Benny McKinney was arrested on July 9, 1988, after a search of the residence located at 2238 East Laurel Street in Springfield, Illinois. Drugs and firearms were confiscated during the search and McKinney was subsequently found guilty of possession with intent to distribute cocaine in violation of 21 U.S.C. Sec. 841(a)(1), use of a firearm during and in relation to a drug trafficking crime in violation of 18 U.S.C. Sec. 924(c)(1), and unlawful possession of a firearm by a convicted felon in violation of 18 U.S.C. Sec. 922(g)(1). McKinney appeals, challenging the validity of the search, the sufficiency of the evidence to support his convictions, and the appropriateness of the sentences he received. We affirm his convictions on all counts, but remand for resentencing.

On July 8, 1988, Sergeant Kettlekamp of the Illinois State Police signed an affidavit which related that an individual named Carla Brown came to him and stated that on July 7, she was in the residence at 2238 East Laurel and observed cocaine, marijuana, drug paraphernalia, and firearms. She stated that the cocaine was located in the front bedroom and that the firearms were in the closet in the front bedroom. The affidavit also stated that Brown observed McKinney distribute one-half gram of cocaine to a female customer and, that later that day, as Brown was walking along the street, McKinney shot at her with a machine gun. A search warrant was issued by a United States Magistrate based solely on the information in the officer's affidavit.

On July 9, the warrant was executed by agents of the Illinois State Police and the Bureau of Alcohol, Tobacco, and Firearms. After announcing their offices, the agents entered the residence and observed McKinney in the front bedroom standing next to the bed. A search of the bedroom produced a packet of white powder containing 4 grams of a cocaine/inositol mixture and a can of Right Guard deodorant with a false bottom which contained seven plastic bags of cocaine. Six firearms were also discovered; a twenty-two caliber pistol found in a dresser drawer, a .32 handgun in another dresser drawer, a twenty-two caliber pistol with a loaded ammunition clip on the shelf in the closet, a twenty-two caliber rifle in the closet, a .357 magnum in a briefcase under the bed, and a rifle and ammunition clip under the bed. Also discovered in the house were a triple beam scale, a bottle of inositol (used to cut the cocaine), and approximately forty rounds of ammunition. The defendant was subsequently arrested.

A thirteen count superseding indictment was filed on September 21, 1988, charging McKinney as follows: Count I, possession with intent to distribute cocaine in violation of 21 U.S.C. Sec. 841(a)(1); Counts II-VII, use of a firearm during and in relation to a drug trafficking crime in violation of 18 U.S.C. Sec. 924(c)(1); and Counts VIII-XIII, unlawful possession of a firearm by a convicted felon in violation of 18 U.S.C. Sec. 922(g)(1).

On November 2, 1988, the defendant filed a petition challenging the facial validity of the search warrant and moving to quash the search warrant and suppress the evidence. Specifically, McKinney contended that Brown was a confidential informant whose information should have been corroborated before the issuance of a search warrant. Furthermore, McKinney asserted, corroboration of Brown's statements was unlikely because her reliability was suspect as she has a prior criminal record and was a drug addict. The district court, however, was persuaded that the warrant was adequately supported by the affidavit and that the information was sufficient under the totality of the circumstances test of Illinois v. Gates, 462 U.S. 213, 238, 103 S. Ct. 2317, 2332, 76 L. Ed. 2d 527 (1983). In this respect, the court stated, "Brown claimed to have witnessed and been victim of criminal acts, see United States v. Wilson, 479 F.2d 936, 940 (7th Cir.1973), ... Brown's name was included in the affidavit, see United States v. Spach, 518 F.2d 866, 870 (7th Cir.1975), and the fact that Brown provided highly detailed and specific, as opposed to general information was sufficient reliability for the affidavit." Accordingly, the district judge determined the issuing magistrate had a substantial basis from which to conclude that probable cause existed and denied McKinney's petition.

After McKinney waived his right to a jury trial, the court found McKinney guilty of all counts as charged. The court sentenced the defendant to 121 months on Count I, the possession with intent to distribute cocaine charge, and 60 months on each of counts VIII-XIII, unlawful possession of a firearm by a convicted felony, to be served concurrently with each other and with Count I. The court also imposed independent, five-year, consecutive sentences on Counts II-VII, use of a firearm during a drug crime to be served consecutively to Count I. A six year probation period after release from prison was also ordered. McKinney appeals, and we affirm the convictions but remand for resentencing consistent with this opinion.

McKinney first contests the district court's denial of his motion to quash the search warrant and suppress the evidence. He contends, as he did in the district court, that the statements by Brown do not provide sufficient reliability to support a warrant with probable cause. The evidence seized on the basis of the warrant was therefore tainted, he asserts, and should be excluded. The district court, in denying defendant's motion to suppress, found there was sufficient probable cause to issue a warrant to search the residence.1

When reviewing a decision to issue a warrant, our task is "simply to ensure that the magistrate had a substantial basis for concluding that probable cause existed." Illinois v. Gates, 462 U.S. 213, 250, 103 S. Ct. 2317, 2338, 76 L. Ed. 2d 527 (1983); see also United States v. Zambrana, 841 F.2d 1320, 1332 (7th Cir.1988); United States v. Griffin, 827 F.2d 1108, 1111 (7th Cir.1987).2 In Gates, the Supreme Court rejected the rigid two-pronged "Aguilar-Spinelli " test for probable cause, adopting in its stead a "totality of the circumstances" test. At the same time, the Court concluded that adoption of "the flexible, common-sense standard" embodied in the totality of the circumstances approach required a more deferential standard of appellate review than a de novo standard of appellate review, recognizing that "[a] grudging or negative attitude by reviewing courts toward warrants is inconsistent with the Fourth Amendment's strong preference for searches conducted pursuant to a warrant...." 462 U.S. at 236, 103 S.Ct. at 2331. The Court observed that, "so long as the magistrate had a 'substantial basis for ... conclud[ing]' that a search would uncover evidence of wrongdoing, the Fourth Amendment requires no more," id. at 236, 103 S. Ct. at 2331, and concluded that substantial basis review provides a "flexible, easily applied standard [to] achieve the accomodation of public and private interests that the Fourth Amendment requires...." Id. at 239, 103 S.Ct. at 2332.

Nevertheless, Judge Posner's concurrence disparages the utility and feasibility of applying the Gates standard. In its place, Judge Posner "proposes a change in the formulation of the standard of review" of probable cause determinations, and urges us to take two steps: first, to abandon the substantial basis approach and to employ instead a clearly erroneous standard when reviewing probable cause determinations in warrant cases; and second, to abandon de novo review of probable cause determinations in nonwarrant cases and to adopt the clearly erroneous standard for reviewing all probable cause determinations. To justify these bold initiatives, Judge Posner argues that the difference between the substantial basis and clearly erroneous review is a matter of semantics and is, at any rate, at odds with "a steady trend toward limiting the scope of appellate review of determinations of fact-specific issues." Infra at 420. And lest the resistance to the tide of change prove too unyielding, Judge Posner suggests that any attempt to preserve a less deferential degree of review in warrant cases is futile, since in United States v. Leon, 468 U.S. 897, 104 S. Ct. 3405, 82 L. Ed. 2d 677 the Court decided to "scuttle" the exclusionary rule by adopting a "good faith" exception. According to Judge Posner, we might as well abandon ship.3

