Good-Faith Exception

Good-Faith Exception

The attenuation factors just covered ask whether the chain between an illegality and a piece of evidence has weakened enough to let the evidence in. The good-faith exception puts a different question, which is whether the officers who gathered the evidence acted reasonably in relying on a warrant that a neutral magistrate had issued. In United States v. Leon, 468 U.S. 897 (1984), the Court held that the exclusionary rule does not require suppression of evidence obtained by officers acting in objectively reasonable reliance on a search warrant issued by a neutral magistrate, even though the warrant is later found to lack probable cause.¹ The good-faith exception therefore begins from an assumed defect, because the warrant is invalid and the search was unconstitutional in that sense. A violation does not by itself bring exclusion, since the rule applies only where its deterrent benefit outweighs the substantial social cost of losing reliable evidence.² Leon addressed the use of such evidence in the prosecution's case in chief.³

Before Leon, the Court had not recognized any good-faith exception to the Fourth Amendment exclusionary rule, but its decisions had been moving in that direction. It had limited who may invoke the rule to the victim of the police misconduct, as in Rakas v. Illinois, and it had refused to suppress the fruits of an arrest made in good-faith reliance on a statute later declared unconstitutional, as in Michigan v. DeFillippo.⁴ Those decisions reflected the balancing approach of cases such as United States v. Calandra, which weighs the cost of applying the rule in a particular setting against the added deterrence it would produce.⁵ Leon ran the same balance for searches under warrants and concluded that the substantial costs of exclusion outweigh its marginal benefits.⁶

The reasoning has two halves, one about the magistrate and one about the officer, and a cost side that completes the balance. The exclusionary rule exists to deter police misconduct and is not designed to punish the errors of judges and magistrates, who have no stake in the outcome of particular prosecutions and so have nothing to learn from the loss of evidence.⁷ The officer's position is the subtler half. Suppression works by teaching the particular officers, and their successors, to take greater care with the rights of the accused. An officer who has gone to a neutral magistrate and acted within the warrant has already done everything the law asks, and once the warrant issues there is nothing more the officer can do to comply.⁸ Penalizing the officer for the magistrate's mistake therefore cannot deter anything, and excluding the evidence could only make officers less willing to do their duty.⁹ On the other side of the scale the cost stays high, because probative evidence is lost and the factfinder must decide the case without it. When the benefit of exclusion falls toward zero and the cost remains, the balance favors admission.

Because the deterrence logic depends on what officers should have known, the test of reliance is objective. Courts ask whether a reasonably well-trained officer would have known that the search was illegal despite the magistrate's authorization.¹⁰ The Court itself later conceded that calling this objectively reasonable reliance good faith was perhaps confusing.¹¹ The common mistake is to ask whether the officer honestly believed the warrant was valid, when an honest belief alone does not satisfy the test. An objective standard also removes any incentive to stay ignorant. If sincere belief were enough, an officer who never learned the basic requirements for an affidavit or a warrant would be safer than one who had, whereas under the objective test the officer is charged with what a trained officer would know.

Reliance is reasonable in the ordinary case, but Leon identified four situations in which an officer cannot claim it. Each marks a point where the premise of the exception fails: either the officer rather than the magistrate is the source of the error, or the magistrate was not supplying the neutral check that reliance presupposes, or the defect was visible on the face of the papers.¹²

Step 1 does real work because Leon does not apply to improper execution of a search, only to flaws in the warrant itself or in the basis for its issuance.¹³ An officer who searches a room the warrant did not cover cannot invoke the magistrate's approval, since the magistrate approved nothing about that room. Steps 2 through 5 then test whether the reliance on what the magistrate did approve was reasonable.

The same reasoning reaches one kind of error that no officer could have prevented. In Massachusetts v. Sheppard, 468 U.S. 981 (1984), the companion case to Leon, the Court held that the exclusionary rule did not apply when a warrant was invalid because the judge forgot to make clerical corrections to it.¹⁴ The Court later extended the good-faith reasoning to a search made in reliance on a statute later declared unconstitutional and to an arrest made on mistaken information in a court's database, and Herring v. United States then took up errors made by police personnel themselves, a decision developed in a later section.¹⁵ The four limits come first, beginning with the baseline rule in its simplest form: officers who search under a facially valid warrant issued by a neutral magistrate.

