Jennifer's Laws · I. The Ten Foundational J-Laws
J-Law 6 — End the Presumption of Judicial Impartiality
No reviewing court shall presume that a judge was impartial, that a judge read or considered the parties' filings, or that an unexplained ruling rested on evidence or legal authority. Boilerplate recitals — "having heard the evidence," "being fully advised in the premises" — shall carry no evidentiary weight of any kind. Where the record does not affirmatively show what a court considered, every doubt shall be resolved in favor of the appealing party rather than against them. Where the appealing party was self-represented, the reviewing court shall presume that the trial court did not afford that party's arguments equal consideration, and the burden of demonstrating otherwise shall rest on the judge who issued the ruling.
No statute created the presumption of judicial regularity. No constitution mentions it. Judges built it, applied it to themselves, and have extended it for more than a century — the same construction, by the same hands, as the immunity doctrine it operates alongside. The two work as a pair and neither is fully effective without the other. Immunity ensures a judge cannot be sued for the harm a ruling causes. The presumption ensures the ruling itself is nearly impossible to disturb. Remove one and the other still functions. Remove both and a judge finally occupies the position every other American already occupies: answerable for the work, on the record, to somebody.
The presumption's central instrument is a phrase that means nothing. Orders across every courtroom in the country recite that the court heard the evidence, considered the arguments, and is fully advised in the premises. The words arrive in the document before the hearing does. They appear identically on orders entered after weeks of briefing and hours of argument by judges who read every page, and on orders entered in 40 seconds by judges who read one side's filing, or nothing at all, and announced the result before anyone finished sitting down. A phrase that cannot distinguish between those two proceedings contains no information about either. Reviewing courts credit it anyway, and the consequence is exactly backward: the judge who explains nothing becomes harder to review than the judge who explains something, and silence is converted from a defect into a shield.
Illinois supplies the doctrine's clearest illustration. In Foutch v. O’Bryant, the state's highest court affirmed the denial of a motion to vacate a default judgment where the record included the deputy clerk's sworn statement that the clerk's office never sent the defendants notice of the trial date, her sworn statement that she did not recall giving testimony at the hearing, and the trial judge's own sworn affidavit that he did not recall any sworn testimony being offered. The court obtained those affidavits itself, read them, and held that it must be presumed the ruling was supported by evidence. The recital on the order outweighed the sworn recollection of the judge who signed it. The prevailing party, meanwhile, filed no brief and made no appearance in the supreme court; the institution defended the judgment on its own initiative. That case is cited across Illinois to this day, and versions of it govern in every state.
A presumption earns its place in law on exactly one ground — that it encodes what is usually true, sparing the system the cost of proving the obvious. Measured against its own justification, this one fails completely. Federal administrative data covering civil cases in United States district courts from 1998 through 2017, excluding prisoner litigation, shows that when both parties had lawyers and a case reached final judgment for one side, plaintiffs won 51 percent of the time. When the plaintiff was self-represented and the defendant had counsel, the plaintiff won 4 percent of the time and lost 96 times out of 100. Represented plaintiffs prevailed at roughly 13.7 times the rate of self-represented ones, and the disparity holds across nearly every category of case — about 8.8 to one in contract, 9.8 to one in tort, 11.5 to one in civil rights, 13.3 to one in insurance, and 42 to one in property matters. In state courts, where most civil litigation happens, a study of more than 925,000 disposed civil cases across 10 urban counties found at least one self-represented party in over three-quarters of them. The presumption of equal consideration is applied, every day, to a population the data shows is not receiving it.
The customary explanation is selection bias: contingency lawyers accept strong cases, so weaker cases proceed without counsel. That accounts for part of the gap and cannot account for what happened when the system tried to close it. Of 90 responding federal district courts, 87 implemented programs to assist self-represented litigants — electronic filing, docket access, direct communication with pro se law clerks, handbooks, standardized forms, mediation, public information. Comparing outcomes in reforming courts against non-reforming courts produced a difference of no more than one percentage point, most likely statistical noise. The Eastern District of New York went furthest of all, creating a dedicated pro se office headed by a magistrate judge with authority over the entire self-represented docket; five years of data later, the estimated effect on win rates was 0.04 percent, statistically indistinguishable from zero. Supply every tool the system has ever proposed short of a lawyer, and the outcome does not move. Whatever produces a 96 percent loss rate is not the litigant's paperwork, preparation, information, or access — it is something none of those reforms touch, and the profession has known it for years. Judge Richard Posner left the United States Court of Appeals for the Seventh Circuit in 2017 over his colleagues' treatment of these litigants, after all 11 of them rejected his proposals to examine how their appeals were handled, and said publicly that most judges regard them as "kind of trash not worth the time."