We note initially that, as circuit judges, we must await an explicit order from our superiors before scuttling the fourth amendment. The Supreme Court has repeatedly employed the "substantial basis" standard without ever equating it to the clear error standard. See, e.g., New York v. P.J. Video, Inc., 475 U.S. 868, 876, 106 S. Ct. 1610, 1615, 89 L. Ed. 2d 871 (1986) ("duty of reviewing court is simply to ensure that the magistrate had a substantial basis ...") (citation omitted); Leon, 468 U.S. at 915, 104 S.Ct. at 3416 ("reviewing courts will not defer to a warrant based on an affidavit that does not provide the magistrate with a substantial basis for determining the existence of probable cause"); Massachusetts v. Upton, 466 U.S. 727, 733, 104 S. Ct. 2085, 2088, 80 L. Ed. 2d 721 (1983) ("Examined in light of Gates, [the] affidavit provides a substantial basis for issuance of the warrant."); Gates, supra; United States v. Harris, 403 U.S. 573, 579, 91 S. Ct. 2075, 2079-80, 29 L. Ed. 2d 723 (1971) ("The Court of Appeals ... believed ... that there was no substantial basis for believing that the tip was truthful."); Jones v. United States, 362 U.S. 257, 271, 80 S. Ct. 725, 736, 4 L. Ed. 2d 697 (1960) ("there was a substantial basis for ... conclud[ing] that narcotics were probably present in the apartment"). In the absence of any evidence at all that the Court employs the two standards interchangeably, one should not independently conclude that they mean the same thing. The Court has certainly not abandoned use of the "clearly erroneous" standard, see, e.g., Citibank v. Wells Fargo Asia, Ltd., --- U.S. ----, 110 S. Ct. 2034, 2040, 109 L. Ed. 2d 677 (1990); Clemons v. Mississippi, --- U.S. ----, 110 S. Ct. 1441, 1459, 108 L. Ed. 2d 725 (1990), and we should not presume that it is implicitly invoking "clearly erroneous" when it employs other terms to define a standard of review. Cf. Michigan v. Mosley, 423 U.S. 96, 109, 96 S. Ct. 321, 329, 46 L. Ed. 2d 313 (1975) (White, J., concurring) ("The Court showed ... that when it wanted to create a per se rule ... it knew how to do so."). Courts of Appeals are not at liberty to treat the substantial basis language as a semantic sleight of hand; it is a distinct prescription for review that the Court has fashioned and consistently applied to the unique requirements of the fourth amendment.

Unlike Judge Posner, we do not read Leon to contain an implicit order to abandon close scrutiny of warrants. Indeed, in adopting the "good faith" exception, the Court explicitly disavowed that notion, stating that "[t]he good faith exception for searches conducted pursuant to warrants is not intended to signal our unwillingness strictly to enforce the requirements of the Fourth Amendment." 468 U.S. at 924, 104 S.Ct. at 3421. Moreover, the Court cautioned that "[d]eference to the magistrate ... is not boundless," 468 U.S. at 914, 104 S.Ct. at 3416, and reiterated that "courts will not defer to a warrant based on an affidavit that does not 'provide the magistrate with a substantial basis for determining the existence of probable cause.' " Id. at 915, 104 S. Ct. at 3416 (quoting Gates, 462 U.S. at 239, 103 S. Ct. at 2332-33). Judge Posner apparently assumes that there is no need to review probable cause determinations unless exclusion of the evidence is at stake, but the Court rejected this position as well, observing that "a reviewing court may properly conclude that, notwithstanding the deference that magistrates deserve, the warrant was invalid because the magistrate's probable-cause determination reflected an improper analysis of the totality of the circumstances or because the form of the warrant was improper in some respect." Id. at 915, 104 S. Ct. at 3416-17 (citation omitted); see also Gates, 462 U.S. at 223, 103 S.Ct. at 2324 ("The question whether the exclusionary rule's remedy is appropriate in a particular context has long been regarded as an issue separate from the question whether the Fourth Amendment rights of the party seeking to invoke the rule were violated by police conduct."). The Court concluded that the need to resolve fourth amendment questions to provide guidance to magistrates and law enforcement officials will often prompt courts to reach the question of whether the police violated constitutional rights before deciding whether they acted in good faith. Id. at 926, 104 S.Ct. at 3422.

Moreover, the Court observed that "it frequently will be difficult to determine whether the officers acted reasonably without resolving the Fourth Amendment issue." Leon, 468 U.S. at 925, 104 S.Ct. at 3422. Leon and Gates both made clear that exclusion remains a proper remedy when police rely on a warrant that was issued without adequate review by the magistrate; the "good-faith inquiry is confined to the objectively ascertainable question whether a reasonably well trained [sic] officer would have known that the search was illegal despite the magistrate's authorization." Leon, 468 U.S. at 922 n. 23, 104 S. Ct. at 3420 n. 23; see also id. at 923, 104 S. Ct. at 3420-21 (citing examples of inadequate review); Gates, 462 U.S. at 239, 103 S.Ct. at 2333 ("In order to ensure that such an abdication of the magistrate's duty does not occur, courts must continue to conscientiously review the sufficiency of affidavits on which warrants are issued."). Leon itself was not a case in which a magistrate lacked a substantial basis for deciding that probable cause existed; the officer's reliance on the warrant was reasonable because the magistrate had a substantial basis for issuing the warrant. Id. at 926, 104 S.Ct. at 3422 ("The affidavit ... provided evidence sufficient to create disagreement ... as to the existence of probable cause. Under these circumstances, the officers' reliance on the magistrate's determination of probable cause was objectively reasonable...." (emphasis supplied)); see also United States v. Hove, 848 F.2d 137, 139 (9th Cir.1988) ("The test for reasonable reliance is whether the affidavit was sufficient to 'create disagreement among thoughtful and competent judges as to the existence of probable cause.' ") (quoting Leon, at 926, 104 S. Ct. at 3422). Clearly, then, Leon did not obviate the need for reviewing courts to determine whether there was a substantial basis in the affidavit for granting a warrant.

More generally, we take issue with the view that our Circuit's law regarding appellate review of probable cause determinations is at odds with a "trend" that treats the application of law to fact as "fact" for the purposes of appellate review. The "steady trend" the concurrence cites as limiting appellate review of fact-related issues in the name of judicial economy is, we assert, neither steady nor a trend.4 The "difficulty of distinguishing between legal and factual issues," Cooter & Gell v. Hartmarx Corp., --- U.S. ----, 110 S. Ct. 2447, 2458, 110 L. Ed. 2d 359 (1990), leads us to acknowledge "the practical truth that the decision to label an issue a 'question of law,' a 'question of fact,' or a 'mixed question of law and fact' is sometimes as much a matter of allocation as it is of analysis."5 Miller v. Fenton, 474 U.S. 104, 106 S. Ct. 445, 451, 88 L. Ed. 2d 405 (1985) (emphasis supplied). "Where ... the relevant legal principle can be given meaning only through its application to the particular circumstances of a case, the Court has been reluctant to give the trier of fact's conclusions presumptive force and, in so doing, strip a federal appellate court of its primary function as an expositor of the law." Id.

This is particularly true when constitutional rights are at stake. The Supreme Court manifested its reluctance to abrogate the role of appellate courts in defining the law of probable cause required for warrants by incorporating an approach less deferential than clearly erroneous review. Judge Posner infers from the adoption of a clearly erroneous standard in some fact-intensive contexts that clearly erroneous is an appropriate standard for all such contexts.6 The fact that the Court has advanced the clearly erroneous standard in the Rule 11 context, or that we use it in negligence cases, says little about its appropriateness in reviewing probable cause determinations; there is--and should be--a significant difference between reviewing litigation sanctions imposed by a trial court and alleged violations of rights secured by the fourth amendment. Constitutional review is often of a different order than other appellate review, because constitutional issues are often of a greater magnitude than other legal issues. There is nothing either extraordinary or inexplicable about the proposition that constitutional rulings--even fact specific ones--should be reviewed under a more demanding standard.7 See, e.g., Neil v. Biggers, 409 U.S. 188, 193 n. 3, 93 S. Ct. 375, 379 n. 3, 34 L. Ed. 2d 401 (1972) ("It is said that we should not '... reverse findings of fact ... unless shown to be clearly erroneous.' This rule of practice ... is a salutary one to be followed where applicable. We think it inapplicable here where the dispute between the parties is not so much over the elemental facts as over the constitutional significance to be attached to them.").