Reasonable Reliance on a Warrant

The baseline rule is easiest to see in the case that created it. In United States v. Leon, the respondents were indicted for conspiracy to possess cocaine for distribution and related offenses, and part of the government's evidence came from a search by officers acting under a facially valid warrant.¹⁷ The District Court suppressed that evidence because the affidavit behind the warrant did not establish probable cause, and the Court of Appeals affirmed.¹⁸ In the lower courts, then, the insufficiency of the affidavit was enough to keep the evidence out. The Supreme Court asked a further question, namely whether the officers had acted in objectively reasonable reliance on a warrant issued by a detached and neutral magistrate. It held that the exclusionary rule does not bar use of evidence obtained that way in the prosecution's case in chief, even though the warrant was ultimately found unsupported by probable cause.¹⁹

On the facts before it, the Court found the reliance reasonable. The affidavit was supported by much more than a bare-bones recital, and the fact that it failed to set out the circumstances with the particularity needed to make the warrant valid did not negate the officers' objective good faith.²⁰ That result shows how generous the baseline is. A court may agree with the defendant that probable cause was missing and still admit the evidence, because the question has moved from whether the magistrate made the probable-cause call correctly to whether a reasonably well-trained officer would have known the search was illegal despite the magistrate's authorization.²¹ Thin affidavits can therefore sit on the protected side of the line, and the point at which an affidavit becomes too thin to rely on is the subject of the bare-bones limit developed below.

Stated as a rule, reasonable reliance on a warrant has three parts that must line up. The officers must have searched under a warrant and acted within its scope, the warrant must have come from a judge or magistrate who was detached and neutral, and the officers' reliance both on the magistrate's probable-cause determination and on the technical sufficiency of the warrant must have been objectively reasonable.²² The last part is where a warrant helps the prosecution most, because a warrant issued by a magistrate normally suffices to show that the officers acted in good faith, so searches under warrants rarely call for any deep inquiry into reasonableness.²³ The reason follows from the officer's position. In the ordinary case an officer cannot be expected to second-guess the magistrate's judgment that probable cause exists or that the form of the warrant is technically sufficient, which means that in most warrant cases there is no police illegality and therefore nothing for exclusion to deter.²⁴

Two qualifications keep the rule from being overstated. The first concerns the setting: the holding addresses use of the evidence in the prosecution's case in chief, and the question the Court framed was stated in those terms.²⁵ An exam answer should carry that qualifier rather than announcing that evidence from a defective warrant is admissible for every purpose. The second concerns the warrant itself. The Court said it did not suggest that exclusion is always inappropriate when an officer has obtained a warrant and abided by its terms, and it acknowledged circumstances in which an officer will have no reasonable grounds for believing the warrant was properly issued.²⁶ Students lose this issue when they treat the existence of a warrant as an automatic shield, which is why each of the four limits that follow must be checked before the evidence is admitted.

Courts apply the baseline with similar generosity to warrants whose defect lies in breadth rather than in probable cause. In United States v. Diaz, the First Circuit reversed a ruling that a warrant in a fraud investigation was facially deficient for lack of particularity, reasoning that given the complexity of the scheme a reasonably well-trained officer would not necessarily have known the search was illegal, and that even a well-trained investigator is not expected to be a legal technician and may rely on the magistrate's greater sophistication.²⁷ The Third Circuit reached a similar result for an overbroad warrant covering more than ten years of records in United States v. Ninety-Two Thousand Four Hundred Twenty-Two Dollars, observing that officers, who are rarely attorneys, may rely on the magistrate's judgment except in rare circumstances, as when a warrant is so plainly defective in form that even a lay officer could not believe it proper.²⁸ Both decisions rest on the premise that drives Leon: the magistrate, and not the officer, is the person charged with drawing the legal line.

Only one fact changed between the two versions, and it is the fact on which the rule is built. When officers go to a magistrate, the legal judgment about probable cause is made by someone with no stake in the outcome of the prosecution, and the officers have done all that can be asked of them.²⁹ When officers rely on their own assessment, the only judgment at issue is the police's own, and that is the conduct exclusion exists to discipline. The exception therefore gives the warrantless search no shelter. Whether some other doctrine, such as attenuation or inevitable discovery, rescues the evidence is a separate question.

The Court has since restated the baseline in the same terms: when police act under a warrant that is invalid for lack of probable cause, the exclusionary rule does not apply if they acted in objectively reasonable reliance on it.³⁰ That formulation is the one to carry into the four limits taken up next, each of which identifies a way in which reliance can fail to be reasonable.