Reversing the default is the entire reform, and it is a modest one. A judge who genuinely read the filings, weighed the arguments, and applied the law can demonstrate it in a paragraph and a citation, which J-Law 2 already requires. A judge who cannot has revealed something the parties are entitled to know. Where the record is silent, the doubt belongs to the institution that controlled the courtroom, employed the reporter, drafted the order, and possessed every means of preserving what happened — never to the citizen who possessed none of them and was told afterward to purchase the transcript proving her own case. And where the losing party was self-represented, the reviewing court should begin from what the evidence actually shows rather than from a fiction: that equal consideration is the exception rather than the rule, and that the judge who claims otherwise should be the one to say so, in writing, on the record.
The expected counterargument: presuming judges are biased insults the bench, destroys public confidence in the judiciary, and will flood appellate courts with challenges to every unexplained order. Notice first what the presumption actually is. It is not a compliment paid to judges; it is an allocation of the burden of proof, and it currently sits on the party with the least power, the least money, and no ability to create the record it demands. Moving a burden is not an accusation, and every American professional already works under the opposite default — a surgeon's outcome is reviewed against the chart rather than against the surgeon's assurance that proper care was given, and no one calls the chart an insult. Public confidence deserves the same scrutiny: confidence resting on a presumption that the measured data contradicts is not confidence, it is unfamiliarity, and it survives only while the numbers stay unpublished. The flood objection is the most revealing of the three. If requiring courts to show their work would overwhelm the appellate system, that is a measurement of how many rulings currently could not survive being shown, offered by the institution issuing them — and volume is the scandal rather than the defense. The workload also runs the other direction, since a trial judge who states reasons produces a reviewable order and forecloses the satellite litigation that unexplained rulings generate. What remains, stripped of its borrowed concerns, is a request that the most powerful participant in the room continue to be believed without evidence, while everyone else proves everything. Stated plainly, no one would propose it.
The full case for this J-Law, including the decision that built the doctrine, is set out in the accompanying article: "Fully Advised in the Premises": How a Boilerplate Phrase Became Illinois's Most Powerful Rule of Evidence.
“Fully Advised in the Premises”: How a Boilerplate Phrase Became Illinois’s Most Powerful Rule of Evidence
In 1984, the Illinois Supreme Court decided a dispute over $3,119.80 worth of work performed on a piece of Knox County real estate. The amount was small and the question looked technical. Foutch v. O’Bryant nonetheless became one of the most heavily cited decisions in Illinois civil practice, and it is cited today for a proposition that should alarm anyone who believes courts exist to be checked: when a trial judge's order recites that the judge heard the evidence, a reviewing court must presume the judge heard the evidence — even when the record in front of the reviewing court contains sworn statements indicating that no evidence was ever taken.
That is not a strained reading of the case. It is what happened in it.
What Actually Happened in Foutch
Raymond O'Bryant and Terry Hainline were sued in February 1979. They answered the complaint, denied the allegations, and then heard nothing for nearly three years. In December 1981 the circuit court held a docket call at which trial dates were assigned. They say they never received notice of it. A trial date was set for March 1982. They say they never received notice of that either. When the case was called, they did not appear, an ex parte hearing was held, and judgment was entered against them. According to their filings, they learned of the judgment through a credit bureau publication.
They moved to vacate within 30 days, as the statute allowed. A hearing was held on May 10, 1982. No court reporter was present. The motion was denied a week later, in an order stating that the court, "having heard the evidence, adduced the arguments of counsel and now being fully advised in the premises," found that the defendants' motion should not be granted.