Judge Posner cites "the dramatic increases in federal judicial workloads" as additional justification for restricting the scope of appellate review of fourth amendment issues to the clearly erroneous standard, but this factor should actually counsel against adopting a clearly erroneous standard, since, under the pressures of mounting caseloads, magistrates and district judges would not be better positioned to conduct thorough reviews of warrant applications. Goals--like achieving uniformity and efficiency--that inform the allocation of responsibility between trial and appellate courts have diminished import when weighed against the need to safeguard constitutional rights; expediency is a weak substitute for meaningful appellate review of alleged fourth amendment violations. In Malley v. Briggs, 475 U.S. 335, 345-46, 106 S. Ct. 1092, 1098-99, 89 L. Ed. 2d 271 (1986), the Court cited the same pressures as justification for resisting erosion of fourth amendment rights. Justice White, who also authored Leon, stated:

It is true that in an ideal system an unreasonable request for a warrant would be harmless, because no judge would approve it. But ours is not an ideal system, and it is possible that a magistrate, working under docket pressures, will fail to perform as a magistrate should. We find it reasonable to require the officer applying for the warrant to minimize this danger by exercising reasonable professional judgment.

Gates and Leon, then, clearly cannot be read as signals for us to curtail the exercise of our own reasonable professional judgment.

In making the case for clarity and uniformity, Judge Posner blurs our law by failing to distinguish warrant and nonwarrant cases, and by claiming that we review probable cause determinations in both de novo.8 As noted, we have consistently followed Gates in limiting our review of probable cause determinations in warrant cases to whether there was a substantial basis for finding probable cause. It is only in nonwarrant cases that we apply a de novo review of probable cause determinations.9 See, e.g., United States v. Towns, 913 F.2d 434, 439-40 (7th Cir.1990) (a "district court's legal determination of probable cause" for warrantless arrest is reviewed de novo ); United States v. Lima, 819 F.2d 687, 688 (7th Cir.1987) (same). Judge Posner asks what sense it makes to give a magistrate's finding greater deference than the district court's, but we respectfully submit that this question does not accurately frame the issue. Deference is determined not by who is making the determination, but rather by the context in which the determination is made. The district court is entitled to the same deference given to the magistrate on those rare occasions when the judge actually issues the warrant; by the same token, we would review de novo a magistrate's determination of probable cause in a nonwarrant context if that situation would ever arise.

Once this distinction is clarified, it is easy to see why we review probable cause determinations underlying warrants less intensively than those made in nonwarrant cases. The fourth amendment prohibits unreasonable searches, and the Supreme Court has long acknowledged that, as a general rule, warrantless searches are "presumptively unreasonable." Leon, 468 U.S. at 960, 104 S.Ct. at 3446 (Stevens, J., concurring in part); see also United States v. Ross, 456 U.S. 798, 807, 102 S. Ct. 2157, 2163-64, 72 L. Ed. 2d 572 (1982) (" 'In cases where the securing of a warrant is reasonably practicable, it must be used.' ") (quoting Carroll v. United States, 267 U.S. 132, 156, 45 S. Ct. 280, 286, 69 L. Ed. 543 (1924)); Payton v. New York, 445 U.S. 573, 586, 100 S. Ct. 1371, 1380, 63 L. Ed. 2d 639 (1980); Katz v. United States, 389 U.S. 347, 357, 88 S. Ct. 507, 514, 19 L. Ed. 2d 576 (1967) ("Searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment--subject only to a few specifically established and well-delineated exceptions."). When the search is conducted with a warrant, however, the intrusion is less egregious because the warrant assures "the individual whose property is searched or seized of the lawful authority of the executing officer, his need to search, and the limits of his power to search." Gates, 462 U.S. at 236, 103 S. Ct. at 2331 (quoting United States v. Chadwick, 433 U.S. 1, 9, 97 S. Ct. 2476, 2482, 53 L. Ed. 2d 538 (1977)). Moreover, police action "without a warrant bypasses the safeguards provided by an objective predetermination of probable cause, and substitutes instead the far less reliable procedure of an after-the-event justification for the arrest or search, too likely to be subtly influenced by the familiar shortcomings of hindsight judgment." Beck v. Ohio, 379 U.S. 89, 96, 85 S. Ct. 223, 228, 13 L. Ed. 2d 142 (1964). Consequently, we apply a higher level of review in nonwarrant cases than in warrant cases. The mitigating effect of a warrant has even prompted the Court to suggest that the probable cause standard itself is stricter in nonwarrant cases, noting that "in a doubtful or marginal case a search under a warrant may be sustainable where without one it would fall." Leon, 468 U.S. at 914, 104 S. Ct. at 3416 (quoting United States v. Ventresca, 380 U.S. 102, 106, 85 S. Ct. 741, 744, 13 L. Ed. 2d 684 (1965)); see also Jones, 362 U.S. at 270, 80 S.Ct. at 736 ("when the officer does not have clearly convincing evidence of the immediate need to search, it is most important that resort be had to a warrant") (emphasis supplied). By following different approaches in warrant and nonwarrant cases, then, we have not, as Judge Posner suggests, been hiding our heads in the sand; to the contrary, we have been vigilantly observing the Court's strictures and signals concerning warrants.

Our decision in Llaguno v. Mingey, 763 F.2d 1560 (7th Cir.1985) (en banc) is not inconsistent with our approach to probable cause determinations made by judges and magistrates. In Mingey, we held that, in damages trials for alleged civil rights violations involving the fourth amendment, probable cause is a jury question subject to the clearly erroneous standard. Again, the context in which the probable cause determination was made is critical. At trial, a jury cannot find probable cause unless the jurors agree unanimously that the preponderance of the evidence establishes it. This degree of review compensates for a lesser degree of appellate review--and ensures an adequate degree of certainty about the probable cause determination--but only in the context of civil cases; we do not allow juries to make probable cause determinations in criminal trials. Indeed, we do not always allow juries to make them in civil trials. See, e.g., Gramenos v. Jewel Cos., 797 F.2d 432, 438-39 (7th Cir.1986) (Easterbrook, J., joined by Posner and Campbell, JJ.) (affirming summary judgment on issue of probable cause while noting that a jury's determination that a warrantless arrest was unreasonable would not preclude court's legal determination that police had probable cause). Our willingness to treat probable cause as an issue of fact turns on our willingness to merge the concepts of probable cause and reasonableness.10 In many contexts, however, we have found it both possible and "useful" "to have an understandable definition of probable cause even though 'reasonableness' remains as a separate issue." Id. at 439.

With all due respect, we also question the premise of Judge Posner's desire to minimize the number of legal formulas to be applied by judges. He urges, in the name of simplicity and uniformity, that we abandon the substantial basis standard in favor of the clearly erroneous standard. We agree, of course, that uniformity and simplicity are desirable goals. We suggest, however, that it is with appropriate intermediate appellate review that uniformity in magistrate's probable cause decisions can best be achieved. Judge Posner correctly notes that facts presented in a particular case will never recur exactly. Nonetheless, the nature of the conduct in cases may be sufficiently similar so that less deferential probable cause determinations by this Court will provide meaningful guidance to the district courts.11

Judge Posner apparently acknowledges that judges can, and must, give varying degrees of scrutiny depending on the type of case before them, for he maintains that the clearly erroneous standard can, and should, be applied with varying degrees of rigor. It seems, then, that his quarrel is not so much with the proposition that probable cause determinations require more appellate scrutiny but rather with the feasibility of applying the levels of review mandated by the Court.12 It is not obvious, however, that a chameleonic "clearly erroneous" standard offers any greater assistance to reviewing courts. "Words are useful only to the extent they distinguish some things from others; symbols that comprise everything mean nothing." Chicago Mercantile Exchange v. SEC, 883 F.2d 537, 547 (7th Cir.1989) (Easterbrook, J.). It is not clear how our exclusive adoption of the clearly erroneous standard would be interpreted. In what cases would reviewing judges apply the "clearly erroneous" standard rigorously and in which would they treat it as it is more commonly perceived--as a symbol of appellate acquiescence? At a minimum, the Court's adoption of a standard distinct from the de novo and clearly erroneous standards tells courts that they "must continue to conscientiously review the sufficiency of affidavits on which warrants are issued," Gates, 462 U.S. at 239, 103 S.Ct. at 2333. We fail to see how a single standard, applied differently according to the context of a case, will clarify or standardize the law more than different levels of review, each applied to specific classes of cases. Finally, one must question the "honesty" of an approach that camouflages multiple degrees of appellate review with the same label. Thus, although constrained by Supreme Court precedent to review probable cause determinations underlying warrants for a substantial basis, we do so without difficulty or reluctance.