Misleading the Magistrate

The baseline rule protects an officer who relies on a warrant because the legal judgment behind the warrant belonged to someone else. The first of the four limits applies when that premise fails in a particular way: the magistrate's judgment rested on information that the officer supplied and had no business supplying. Reliance is not objectively reasonable when the magistrate was misled by information in an affidavit that the affiant knew was false or would have known was false except for reckless disregard of the truth.³² The affiant is the person who swears to the affidavit, ordinarily the officer seeking the warrant, so the limit looks to that person's state of mind and not to the magistrate's. The limit is also commonly stated to reach false information the officer provides in seeking the warrant, and not only the words on the page.³³

The reason for the limit comes straight from the deterrence logic behind the exception. Exclusion exists to deter police misconduct, and it cannot be expected to deter magistrates, who have no stake in the outcome of particular prosecutions, so the exception admits the evidence when the mistake was the magistrate's.³⁴ A falsified affidavit reverses that picture. The error now begins with the officer, which puts the person the rule is meant to discipline at the origin of the problem, and the deterrent benefit of exclusion returns at full strength. A magistrate who decides on the papers the officer submits is also making a judgment about the facts as the affiant described them. When those facts were invented or recklessly stated, the approval is not a neutral judgment about the real situation, and the officer has no genuine judicial determination to lean on.

The incentive reason is subtler. If a magistrate's signature could cleanse a false affidavit, an officer who fell short of probable cause would have every reason to fill the gap with an embellishment, collect the warrant, and then invoke good faith, and the exception would reward the very conduct the exclusionary rule exists to discourage. Leon itself acknowledged, in a footnote, that a warrant affidavit shown after the fact to contain a deliberately or recklessly false statement cannot be placed beyond challenge, because that would impose on the magistrate's authority.³⁵ The limit therefore closes the one route by which an officer could manufacture the reliance that the exception protects.

The mental-state wording has two branches. The first covers an affiant who knew the statement was false. The second covers an affiant who would have known it was false except for reckless disregard of the truth, which reaches the officer who ignored obvious reasons for doubt instead of checking.³⁶ Neither branch describes an honest mistake. An affiant who accurately relays what a source reported, with no reason to doubt it, has neither known of any falsity nor disregarded the truth, so the limit does not apply and good faith can still be asserted, although the other three limits remain to be checked. The limit is also framed around a magistrate who was misled, which ties it to falsehoods that fed into the decision to issue the warrant.

The informant's story was false in all three versions, so falsity alone cannot be what moves the answer. The facts that change are what the affiant knew and what the affiant was willing to ignore. In the first and third versions the affiant is the source of the magistrate's mistake, and the premise of the exception collapses. In the second, nothing in the affiant's conduct calls for deterrence, so the baseline rule governs. This limit also differs in kind from the next one. Here the magistrate may have been perfectly neutral and careful, and the defect lies in what the affiant fed into the process, whereas the next limit asks whether the magistrate wholly abandoned the judicial role and so locates the failure in the issuing judge. In the screen laid out earlier, this limit is step 2, and a yes answer at that step makes good faith unavailable.

Abandoned Judicial Role

The second limit on good faith shifts attention from the officer's affidavit to the judge who acted on it. Reliance on a warrant is not objectively reasonable when the issuing magistrate wholly abandoned the judicial role, which happens when the judge acts as a rubber stamp for the police or as an adjunct of law enforcement instead of weighing the application independently.³⁸ A rubber stamp approves whatever is placed in front of it, and an adjunct joins the investigation instead of judging it. In both situations the warrant carries a judge's signature without reflecting a judge's judgment.

The reason for the limit follows from the logic of the baseline rule. Good faith protects an officer who relies on a warrant because exclusion is meant to deter police misconduct, and it cannot be expected to deter magistrates, who have no stake in the outcome of particular prosecutions.³⁹ That reasoning assumes the magistrate really is a person standing apart from the police, and the assumption holds when the judge reads the application and decides. It fails when the judge approves papers without reading them or takes part in the search, because the judge's role has then been absorbed into the police function. The officer cannot claim that independent review caught or excused anything, since no independent review occurred. An incentive problem also appears. If any signature could launder a search, officers would have reason to look for the most compliant judge available, and the exception would reward that search for a rubber stamp.