That sentence decided the case. Not the notice failure. Not the merits of the underlying claim. Not whether two men had ever been given a real opportunity to defend themselves against a money judgment. The sentence.
The Court Had the Evidence That Rebutted the Presumption, and Applied the Presumption Anyway
Here is the part of Foutch that deserves to be read slowly, because it is the part the citing cases never quote.
The appellate court's decision had referenced an affidavit from a deputy in the circuit clerk's office. That affidavit was not in the materials before the supreme court, so the supreme court went and obtained a copy from the appellate court clerk's office. Having done so, this is what the court possessed when it wrote its opinion:
An affidavit from deputy clerk Shirley Carpenter stating that neither she nor any other member of the clerk's office had sent defense counsel a copy of the order setting the case for trial. That is a sworn statement that the notice failure the defendants alleged actually occurred.
The same affidavit stating that Carpenter had "no recollection of giving sworn testimony at [the] hearing" in May 1982.
A representation in the appellant's own motion that Carpenter had "only made off the record statements to the court."
And an affidavit from the trial judge himself, stating that he had no "independent recollection of the specifics of what was said at the hearing, nor do I recall any sworn testimony being offered."
Read that last one again. The judge who signed the order reciting that he had heard the evidence swore, under oath, that he did not recall any sworn testimony being offered. The clerk whose off-record remarks appear to have been the entire substance of whatever the court heard swore that she did not recall testifying. The court of last resort in Illinois held both affidavits in its hands.
It affirmed anyway. "It must be presumed," the court wrote, "that the denial of the motion was in conformity with the law and was properly supported by evidence."
The presumption was not applied in the absence of contrary information. It was applied over contrary information, supplied by the judge who signed the order. The recital of a form phrase was treated as more reliable evidence of what happened in that courtroom than the sworn recollection of the man who presided over it.
One further detail belongs in any honest account of this case. The plaintiff, who won, filed no brief in the Illinois Supreme Court and made no appearance there. The court affirmed in favor of a party who did not show up, against a party who appeared and briefed the case, on a theory the winning side never bothered to argue. The court constructed the defense of the judgment itself.
The Phrase Means Nothing, and Everyone in the System Knows It
Anyone who has spent time inside an American courtroom recognizes the language immediately. "Having heard the evidence and the arguments of counsel and being fully advised in the premises." It appears on orders across every division of every circuit, in every kind of case, year after year. It is not a finding. It is a template — words that arrive in the document before the hearing does.
Those exact words appear on orders entered after months of briefing, hours of argument, and weeks of judicial research, by judges who read every page and thought hard about the result. Those same exact words appear on orders entered by judges who read one side's filing and none of the other's, asked no questions, heard no argument, and announced the ruling before anyone sat down. The phrase cannot distinguish between the two, because it was never written to describe either one. It is boilerplate, and boilerplate by definition carries no information about the specific proceeding it is attached to.
Reviewing courts nonetheless treat this phrase as though it carries evidentiary weight. That is the central absurdity of Foutch and the line of cases behind it. A doctrine built to respect the trial court's superior view of the proceedings has been converted into a doctrine that credits a phrase over the proceedings — and over the sworn recollection of the judge who conducted them.
The Recharacterization Problem: What Skaggs Actually Shows
Foutch rests on Skaggs v. Junis, a 1963 decision the court quotes at length. In Skaggs, the plaintiff complained that the record failed to disclose the evidence supporting an award. The Foutch court describes the argument this way: the plaintiff "contended that it must be assumed that none was heard."
That is not the plaintiff's argument. It is the reviewing court's summary of the losing party's argument, written by the judges who rejected it, published in an opinion the losing party had no hand in drafting.
This is the recharacterization problem, and it runs through the entire body of American case law. Every published opinion is a document authored by the side that prevailed, describing what the side that lost said. The losing brief is not published alongside it. The transcript is not published. The exhibits are not published. What survives into the permanent record is the court's account of the dispute, and the court's account is written after the court has already decided how the dispute comes out.
We have no way of knowing what the Skaggs plaintiff actually presented. The plaintiff may have offered nothing more than the bare inference the opinion describes. The plaintiff may equally have produced substantial material demonstrating that no evidence was taken — the same kind of material the Foutch defendants produced 20 years later, and which the Foutch court read and set aside. The opinion is the only surviving witness, and it is a witness with an interest in the outcome.