Based on these principles, we hold that "a magistrate's determination of probable cause is to be given considerable weight and should be overruled only when the supporting affidavit, read as a whole in a realistic and common sense manner, does not allege specific facts and circumstances from which the magistrate could reasonably conclude that the items sought to be seized are associated with the crime and located in the place indicated." Griffin, 827 F.2d at 1111 (7th Cir.1987) (quoting United States v. Pritchard, 745 F.2d 1112, 1120 (7th Cir.1984)). An affidavit has made a proper showing of probable cause when it sets forth facts sufficient to induce a reasonably prudent person to believe that a search thereof will uncover evidence of crime. Berger v. New York, 388 U.S. 41, 55, 87 S. Ct. 1873, 1881-82, 18 L. Ed. 2d 1040 (1967). In conducting this review, our duty is to ensure that the "magistrate had a substantial basis for concluding that a search would uncover evidence of wrongdoing." Griffin, 827 F.2d at 1111 (quoting Gates, 462 U.S. at 236, 103 S. Ct. at 2331).

Considering the magistrate's decision in this case, we note that the affidavit Officer Kettlekamp presented to the magistrate contained personal observations of a citizen who promptly reported to the police details of specific evidence of criminality that were located inside the residence at 2238 East Laurel. The informant also asserted that she was the victim of a violent crime committed by the defendant. The magistrate was thus presented with specific details of a crime; neither Brown's drug addiction nor criminal record warranted disregarding her report. Although an informant's reliability is a factor to be considered by a court, it is just one relevant consideration in the totality of the circumstances analysis. Gates, 462 U.S. at 234, 103 S.Ct. at 2330. As the Supreme Court stated in Gates, "even if we entertain some doubt as to an informant's motives, [her] explicit and detailed description of alleged wrongdoing, along with a statement that the event was observed first hand, entitles [her] tip to greater weight than might otherwise be the case." Id. at 234, 103 S. Ct. at 2330. We have such a detailed first hand observation in this case. These statements provided the magistrate with ample information from which to conclude that a search would uncover evidence of crime. We conclude, therefore, that there was probable cause to issue the warrant. Even if we did not agree with the magistrate's determination, however, we could not invalidate the warrant. Adhering to the Supreme Court's statement that "after-the-fact scrutiny by courts of the sufficiency of an affidavit should not take the form of de novo review," Gates, at 236, 103 S. Ct. at 2331, we could not reasonably conclude that the magistrate lacked a substantial basis for determining that, under the totality of the circumstances, probable cause existed. The affidavit was more than sufficient to provide the magistrate with a substantial basis to issue a warrant based on probable cause.

McKinney's next contention on appeal is that the evidence presented was not sufficient to prove him guilty beyond a reasonable doubt. "In evaluating the defendant's sufficiency of the evidence challenge, we note that he bears a heavy burden.... The test is whether, after viewing the evidence in the light most favorable to the government, any rational trier of fact could have found the essential elements of crime beyond a reasonable doubt." United States v. Nesbitt, 852 F.2d 1502, 1509 (7th Cir.1988) (emphasis in original) (citations omitted). In fact, only when the record contains no evidence from which the trier of fact could find guilt beyond a reasonable doubt, may an appellate court overturn the verdict. Id.

We conclude that the district court was presented with evidence from which it could conclude the defendant was guilty beyond a reasonable doubt. There was testimony presented at trial that McKinney was the actual owner and resident of the residence at 2238 East Laurel where the weapons and cocaine were found. His fingerprint was found on one of the weapons seized from the bedroom and various witnesses testified that he had a firearm present during a drug transaction. The evidence also demonstrated that McKinney possessed the cocaine with the intent to distribute it. We conclude that viewing the evidence in the light most favorable to the government, a rational trier of fact could have found that the defendant was guilty beyond a reasonable doubt on all counts.

McKinney's final challenge on appeal concerns the appropriateness of the sentences he received for unlawful possession of a firearm by a convicted felon and use of a firearm during and in relation to a drug trafficking crime. He received 5 years on each of the six firearm possession counts to run concurrently with each other and with the drug charge. He also received 5 years on each of the six use of a weapon counts to run consecutively to each other and consecutively to the underlying cocaine and firearm possession sentences. McKinney argues that the court should have imposed sentence on either the charge of possession of a weapon or use of a weapon, but not both. He alleges that sentencing him under both violates the double jeopardy clause of the fifth amendment, or in the alternative that the multiple sentences received in each count violates that clause.

In Blockburger v. United States, 284 U.S. 299, 52 S. Ct. 180, 76 L. Ed. 306 (1932), the Supreme Court devised a test for determining whether two statutory violations constitute two offenses for purposes of the double jeopardy clause. The Court held that "the test to be applied to determine whether there are two offenses or only one, is whether each provision requires proof of a fact which the other does not." Blockburger, 284 U.S. at 304, 52 S.Ct. at 182. A reading of the two applicable statutes makes it clear that these two offenses consist of separate elements, and therefore, the defendant could receive separate sentences for each offense. Section 922(g) provides:

It shall be unlawful for any person(1) who has been convicted in any court of a crime punishable by imprisonment for a term exceeding one year to ... possess in or affecting commerce, any firearm or ammunition; or to receive any firearm or ammunition which has been shipped or transported in interstate or foreign commerce.

Whoever during and in violation to any crime of violence or drug trafficking crime ... uses or carries a firearm, shall....

As the Ninth Circuit determined in United States v. Hunter, 887 F.2d 1001, 1003 (9th Cir.1989), "[i]t is obvious that conviction of the offense under Section 924(c)(1) requires proof of elements not required for conviction under 922(g)(1), and vice versa." Section 924(c)(1) is a penalty for a felon who possesses a firearm, while section 924(g)(1) addresses Congress' concern with the more heinous crime of drug trafficking with a firearm. Therefore, the two charges satisfy the Blockburger test.13

McKinney's contention concerning the multiple sentences he received for each of the firearm offenses holds more weight. He argues that he should not receive multiple punishment for what is in effect one offense. We agree. In the context of multiple punishments imposed in a single prosecution, "the Double Jeopardy Clause does no more than prevent the sentencing court from prescribing greater punishment than the legislature intended." Grady v. Corbin, --- U.S. ----, 110 S. Ct. 2084, 2091, 109 L. Ed. 2d 548 (1990) (quoting Missouri v. Hunter, 459 U.S. 359, 366, 103 S. Ct. 673, 678, 74 L. Ed. 2d 535 (1983)). McKinney was convicted of one narcotics transaction in which he used, and as a convicted felon, possessed a firearm. Yet he was sentenced on six counts of each crime for the six weapons seized. There is nothing in the statutes from which we can conclude that Congress considered the situation we are faced with today and intended such extreme sentencing by the piling up of charges. "[I]n the absence of express Congressional intent to 'fix the punishment for a federal offense clearly and without ambiguity, doubt will be resolved against turning a single transaction into multiple offenses...." McFarland v. Pickett, 469 F.2d 1277, 1278 (7th Cir.1972) (quoting Bell v. United States, 349 U.S. 81, 84, 75 S. Ct. 620, 622-23, 99 L. Ed. 905 (1955)).