The limit marks the line between a wrong judge and a non-neutral judge. A neutral judge who reads the affidavit and makes an honest but mistaken probable-cause call is exactly the figure the baseline rule protects, since the error belongs to the judge and exclusion would not correct it. The word "wholly" in the formulation signals that the limit is reserved for a failure of the judicial role itself, and that a weak or debatable judgment falls short of it. The exception was built on the premise that there is no evidence magistrates are inclined to ignore or subvert the Fourth Amendment, so abandonment is the rare departure from that premise.⁴⁰ It also differs from the preceding limit. Misleading the magistrate places the fault in the affiant's statements, whereas this limit places it in the conduct of the issuing judge.

What separates the three versions is whether the judge exercised judgment, and the correctness of that judgment plays no part. Students lose this issue when they treat a judge's mistake as proof of abandonment, because a later holding that probable cause was lacking is the starting condition for every good-faith question and cannot also answer it. On an exam stem, look for facts showing that the judge did not read, did not decide, or joined the police effort. Absent those facts, a neutral judge's incorrect ruling leaves the baseline rule in place, and the remaining limits still have to be checked.

Bare-Bones Affidavits

The third limit on good faith turns from the people who handled the warrant to the paper that supported it. Reliance on a warrant is not objectively reasonable when the supporting affidavit is "so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable."⁴² The indicia of probable cause are the facts and signs in the affidavit from which a neutral reader could infer that evidence of a crime will be found in the place described. An affidavit that offers a bare conclusion in place of those facts is called a bare-bones affidavit. Along with the facially deficient warrant taken up next, this limit is among the two that courts see most often when warrants for business records are challenged.⁴³

The limit makes sense only once its two layers are separated. Every good-faith question begins after a court has already held that the affidavit failed to establish probable cause, so that holding supplies the premise and cannot also be the answer. Insufficiency puts the case into the good-faith analysis, and the limit then asks something further: was the affidavit so thin that an officer should have seen it gave the magistrate no basis for a decision? The inquiry is objective, which means the officer's sincere belief that the affidavit was enough does not decide it. The question is what a reasonable officer would have believed about the affidavit's contents.⁴⁴

The reason for the limit comes from what the baseline rule assumes. Good faith protects an officer who relies on a warrant because exclusion is meant to deter police misconduct and cannot be expected to deter magistrates, who have no stake in the outcome of particular prosecutions.⁴⁵ That reasoning assumes the officer gave the magistrate something to decide. The affidavit is the officer's own work product, and a reasonably well-trained officer can recognize one that hands the magistrate nothing to evaluate. Deference to the magistrate's signature is unreasonable when the officer should have seen that the signature rested on nothing. An incentive problem sits behind this as well. If any signature could save a search, an officer with a hunch could submit a single conclusory sentence, count on a magistrate to approve it, and keep the evidence even if a court later disagreed, so that careless drafting would cost nobody anything. Suppressing the fruits of a bare-bones affidavit gives the officer a reason to put facts on the page.

The contrast that defines the limit appears in United States v. Leon itself. The district court there suppressed the evidence because the affidavit did not establish probable cause, and the Supreme Court nonetheless admitted it, reasoning that the affidavit was supported by much more than a "bare bones" affidavit.⁴⁶ The affidavit had not set out the circumstances with enough particularity to make the warrant valid, but that shortfall did not negate the officers' objective good faith.⁴⁷ What kept the case on the admissible side of the line was that the affidavit gave the magistrate real material to weigh. A reasonable officer could think such material enough even though a court later disagreed.

Both versions end with a court holding that probable cause was missing, so the holding cannot be what moves the answer. The fact that changes is how much the affidavit gave the magistrate to evaluate. In the first version there was nothing to weigh and every reasonable officer would have seen that. In the second there was something to weigh, and the disagreement between officer and court was the kind that Leon tolerates. Students lose this issue when they treat an insufficient affidavit as automatically a bare-bones one, because that equation would erase the baseline rule: every warrant that is later found invalid for lack of probable cause would then fall within the limit. On an exam stem, look for an affidavit made up of conclusions, such as an officer's belief or an unnamed source's assertion, with no facts showing how anyone knows what is claimed and nothing corroborated.

This limit also differs in kind from the two before it. The limit on misleading the magistrate asks what the affiant knew, and the limit on an abandoned judicial role asks what the judge did. This one asks only what the affidavit says and what a reasonable officer would make of it, so it can apply even when the magistrate was neutral and the officer was entirely honest. The next limit looks at the warrant itself instead of the affidavit behind it.

