Prove It
Any framework that polices weak filings must answer the hardest case honestly: the genuine victim of domestic violence whose abuse happened, as abuse does, behind closed doors, and who cannot corroborate it.
Start with what she is not. She is not a nuisance filer. Her own truthful testimony is evidence—legally sufficient, if credited, to carry the burden. Whether it will prevail against a flat denial is a credibility contest no one can reliably predict, which means she does not “know she can’t meet the standard.” Her case is uncertain, not baseless, and sanctioning uncertain-but-true claims would build a rule whose chief beneficiaries are abusers careful enough not to leave marks. Nothing in this post supports punishing her for filing.
But candor—the kind a lawyer owes a client, not the kind a system uses to lecture citizens—requires telling her what losing costs. A denied petition does not return her to where she started. It becomes an exhibit: she made allegations; a court heard them; a court rejected them. Opposing counsel will convert “not proven” into “not true” at every subsequent hearing, and factfinders hear the two the same way no matter what the law says. Her credibility—the only asset her uncorroborated case had—is now impaired on the one issue where she was telling the truth. If she is assaulted again and this time has evidence, she litigates with a prior loss around her neck.
The honest counsel is therefore not “stay silent.” It is: the courthouse is a better proof machine than it is a truth machine. Feed it a true claim it cannot verify and it will stamp the claim unfounded, and the stamp will be used against you. Build the record the case currently lacks—documentation going forward, disclosures to advocates and providers, exchange structures that eliminate the private encounters where abuse occurs—and preserve your credibility for the day the machine can do something with it. That is not the system abandoning victims. Splitting the baby on unproven claims is the system abandoning victims,[2] because it teaches courts to treat every allegation, true and false alike, as half-true—and a system that cannot tell the difference protects no one.
Weak Cases Are Error Generators
The resource argument against weak filings is real but underpowered—”courts are busy” invites the answer “fund more courts.” The stronger argument is accuracy: weak cases do not just consume resources; they produce wrong answers at a higher rate, in both directions.
A strong case constrains the factfinder. The evidence points somewhere; the ruling follows. A weak case leaves the factfinder unconstrained, and unconstrained discretion in child custody disputes gets filled by exactly the forces that should worry everyone: risk aversion, status quo deference, baby-splitting, and the private priors of whoever drew the assignment. The coin-flip petition is not decided; it is guessed, and the guess is laundered into findings. Sometimes the guess is too lenient and real abuse goes unaddressed. Sometimes it is too harsh and a parent is restricted on nothing. Both errors cluster where the proof is thinnest.
The tempting response is gatekeeping—screening filings for strength at the door. Resist it. Domestic relations practice is fact-dependent to its core; a commissioner deciding at intake which petitions deserve a hearing is just the thin-evidence problem relocated earlier, with less record and less process. The door is the wrong place to guard.
The right place is the courtroom, and the mechanism is embarrassingly simple: make weak cases lose—completely and predictably and without impunity. Denied means denied. No cautionary restrictions on failed proof, no half-relief. Price the loss through fee allocation so the lottery filer’s math changes. Say it in written rulings, so the bar learns case by case. The actual gatekeeper in this system is not the court; it is the lawyer at intake, and lawyers forecast honestly when losing is real and expensive. “We can file this, but on this evidence we will likely lose and you may pay his fees” is a different conversation than “we can file this”—and the second conversation is the one the current payout structure invites.
That is the whole reform, and it requires no new statute, no new rule, no task force. Courts do not need to deter weak cases. They need to stop rewarding them. Hold fast to the standard of proof—no relief without persuasion, full denial at equipoise, real consequences for baseless filings—and the gatekeeping happens by itself: through candid intake meetings, recalibrated client math, and a bar that adjusts to a payout table with teeth. The standard of proof is not an obstacle the system tolerates on the way to doing justice. It is the immune system. It only works when nobody suppresses it.
Utah Family Law, LC | divorceutah.com | 801-466-9277
[1] The “mirror-image error” fundamentally intersects with the doctrine of evidentiary equipoise under the preponderance of the evidence standard. In civil matters, the petitioner bears the burden of proving that their claims are more likely true than not (i.e., a probability exceeding 50%). In a pure “he said/she said” dispute lacking external corroboration, a factfinder who fails to make an explicit credibility determination leaves the competing testimonies in a state of perfect balance, or equipoise (a 50/50 tie). Because a tie must legally result in a judgment for the respondent, a court that grants relief without a clear finding that the petitioner is more credible than the respondent effectively misapplies the burden of proof, awarding a judgment to a claim that has failed to persuasively tip the scales past the required 50% threshold.
[2] And it’s a system that does not understand the wisdom of King Solomon.
Solomon’s wisdom was not that compromise is just in close cases. His wisdom was that a false compromise can reveal, and sometimes reward, the party with no real claim. When courts treat every allegation, true and false alike, as half-true, they protect no one.
Solomon’s judgment was wise precisely because he never intended to split the baby; he used the threatened compromise to expose the truth. A court that actually compromises between true and false allegations does the opposite: it treats every claim as half-true, and a system that cannot tell the difference protects no one.