In United States v. Baugh, 787 F.2d 1131, 1132 (7th Cir.1986), we applied what has been termed the McFarland/Calhoun reasoning and determined that Baugh's sentences on five counts of unlawful receipt of a firearm in violation of 18 U.S.C. Sec. 922(h) should be reversed because absent proof of the weapons being acquired or stored separately he was guilty of only one offense. See United States v. Calhoun, 510 F.2d 861, 869 (7th Cir.), cert. denied, 421 U.S. 950, 95 S. Ct. 1683, 44 L. Ed. 2d 104 (1975) ("absent a showing that two firearms were stored or acquired at different times or places, there is only one offense"); McFarland, 469 F.2d at 1277 (multiple sentences improper when no evidence of separate acquisition or storage). This line of reasoning was also adopted by this Court in United States v. Oliver, 683 F.2d 224 (7th Cir.1982), where we held that ammunition and a revolver cannot support separate offenses of receipt of a firearm or ammunition which has been shipped or transported in interstate or foreign commerce by any person convicted of a crime punishable by a prison term exceeding one year absent evidence that the ammunition and firearm were acquired at different times. In all these decisions, we ascertained that Congress intended to punish undifferentiated possession or receipt of multiple firearms no more severely that the possession or receipt of a single firearm. See also United States v. Hodges, 628 F.2d 350, 352 (5th Cir.1980) (undifferentiated possession of multiple firearms constitutes only one offense).

These cases are directly on point and we conclude that McKinney cannot receive separate sentences for the firearms charges. Sections 924(c)(1) and 922(g)(1), under these facts, are only punishable as single offenses. The government did not demonstrate that the guns were stored or obtained in different locations or used in carrying out different drug trafficking crimes. All the guns were recovered from one location, the bedroom of the residence at 2238 East Laurel and only one had the defendant's fingerprint. Furthermore, the defendant was convicted for use of a firearm in one narcotics transaction, not six. A firearm was only connected by witnesses to this one transaction and we are not presented with proof that a gun was used on six different occasions. Therefore, under Baugh, there can be only one possession and one use offense.14 Accordingly, we conclude that appellant's concurrent and consecutive sentences on the firearms convictions violate the double jeopardy clause's proscription against multiple punishment. Therefore, we vacate the sentences and remand the case to the district court for resentencing of the firearm crimes as one offense of possession of a firearm and one offense of use of a firearm.

For the above stated reasons, we AFFIRM McKinney's convictions on all counts. We REMAND for resentencing on the firearms charges consistent with this opinion.

POSNER, Circuit Judge, concurring.

I have no disagreement with the decision of this case, but I want to flag, for future reference, a disagreement within the court over the standard for reviewing determinations of probable cause. A number of our decisions, as well as decisions in other circuits such as United States v. Muniz-Melchor, 894 F.2d 1430, 1439 n. 9 (5th Cir.1990), and United States v. Suarez, 902 F.2d 1466, 1467 (9th Cir.1990), hold or imply that such determinations, by whomever made, are to be reviewed de novo, that is, as if the reviewing judge or judges were the front-line judicial authority--the magistrate, or the trial judge, or the jury, that made the initial determination. Illustrative are United States v. Malin, 908 F.2d 163, 165-66 (7th Cir.1990); United States v. Sophie, 900 F.2d 1064, 1072 (7th Cir.1990); United States v. Ingrao, 897 F.2d 860, 862 (7th Cir.1990); United States v. Rambis, 686 F.2d 620, 622 (7th Cir.1982). Judge Easterbrook and I have expressed our disagreement with that approach. United States v. Malin, supra, 908 F.2d at 169-70 (concurring opinion). We think that the proper standard is the clearly-erroneous rule used in other areas of law to review the application of a legal standard to a particular set of facts.

Some courts already use the clear-error formula in reviewing determinations of probable cause, at least when made by the district judge, no warrant having been issued. United States v. Fox, 902 F.2d 1508, 1513 (10th Cir.1990); United States v. Williams, 897 F.2d 1430, 1435 (8th Cir.1990); United States v. Santana, 895 F.2d 850, 852 (1st Cir.1990). In like vein (and in tension with the cases from this circuit cited above), this court has said that "a district court's denial of a motion to suppress evidence will be affirmed on appeal unless it is clearly erroneous," United States v. D'Antoni, 856 F.2d 975, 978 (7th Cir.1988), and then, citing D'Antoni, that "in denying the defendants' motion to suppress, the district court found that there had been sufficient probable cause to issue a warrant to search the defendants' residences; we must determine whether that ruling was clearly erroneous." United States v. McNeese, 901 F.2d 585, 592 (7th Cir.1990). Many cases, including the present one, give considerable deference to the determination of probable cause when it is made by the magistrate in issuing a warrant, but decline to equate that deference to review for clear error. To complete the muddle, we have held--en banc, no less--that when the issue of probable causes arises in a suit tried to a jury, the issue is one for the jury, not for the judge, and the jury's determination must therefore be upheld if reasonable (which is to say, not clearly erroneous). Llaguno v. Mingey, 763 F.2d 1560, 1565, 1567 (7th Cir.1985) (en banc). (That was a plurality opinion but the court was unanimous in its characterization of the probable-cause determination. Id. at 1570 (Coffey, J., concurring and dissenting); id. at 1578 (Wood, J., concurring and dissenting); id. at 1580 (Cudahy, J., dissenting).) See also Gramenos v. Jewel Cos., 797 F.2d 432, 438 (7th Cir.1986); Moore v. Marketplace Restaurant, Inc., 754 F.2d 1336, 1347 (7th Cir.1985). I suggest that clearly erroneous is the correct rule whoever makes the initial determination.

Consider a finding of negligence in an ordinary tort suit. It is not a finding of fact in the sense that it could be made by someone uninstructed in the legal standard of negligence. Rather it is the application of the legal standard to the facts of the particular case. But it is not reviewed de novo; it is reviewed for clear error. Cooter & Gell v. Hartmarx Corp., --- U.S. ----, 110 S. Ct. 2447, 2459, 110 L. Ed. 2d 359 (1990); Mucha v. King, 792 F.2d 602, 605 (7th Cir.1986); Nunez v. Superior Oil Co., 572 F.2d 1119, 1126 (5th Cir.1978). The broader principle, endorsed by a growing number of cases in this circuit, such as Mars Steel Corp. v. Continental Bank N.A., 880 F.2d 928, 933 (7th Cir.1989) (en banc), and Foy v. First National Bank, 868 F.2d 251, 254 (7th Cir.1989), is that (with immaterial exceptions) the application of law to fact is itself a question of fact for purposes of appellate review. The point is not that it is "really" a question of fact; that would be absurd. The point is that appellate review of the application of law to fact should be deferential, as it is in the case of rulings on questions of fact, and this for two reasons. The trier of fact, whether judge or jury, is closer to the facts than the appellate judges and is therefore better able, other things being equal, to assess their legal significance. And the dependence of the determination of negligence on the facts of the particular case, facts that will not recur exactly in any other case, would make quixotic an attempt to bring about uniformity of results by close appellate review.

These considerations argue as forcefully for deferential review of determinations of probable cause as for deferential review of determinations of negligence, if only because a determination of probable cause involves applying to the particular facts of the case a legal standard that is quite like the negligence standard. This was one of the reasons given in Llaguno v. Mingey, supra, 763 F.2d at 1565, for treating probable cause as a jury question. Both negligence and probable cause require a judgment of reasonableness that is made after balancing costs and benefits (the costs of care and the benefits of accident avoidance in the negligence case, and the costs of getting evidence of criminal guilt by other means and the benefits in protecting privacy in the probable-cause case) and is based explicitly on probabilities (the probability of an accident in the negligence case and the probability that a search will turn up evidence of crime in the probable-cause case). Compare Brinegar v. United States, 338 U.S. 160, 175, 69 S. Ct. 1302, 1310-11, 93 L. Ed. 1879 (1949), and United States v. Sweeney, 688 F.2d 1131, 1137 (7th Cir.1982), with McCarty v. Pheasant Run, Inc., 826 F.2d 1554, 1556-57 (7th Cir.1987), and United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir.1947) (L. Hand, J).