Facially Deficient Warrants

The fourth limit on good faith looks at the warrant itself, the document the executing officers carry into the search. Reliance on a warrant is not objectively reasonable when the warrant is so facially deficient that the executing officers cannot reasonably presume it to be valid.⁴⁹ The standard example is a failure to particularize, meaning the warrant does not describe the place to be searched or the things to be seized with enough specificity to tell the officers what they may enter and take.⁵⁰ A warrant is facially deficient when the flaw shows on the face of the paper, so that anyone reading the text would see the problem without investigating anything outside it.

The reason for the limit follows from who holds the document. The earlier limits concern events that executing officers may know nothing about: what the affiant told the magistrate, how carefully the judge reviewed the application, and how much the affidavit contained. The warrant is different because it sits in the hands of the officers who carry out the search, and they can read it before they act. Good faith protects reliance on a warrant because exclusion is meant to deter police misconduct, and an officer who acts on a judge's approval has done what the system asks.⁵¹ That description fits poorly when the paper says nothing about where to search or what to take, because an officer who proceeds on such a document has relied on no judicial determination about place or things. An incentive problem sits behind the limit as well. If any signature on any form could save a search, executing teams would have no reason to read what they carried, and careless paperwork would cost nobody anything. Suppression gives officers a reason to check the face of the warrant before they enter.

The inquiry is objective and is pitched at the ordinary officer, who is rarely a lawyer. In United States v. Ninety-Two Thousand Four Hundred Twenty-Two Dollars the Third Circuit applied Leon to a warrant authorizing seizure of more than ten years of records and explained that officers are entitled to rely on the magistrate's judgment except in rare circumstances, such as a warrant so plainly defective in form that even a lay officer could not believe in good faith that it was proper.⁵² The First Circuit reached a similar result in United States v. Diaz, a fraud case in which a district court had found the warrant facially deficient on particularity. The appeals court reversed, reasoning that the scheme was complex and ongoing and that a reasonably well-trained officer would not necessarily have known the search was illegal.⁵³ It added that separating good records from bad is inherently difficult, and that an investigator is not expected to be a legal technician and may rely on the magistrate to know where to draw the line.⁵⁴ What separates those decisions from a true facial defect is the kind of error involved. Whether a category of records may stretch across several years is a question of legal line-drawing on which reasonable officers and courts disagree, while an omitted address or an empty description of items is visible to a layperson holding the paper.

Both warrants contain an error that a reader can see, so visibility alone cannot be what moves the answer. The fact that changes is what the error touches. In the first version the missing information is the very information the warrant exists to supply, whereas in the second the information telling officers where to go and what to take is intact. Students lose this issue when they treat a warrant that a court later calls overbroad as automatically facially deficient, because a later finding of invalidity is the starting condition for every good-faith question and cannot also answer it. On an exam stem, look for a warrant that omits or garbles the place or the items in a way evident to anyone reading the document. A flaw that surfaces only through legal analysis or later investigation points toward the baseline rule instead.

This limit also differs from the one before it. The bare-bones limit asks what the affidavit says, and this limit asks what the warrant says, so the two can diverge: a thin affidavit may support a well-drafted warrant, and a richly supported application may produce a poorly drafted one. The limit examines the warrant as issued, and Leon does not rescue an officer who executes a facially valid warrant improperly, since its good-faith analysis concerns flaws in the warrant or in the basis for its issuance.⁵⁵ A defective-looking form does not always defeat reasonable reliance, however. In Massachusetts v. Sheppard, the companion case to Leon, the Court applied the good-faith analysis to permit use of evidence even though the warrant was found invalid because it was overly broad.⁵⁶ The next section takes up that decision and the circumstances in which an error in the paper leaves reliance reasonable.

Clerical Errors by the Judge

The four limits just covered mark the points at which reliance on a warrant stops being reasonable. A different problem arises when the warrant is wrong only because the judge slipped in completing it, after the officers had done everything the process asked of them. In Massachusetts v. Sheppard, the companion to Leon, the Court held that the exclusionary rule does not apply to evidence seized under a warrant invalidated by a technical error of the issuing judge, so long as the officers' reliance on the warrant was objectively reasonable.⁵⁸ The Court framed the sole question as whether the officers reasonably believed that the search they conducted was authorized by a valid warrant. No one disputed that the officers actually held that belief, so the inquiry came down to whether the belief had an objectively reasonable basis.⁵⁹

The facts show how much the officers did. Police suspected Osborne Sheppard in the beating death of Sandra Boulware, whose burned body had been found in a vacant lot in Boston, and on a Sunday they sought a warrant to search his residence.⁶⁰ The local court was closed, and Detective O'Malley could find only a form previously used in another district for controlled-substance searches. He typed out the controlled-substance subtitle, substituted the correct neighborhood, and typed in Sheppard's name and address, but the reference to controlled substances remained in the part of the form that would become the warrant once signed.⁶¹ O'Malley took that form and his affidavit, which the district attorney had reviewed and approved, to a judge's home. He told the judge that he knew the form dealt with controlled substances and showed him what had been crossed out. The judge said he would make the necessary changes, altered parts of the form, signed it, and returned it with the statement that it was sufficient authority in form and content for the requested search. He had not changed the substantive portion, so the warrant still authorized a search for controlled substances, and he had not incorporated the affidavit into it.