Anyone who has litigated for any length of time has read an opinion describing arguments they did not make, conceding points they did not concede, and omitting the evidence that was the entire basis of their filing. This does not require dishonesty in any provable sense. It requires only the ordinary human habit of recalling events in the shape that justifies what you have already decided to do. The reviewing court arrives at a conclusion, then works backward to build the record that supports it — precisely the process the trial court is presumed not to have used.
"Any Doubts Will Be Resolved Against the Appellant"
Foutch states the operating rule plainly: an appellant bears the burden of presenting a sufficiently complete record, and "any doubts which may arise from the incompleteness of the record will be resolved against the appellant."
Consider the choice embedded in that sentence, because it was a choice. Confronted with an incomplete record, a reviewing court has exactly two options. It can resolve the uncertainty in favor of the citizen who lost and against the government official whose conduct is under review, or it can resolve the uncertainty in favor of the government official and against the citizen. Illinois chose the second, and courts across the country made the same choice, and the choice is now recited as though it were dictated by logic rather than selected by the people it protects.
Nothing compelled it. The opposite default is available and defensible on every principle courts claim to honor. The party with the power to create a complete record is the court itself. The party with the institutional resources to preserve proceedings is the court itself. The party whose conduct is being reviewed is the court itself. Placing the consequences of a missing record on the citizen who did not control the courtroom, could not compel a court reporter, and was not present when the order was drafted is not a neutral allocation of burdens. It is an institution assigning the cost of its own recordkeeping failures to the people who came to it for help.
Foutch also notes that the defendants filed no bystander's report under Rule 323(c) and no agreed statement of facts under Rule 323(d). Examine what those remedies actually require. A bystander's report is prepared by the appellant, served on the opposing party, subject to their proposed amendments, and then presented to the trial court for certification. The remedy for a proceeding the trial judge did not record is a document the trial judge must certify. The cure for the judge's missing record requires the judge's signature. When the dispute is about what that judge did, the appellant is directed to a remedy the appellee and the judge together control.
Stacked Presumptions and the Distance From the Truth
Trace the structure Foutch and its cited cases build, layer by layer:
- The judge is presumed impartial.
- The impartial judge is presumed to have followed the law.
- The boilerplate recital is presumed to be a truthful account of what occurred.
Any gap in the record is presumed to be the appellant's failure.
- Every doubt arising from that gap is resolved against the appellant.
Each layer is defensible in isolation, and each rests entirely on the one beneath it. Stack all five and the outcome is fixed before anyone examines what actually happened. Presumptions are supposed to be shortcuts around facts that are usually true. Stacked five deep, they stop approximating anything. They become a closed system that generates the same answer regardless of input, and the further the stack rises, the further the result travels from the events it purports to describe. Foutch is what the top of that stack looks like: a court holding that a judge heard evidence in a hearing where the judge swore he did not recall evidence being heard.
The Price of Rebutting a Presumption
Set aside the doctrine for a moment and consider the economics, because the economics are how this rule actually operates on human beings.
A litigant who receives a one-sentence order and wants meaningful review must build the record the court did not build. That means purchasing transcripts, at rates that run into thousands of dollars for a case of any length. It means filing motions to correct or supplement the record. It means preparing a bystander's report and pursuing certification from the judge whose conduct is at issue. It means additional hearings, additional filings, and additional months, all of it consumed before a reviewing court will even look at the substance of the complaint.
Most attorneys will not do this work, because clients will not pay for it and the return is poor. Most people cannot afford to have an attorney do it. And a self-represented litigant facing this obstacle does not face it once. She faces it at every hearing, in every order, for the entire life of the case — a fresh unexplained ruling, a fresh demand that she fund and construct the record proving what happened in a room where she had no power to record anything. At some point the time, money, and energy run out. That exhaustion is not an unfortunate side effect of the doctrine. Given the structure, it is the doctrine's most predictable output.
What the Presumption Is Actually Presuming
Legal presumptions are justified on one ground: that they encode what is usually true, sparing the system the cost of proving the obvious in every case. Measured against that standard, the presumption that judges give equal consideration to the parties before them is among the worst-calibrated rules in American law, and the data has been available for decades.