The application of the standard of probable cause to the particular facts of a case may be made by a magistrate in issuing a search warrant, as in this case; by the district judge in a case in which the defendant moves to suppress evidence obtained by a search not pursuant to warrant, so that there will never have been a determination of probable cause by a magistrate; or by the judge or jury in a civil rights suit in which the plaintiff claims that he was the victim of an unconstitutional search or seizure (Llaguno was such a case). In the first type of case the district judge will review the magistrate's determination of probable cause if and when the defendant moves to suppress evidence seized in or as a result of the allegedly improper search or seizure, while in the other types of case the first appellate review will be by the court of appeals. But whatever the type of case, the only question for the reviewing court, whether it is the district court or the court of appeals, should be whether the determination of probable cause made by the first-line judicial officer (magistrate, district judge, or jury, as the case may be) is clearly erroneous.

Because the dramatic increases in federal judicial workloads in recent years have fallen so heavily on the federal courts of appeals, Report of the Federal Courts Study Committee 5, 110 (April 2, 1990), it is no surprise that there has been a steady trend toward limiting the scope of appellate review of determinations of fact-specific issues. The landmarks are noted in the concurring opinion in Malin. This circuit's position on review of determinations of probable cause bucks the trend for no good reason. Although it is true that determinations of probable cause made in connection with the issuance of a warrant are ex parte, our circuit's rule of de novo review applies even when the determination is made not by a magistrate or judge ex parte, but in a full adversarial hearing because no warrant had ever been issued. Indeed, bowing to the Supreme Court's command to give "great deference" to the magistrate's determination of probable cause in a warrant proceeding, Illinois v. Gates, 462 U.S. 213, 236, 103 S. Ct. 2317, 2331, 76 L. Ed. 2d 527 (1983), even circuits such as the Fifth and Ninth, which (in cases like Muniz-Melchor and Suarez, cited earlier) like us review de novo a district judge's determination of probable cause when made in a case where no warrant was issued, defer to the same determination when made by a magistrate in an ex parte warrant hearing. In re Trinity Industries, 898 F.2d 1049, 1050 (5th Cir.1990); United States v. McQuisten, 795 F.2d 858, 861 (9th Cir.1986). McQuisten actually states--correctly as it seems to me--that "we may not reverse a magistrate's finding of probable cause unless it is clearly erroneous." Id.

What sense can it make to defer to the magistrate's determination of probable cause to the same or similar degree as other findings of fact, but give no deference at all to the same determination when made by a district judge after a full adversarial hearing in a case in which there was no warrant proceeding? The Fifth and Ninth Circuits have avoided confronting this anomaly by the unedifying expedient of maintaining separate lines of precedent. Cases that defer to the magistrate's finding of probable cause do not cite cases that review the district judge's finding of probable cause de novo, nor vice versa. And neither line of precedent cites cases in which a jury's determination of probable cause is given the full respect ordinarily given findings of fact whether made by judge or by jury. By such ostrich methods an incoherent approach to the review of probable-cause determinations is perpetuated.

Provoked by the effrontery of a concurring opinion that proposes a change in the formulation of the standard of review (only in law is "innovative" a pejorative), Judge Flaum's opinion forthrightly acknowledges the existence of the conflicting lines of precedent. For this forthrightness I applaud him. At last the issue is out in the open. Now all can see that the circuit's position is a Rube Goldberg invention: a needlessly complex machine, which incidentally does not work.

Judge Flaum's opinion states that a magistrate's finding of probable cause must be upheld if it has a "substantial basis." This is a step in the right direction, because it is a step away from de novo review. It is a step taken unconsciously, in a sense, because the opinion appears to labor under the impression that it is only in cases in which there is no warrant that this court reviews determinations of probable cause de novo. I do not agree. Malin and Rambis, cited at the outset of this opinion, were warrant cases. Malin repeated the standard set forth in Rambis: "Since this determination [whether the information in the affidavit submitted to the magistrate in the warrant proceeding establishes probable cause] involves the application of law rather than an evaluation of factual evidence, on review the appellate court is not limited to a determination of whether the district court's finding was clearly erroneous. It must independently review the sufficiency of the affidavit, recognizing that doubtful cases should be resolved in favor of upholding the warrant." 686 F.2d at 622 (citation omitted), quoted in pertinent part in United States v. Malin, supra, 908 F.2d at 165. If all that is meant by this rather murky prose is that when there is successive review--here, review of the magistrate's determination by the district judge followed by review of the district judge's determination by the court of appeals--the second review is de novo, I would have no quarrel. The district judge must decide whether the magistrate committed clear error, and we must decide the same thing. In re Trinity Industries, Inc., supra, 898 F.2d at 1050; United States v. Castillo, 866 F.2d 1071, 1076 (9th Cir.1988); Parker v. Sullivan, 891 F.2d 185, 187 (7th Cir.1989). The appropriateness of such duplicative review may be questioned, but that is a separate matter. A finding that there was clear error is itself a legal determination, and legal determinations, as opposed to the application of law to fact, are reviewed de novo. But if "independently review the sufficiency of the affidavit" is thought to have some independent force, then despite the concession which follows regarding the doubtful case, we may indeed have a standard of, or at least very close to, de novo review. Read the portions of the Malin and especially Rambis opinions in which the court applies the standard, and you will be driven to conclude that the de facto standard is indeed de novo review. "We hold that the facts and circumstances described in the affidavit were sufficient to establish probable cause and that the issuance of the search warrant was proper." United States v. Rambis, supra, 686 F.2d at 625.

Standing by itself Judge Flaum's formulation of the standard of review in warrant cases ("substantial basis") would differ only semantically from that in McQuisten. But he goes on to explain that he considers his standard an "intermediate" standard, intermediate, that is, between de novo review and review for clear error. I understand where the "substantial basis" formula comes from; it comes from Illinois v. Gates. I do not understand where the idea that it is an intermediate standard comes from. Gates does not describe it so, or purport to reject the clearly-erroneous rule or to adopt an intermediate standard; these issues are not discussed. I particularly fail to see how the Court's directive in Gates that we give "great deference" to the magistrate's finding can be squared with a rule that gives that finding less deference than we would give an ordinary finding of fact. See also United States v. Leon, 468 U.S. 897, 914, 104 S. Ct. 3405, 3416, 82 L. Ed. 2d 677 (1984). I find nothing in Gates inconsistent with declaring the standard of review of the magistrate's determination of probable cause to be that of clear error and I do not see what sense it makes to give more deference to the magistrate's determination than to a district judge's made after a full hearing, which under our circuit's cases would be reviewed de novo.

In suggesting that "substantial basis" be equated to "not clearly erroneous" I am not trying to defy the Supreme Court. If the Supreme Court thinks these are different standards I shall consider myself bound by its view. But I find nothing in the Court's opinions to suggest that it does. Perhaps my brethren do not either; for they are not content to rest upon authority; they offer in two opinions three reasons for wanting an intermediate standard of review for magistrate's determinations. The first is that there should be more searching appellate review generally when constitutional rights are at stake. If this is right, it still leaves unexplained why there should be different standards depending on who the trier of fact is. But it cannot be right, not as a general rule at any rate. (The Supreme Court has described as "rare" those occasions on which it has authorized de novo review of findings of fact in constitutional cases. Miller v. Fenton, 474 U.S. 104, 114, 106 S. Ct. 445, 451-52, 88 L. Ed. 2d 405 (1985).) For this would mean that the clearly-erroneous rule did not apply in any constitutional case, something I would not have thought anyone believed. The second reason given (this one by Judge Will only) is that since the magistrate's determination is nonadversarial there is less reason to believe it correct. If this is right, it leaves unexplained why under our circuit's precedents an adversarial determination by a district judge after a full hearing receives more searching review than the magistrate's ex parte determination. Hence the third reason: to encourage police to seek warrants we give the magistrate's determination of probable cause to support a warrant more respect than a district judge's determination of probable cause when there was no warrant. But actually this is an argument only for giving both determinations the same respect. For we have offsetting considerations: the nonadversarial character of the magistrate's determination and his lack of the guarantees of tenure and undiminished compensation that Article III gives federal judges, which pushes my brethren toward wanting to give magistrate's determinations less respect than judges' determinations; and the desire to encourage the police to makes searches and arrests on the basis of warrants, which pushes them in the opposite direction.