The trial judge concluded that the warrant failed the Fourth Amendment because it did not particularly describe the items to be seized, but admitted the evidence because the police had acted in good faith.⁶² A plurality of the Massachusetts Supreme Judicial Court accepted that the police had searched in reasonable good faith, yet ordered exclusion because this Court had not yet recognized a good-faith exception.⁶³ The Supreme Court, having already decided in Leon that reasonable reliance on a warrant later found invalid does not trigger exclusion, agreed with the trial court that the officers' belief was reasonable, because they took every step that could reasonably be expected of them.⁶⁴ The detective prepared a reviewed affidavit, presented it to a neutral judge who found probable cause, warned that the form might need changes, was told the changes would be made, and watched the judge make some of them. At that point a reasonable officer would have concluded, as O'Malley did, that the warrant authorized a search for the materials listed in the affidavit.

Sheppard answered that O'Malley knew the form was defective and should therefore have read the signed warrant to confirm that the corrections had been made. The Court rejected the premise. Whatever an officer may be required to do when executing a warrant without knowing beforehand what items are to be seized, the Court refused to require an officer to disbelieve a judge who has just advised him, by word and by action, that the warrant authorizes the search he requested.⁶⁵ The determinations of a judge acting within his jurisdiction, even if erroneous, are valid and binding until set aside through a recognized procedure. An officer who must accept at face value a judge's conclusion that a form is invalid can hardly be expected to disregard the judge's assurance that everything is in order, especially after calling the problem to the judge's attention.⁶⁶

Two separate supports hold the rule up. The first concerns the officer's conduct. Exclusion exists to deter police misconduct, and deterrence can only operate on choices that officers control. O'Malley's choices were the ones the system asks of an officer: he drafted a proper affidavit, obtained the prosecutor's review, went to a neutral judge, and disclosed the form's problem. Suppressing the evidence would punish him for a gap he did not create and could not have closed without distrusting the judge, and it would teach later officers nothing about what to do differently. The second support concerns the judge's role. Because a judge's rulings bind until set aside, an officer who has the judge's assurance has no practical means of overriding it on the spot, and a rule requiring officers to audit signed warrants against the judge's word would make the person who must defer to the judge responsible for supervising the judge. Later decisions describe the case in these terms. Herring v. United States cites it for the holding that the exclusionary rule did not apply when a warrant was invalid because a judge forgot to make "clerical corrections" to it.⁶⁷ Herring also recounts that the Court in Evans refused to suppress evidence after a mistake by a court employee, partly because the rule was crafted to curb police rather than judicial misconduct.⁶⁸

Sheppard can look inconsistent with the facially deficient warrant limit, because the paper O'Malley carried described the wrong items and a reader could have seen the flaw. The way to reconcile them is to ask what the officer had been told. In the facially deficient case the defect is apparent and nothing suggests anyone addressed it, so reliance on the paper alone is unreasonable. In Sheppard the defect was also apparent in principle, but the officer had identified it to the judge and been assured that it would be fixed, which gave him a reasonable basis for believing it had been. The reliance that good faith protects in this setting is reliance on the judge's word about the warrant as well as on the warrant's text.

The flawed paper is present in both versions, so the flaw cannot be what moves the answer. The facts that change are the officer's disclosure of the problem and the judge's assurance of a cure. Students go wrong in two directions. Some treat an officer's awareness that a form needed changes as defeating good faith by itself, although O'Malley knew exactly that and the evidence was still admitted. Others treat any judicial slip as automatically excused, which overlooks that Sheppard rested on the officers' having raised the problem and received an assurance. On an exam stem, look for an officer who flags a problem, a judge who promises or represents a cure, and a warrant that still contains the error. Then confirm that the other limits are absent, meaning a neutral judge, a truthful affidavit, and an affidavit that gave the judge real material to weigh.