The most rigorous study of the question examined federal administrative data covering civil cases filed in United States district courts from 1998 through 2017, excluding prisoner litigation. In cases that reached final judgment for one side, where both parties had lawyers, plaintiffs won 51 percent of the time — the coin flip one would expect. Where the plaintiff was self-represented and the defendant had counsel, the plaintiff won 4 percent of the time. The self-represented plaintiff lost 96 times out of 100. Overall, represented plaintiffs prevailed at roughly 13.7 times the rate of self-represented ones. The disparity holds across nearly every category of civil case: about 8.8 to one in contract cases, 9.8 to one in tort, 11.5 to one in civil rights, 13.3 to one in insurance, and 42 to one in property cases.
The standard response is selection bias — contingency lawyers take strong cases, so weak cases proceed without counsel. That explanation is real and it explains part of the gap. It cannot explain what happened next.
Federal district courts spent years building programs to close that gap. Of 90 responding districts, 87 had implemented at least one measure to assist self-represented litigants: electronic filing, docket access, direct communication with pro se law clerks, handbooks, standardized forms, mediation, public information programs. The same study compared outcomes in courts that adopted these reforms against courts that did not. The difference in win rates was no more than one percentage point, and most likely statistical noise.
The Eastern District of New York went further than anyone. It created a dedicated pro se office headed by a magistrate judge with broad authority over the entire self-represented docket. Five years of data later, the estimated effect on win rates was 0.04 percent, statistically indistinguishable from zero.
That finding is the one that matters, and it deserves to be stated without softening. Give self-represented litigants electronic access, forms, handbooks, clerk assistance, mediation, and a dedicated judicial officer assigned to their cases — every tool the system has ever proposed short of handing them a lawyer — and the outcomes do not move. Whatever is producing a 96 percent loss rate is not the litigant's paperwork, preparation, information, or access to the docket. It is something none of those reforms touch.
This is not a marginal population. Federal data shows that 27 percent of all civil cases filed between 2000 and 2019 involved at least one self-represented party. In state courts, where the overwhelming majority of civil litigation occurs, a study of more than 925,000 disposed civil cases across 10 urban counties found at least one self-represented party in over three-quarters of them.
And the profession is not unaware. Judge Richard Posner resigned from the United States Court of Appeals for the Seventh Circuit in 2017 specifically over his colleagues' treatment of self-represented litigants, after his proposals to review how their appeals were handled were rejected by all 11 of them. He said publicly that most judges regard such litigants as "kind of trash not worth the time." He described a process in which staff attorney recommendations on pro se appeals were generally rubber-stamped. A federal appellate judge with 35 years on the bench left the job over it and wrote a book about it, and the doctrine did not change by a single word.
What Foutch Is For
Every element of this structure is defended on grounds of judicial efficiency, respect for the trial court's superior vantage point, and the orderly administration of appeals. Examine what it actually accomplishes and a simpler description emerges.
Foutch and the cases it rests on ensure that a judge who explains nothing is harder to review than a judge who explains something. They ensure that the citizen, not the court, bears the financial cost of the court's failure to make a record. They ensure that a form phrase carries more weight on review than the sworn statement of the judge who signed it. They ensure that every ambiguity created by an institution's own incomplete recordkeeping is charged to the person who came to that institution seeking justice. And they do all of it while reciting a presumption of judicial regularity that the measured outcomes of American civil litigation contradict on their face.
A system does not arrive at a rule like this by accident, and it does not keep it for 40 years by oversight. It keeps it because the rule works — for the institution. It is the mechanism by which a court, at every level, protects itself from having to answer for what it did in a room where nobody was allowed to keep a record.
The remedy is not complicated. Require judges to state their reasons and cite their authority, so there is something real to review. Strip boilerplate recitals of any evidentiary weight, because they contain no evidence. Reverse the default, so that doubts arising from an incomplete record are resolved against the institution that failed to make one rather than the citizen who could not. And retire the presumption of equal consideration, because two decades of federal data and every honest practitioner's experience agree that it describes something that does not happen.