My brethren in defending the intermediate standard may in any event be spitting into the wind. Since the Supreme Court's adoption in Leon of the good-faith exception to the exclusionary rule, 468 U.S. at 924, 104 S. Ct. at 3421, judicial review of the magistrate's determination of probable cause in issuing a search or arrest warrant has had much less practical importance than it did before. The evidence obtained through the warrant will not be suppressed even if there was no probable cause, provided only that the officers who sought and executed the warrant were acting in good faith, albeit erroneously. In plumping for an intermediate standard my brethren are bailing water from a ship that the captain has decided to scuttle.

There is a more fundamental objection to their warm endorsement of multiple standards of review. It greatly exaggerates the utility of verbal differentiation. It reflects the lawyer's exaggerated faith in the Word. I think I understand the difference between plenary review and deferential review. In the former setting the appellate judge must say to himself, "The issue has been given to me to decide, and while I shall pay due attention to what the district judge (or other trier of fact) had to say on the question the ultimate decisional responsibility is mine and must be exercised independently." In the latter setting the appellate judge must say to himself, "The issue is not mine to decide; because the district judge (or magistrate or administrative agency or whatever) has a better feel for it, or for other institutional reasons (such as to discourage appeals), the responsibility for deciding has been given to him and I must go along unless persuaded that he acted unreasonably, or in other words unless I am clear in my mind that he erred." What is the intermediate position? There is none; the proliferation endorsed by my colleagues on this panel has only semantic significance. The reasons that lead the court to support an intermediate standard can, as we shall see in a moment, be accommodated without a multiplication of formulas. Nor, finally, do I see how the words "substantial basis" can be thought, even on the semantic level, to imply a less deferential standard of review than clear error. "Substantial basis" sounds like "substantial evidence," the formula for judicial review of administrative factfindings--a formula usually thought, if anything, more deferential than the standard of clear error is. My brethren have become hopelessly entangled in words--they have forgotten Justice Holmes's admonition to think things not words--but even at the level of semantics they strike out.

They fear that the clearly-erroneous standard has become a synonym for perfunctory review. I do not share that fear. It is true that when a finding by the first-line decision-maker resolves an issue of which witnesses to believe, the scope of appellate review is exceedingly narrow because the reviewing court has not seen the witnesses testify. But when the finding to be reviewed is the evaluation of the legal significance of conceded facts, the reviewing court is not similarly handicapped and need not and should not consider its function that of a rubber stamp. This is an example of the broader point, which is both fundamental and ignored, that the nature of the issue to be reviewed in relation to the comparative institutional advantages of trial and reviewing courts, and other pertinent practical considerations, rather than the words in which standards of appellate review are formulated, should--and usually do--determine the scope of review. It is recognized that "abuse of discretion" is a chameleon, or more politely a range, not a point. Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 388-89 (7th Cir.1984), and cases cited there; Friendly, Indiscretion About Discretion, 31 Emory L.J. 747 (1982). So is "clear error" (and they may be the same chameleon, the same range). Of course the findings of fact in a capital case will receive more searching appellate review than the findings in a case in which a small fine is imposed; of course appellate courts will be, or at least should be, zealous to protect the right to be free from unreasonable searches and seizures. The quality of the factual inquiry--whether adversarial or ex parte--and of the factfinder should also, and does, affect the willingness of appellate courts to reverse findings governed by the clearly-erroneous rule. Cf. River Road Alliance, Inc. v. Corps of Engineers, 764 F.2d 445, 449 (7th Cir.1985). The impression that "substantial evidence," the formula used in reviewing the decisions of administrative agencies, is more deferential than "clearly erroneous" derives not from the semantic difference in the formulas, which is opaque to meaning, but from the fact that appellate judges feel more comfortable with the issues on which district judges make findings (many federal appellate judges are former district judges) than they do with the issues that administrative agencies resolve.

What is needed at this juncture in the evolving law of appellate review is not a multiplicity of rigid rules stated in empty jargon, or even three rigid rules that hack crudely at a complex reality, but the sensitive application of the clear-error standard, understood as such, across the board. This is a cleaner as well as a more honest approach than attempting the legerdemain of deriving from the words "great deference" a warrant for a nondeferential rule of appellate review to be squeezed between de novo review and clear-error review.

American law is too vague, too complicated, too expensive; and it is these things in part because judges are too fond of sterile verbalisms and outmoded distinctions. A tripartite standard of appellate review of determinations of probable cause is confusing, unworkable, and unnecessary. We should not fear to reject it for fear of being called innovative.

WILL, Senior District Judge, concurring.

I join in the judgment and agree with most of the reasoning of Judge Flaum's opinion. I write separately, and with reluctance, in light of Judge Posner's concurrence, to elaborate further on the reasons why, absent a decision from the Supreme Court changing the established rule, I believe that fourth amendment findings of probable cause to conduct a search both are and should continue to be reviewed under a "substantial basis" standard rather than some less demanding standard such as "clear error."

1. Obviously, negligence and probable cause are similar in that both are mixed questions of law and fact. See Posner concurrence, supra at 419-20. I also agree that "appellate review of the application of law to fact should be deferential." Id. But it does not follow from there, as Judge Posner would have it, that the standard of appellate review, even though generally deferential for mixed questions, is (or even ought to be) "clear error" for all mixed questions.

Probable cause is a constitutionally mandated standard. Negligence isn't. A determination of probable cause for the issuance of a search warrant is made in an ex parte proceeding. A finding of negligence is not. These are key differences. Our review of the ex parte application of important constitutional standards to facts both is and ought to be more rigorous and searching than the review, after a trial, of a finding of negligence, by jury or judge, in which no constitutional standards are implicated. Even appellate review of facts is sometimes more rigorous and searching when a constitutional norm is at issue. See Rosenbloom v. Metromedia, 403 U.S. 29, 54, 91 S. Ct. 1811, 1825, 29 L. Ed. 2d 296 (1971) (first amendment questions of "constitutional fact" compel de novo review); Bose Corp. v. Consumer Union, 466 U.S. 485, 104 S. Ct. 1949, 80 L. Ed. 2d 502 (1984) (same).

It is simply inaccurate to say that reviewing findings of probable cause under a more searching standard than clear error somehow "implies that probable cause is a question of 'law.' " Malin, supra, 908 F.2d at 169 (Easterbrook, joined by Posner, concurring). All that is "implied"--or, more accurately, what the use of a substantial basis standard reflects--is that the application of important constitutional rules demands greater scrutiny and review than the application of some other rules. See also United States v. Hawkins, 823 F.2d 1020, 1022 (7th Cir.1987) (de novo review for another mixed question of constitutional law and fact--the voluntariness of confessions).

Cooter & Gell v. Hartmarx Corp., --- U.S. ----, 110 S. Ct. 2447, 110 L. Ed. 2d 359 (1990) and related decisions are not pertinent here. There is a world of difference between enforcing Rule 11 and vindicating rights secured by the fourth amendment. Rule 11 determinations of "frivolousness," as well as findings of negligence or determinations about who is and who isn't a "buyer in the ordinary course" (Foy v. First National Bank, Posner supra ), and even rulings that a defendant arrested for a minor offense was not arrested as a pretext for collecting statements and evidence relating to his involvement in other crimes (United States v. D'Antoni, Posner supra ), are all properly reviewed for clear error only. But when it comes to basic constitutional protections, such as the guarantee that "no Warrant shall issue, but upon probable cause," our duty is higher and so is the standard of review--as it ought to be.

There is no "trend" toward revising the standards by which we review the application of constitutional standards to facts or to mixed questions of facts and law. Nor should there be any such trend. United States v. D'Antoni is not contrary. There was no review in D'Antoni of the application of any constitutional rule. The issue in the case was ordinary fact--was D'Antoni arrested for the traffic offense or only so that he could be grilled about other crimes? The broad language in the opinion--"motions to suppress evidence will be affirmed on appeal unless ... clearly erroneous"--comes out of nowhere, and the citation that follows it, to United States v. Binder, 794 F.2d 1195 (1986), is no support for it. Binder stands only for the simple proposition that "[t]he credibility of witnesses at a suppression hearing is a matter for the trial judge to determine, and his credibility findings will not be reversed unless they are clearly erroneous." Binder, 794 F.2d at 1199. Neither D'Antoni nor Binder holds, or offers any rationale, argument, or Supreme Court precedent to support the broad proposition that a constitutional ruling made during the course of a suppression hearing should be reviewed for clear error only. Take, for example, a due process ruling that a witness identification was not prompted by undue suggestion, was reliable under the totality of the circumstances, and was therefore admissible. Our review of that ruling would not be limited to a run-through check for clear e See Neil v. Biggers, 409 U.S. 188, 193 n. 3, 93 S. Ct. 375, 379 n. 3, 34 L. Ed. 2d 401 (1972) ("It is said that we should not '... reverse findings of fact ... unless shown to be clearly erroneous.' This rule of practice ... is a salutary one to be followed where applicable. We think it inapplicable here where the dispute between the parties is not so much over the elemental facts as over the constitutional significance to be attached to them.")

Constitutional review is of a different order than most other appellate review, because constitutional rights are of a different magnitude than most other rights, and there is nothing either extraordinary or inexplicable about the proposition that district judges' and magistrates' constitutional rulings, applying constitutional standards to facts, should be reviewed under a more stringent standard than clear error. This is not to say that the "clearly-erroneous rule [does] not apply [to any ruling] in any constitutional case." Posner, supra at 422. It is simply to say that in constitutional cases appellate review of mixed questions should proceed under a more searching standard than simple clear error, a proposition which Judge Posner offers no reasons for doubting.

In any event, unless and until the Supreme Court mandates a change in the standards of review, I see no reason for lower courts to revise them and thereby diminish constitutional protections. I firmly believe the present standards to be correct and in my opinion, the old maxim "If it ain't broke, don't fix it" applies here.

2. The difference between the substantial basis standard and a clearly erroneous standard of review is significant. At first glance it may look like semantics--define a clearly erroneous decision as a ruling without substantial evidentiary support, and the difference between a ruling without a substantial basis and a clearly erroneous one seems nonexistent. But there is at least a psychological (and accordingly, in practice, a quite real) difference between saying (a) that we will reverse only if the decision below is clearly erroneous and (b) that we will affirm, if, giving deference to the lower court's determination, that determination has a substantial basis. The first formulation amounts to a recipe for almost routine affirmance. The second, by contrast, ensures a more detailed and searching review than a clear-error standard does and is the appropriate and correct standard for reviewing constitutional determinations. There are cases that will be decided differently, and the fourth amendment protection will be different, depending on whether a "substantial basis" or a "clearly erroneous" standard is used.

3. Judge Posner urges, in part, that review of determinations of probable cause should be limited to review for clear error because an appellate panel will be farther removed from the facts and less able to assess their significance than the district judge or magistrate. Posner, supra, p. 420. This is only rarely true. In cases where a hearing was held before the warrant issued, virtually all the facts known to the district judge or magistrate will be in the officer's sworn affidavit. A warrant hearing is not an adversarial hearing, and no substantial record is developed. The officer is not normally cross examined and the informant referred to in the affidavit, if there was an informant, does not usually appear in court. There is no weighing of the testimony of disagreeing witnesses.

In my experience, district judges and magistrates do not generally refuse to issue warrants based on credibility determinations or go behind the statements in the affidavit. If the affidavit appears to satisfy fourth amendment criteria, they issue the writ. Almost always, the decision to issue or refuse a warrant is made on the basis of the facial sufficiency of the accompanying affidavit--an affidavit which an appellate panel can review just as easily as the district judge or magistrate did in the first instance.

In cases where a warrant issued, and the only hearing was ex parte, an appellate panel will generally be in the same position to evaluate the facts as the district judge or magistrate was. Distance from the facts is no justification for a clear error standard in such a case. In this connection, it is important to recognize that substantial basis review does not mean no deference and that the deference inherent in the substantial basis standard allows for sufficient consideration of any advantage in observing the evidence which the lower court may have had.

In cases where a hearing was held only after the search, no warrant having issued, the circumstances are somewhat different. A hearing after a search, as opposed to a hearing on a request for a warrant, will not generally be ex parte. Frequently it will be fully adversarial and often it will result in a substantial record with live testimony from several witnesses. I see no compelling reason why such fully adversarial hearings, filled with live testimony, should be reviewed de novo, though that has apparently been the rule in this circuit. See Flaum, supra p. 413. I agree with Judge Posner that in such "nonwarrant" cases the district judge will generally be closer to the facts than an appellate panel and that de novo review is, accordingly, inappropriate. I do not agree with Judge Posner, however, that the solution is to adopt a clearly erroneous standard for all probable cause determinations. I would review "nonwarrant" cases under a substantial basis standard, the same as cases in which warrants issued.

4. As a general principle, and all other things being equal, legal rules should, if sound, also be simple and uniform. Rules that are vague or needlessly complex are inefficient--because of the uncertainty over how far they reach and what they mean, they commonly deter more behavior than they were meant to. For that reason, rules governing conduct should, generally, be simple and uniform. But the point has much less force in the context of standards of review, which are not rules that have any significant effect on conduct. There is no per se reason that standards of review should be either simple or uniform. A proliferation of detailed standards of review, with different standards in different areas, is not a problem in itself--so long as each standard is clear and administrable and the categories of rulings to which it applies is plainly defined. Occam's razor to the contrary notwithstanding, it is OK, desirable, and the law, see Illinois v. Gates, supra, that negligence and probable cause are reviewed under different standards and that the standard for probable cause is whether the factfinder below had a "substantial basis" for concluding that probable cause existed. In the last analysis, it is more important that rules be just than that they be simple and uniform.

Whether uniform use of a clear-error standard would be a "cleaner" approach to appellate review, Posner supra at 423, is beside the point. Deciding whether a finding of probable cause has a substantial basis is often more difficult, and less tidy, than deciding only whether the finding was clearly erroneous or not. Judges, however, are called on to make hard decisions. Justice is not separated from injustice by a bright line. The difficulty of discerning the difference between the existence of probable cause and the lack of it is no reason for shirking a constitutionally mandated responsibility to distinguish them. Clear error may be a less complicated and less time-consuming standard to apply than substantial evidence. But it is a standard that will diminish constitutional protections and lead to less just results in some cases, and caving in to injustice, because guaranteeing justice is "complicated" or "expensive," is not the role of a judge.

Judge Posner also suggests that using a clear-error standard across the board would be a "more honest" approach to appellate review. Posner, supra p. 423. But there is nothing dishonest about insisting on a more rigorous level of review for probable cause determinations than clear error review entails and nothing dishonest about using different levels of review in different cases. And as Judge Flaum points out, the traditional position, that different phrases mean different things (i.e., that "substantial basis" does not mean "clear error") is at least as "honest" and no more dependent on a faith in words than Judge Posner's suggestion that one phrase, "clear error," means different things in different contexts.

5. The government has not argued that the standard for reviewing probable cause determinations should be lowered from "substantial basis" to "clear error." That circumstance would generally spell waiver in this circuit. And this court has, on at least one past occasion, stated that the government not having challenged the present standard of review, the issue is "all by the way" and "we must therefore" refrain, in such a case, from switching to a "clear error" standard. See United States v. Rutledge, 900 F.2d 1127, 1129 (7th Cir.1990) (Posner, J.).

This is not a hard case for which the standard of review makes a difference. The information in Officer Kettlekamp's affidavit easily supported the issuance of a warrant to search the residence at 2238 Laurel. The magistrate's finding of probable cause would obviously have sailed through clearly erroneous review but our affirmance under the more searching substantial basis standard is, on these facts, equally unmistakable.

It should be noted that challenging a search conducted pursuant to a warrant, on grounds that there was no probable cause, is an uphill exercise. That's not because probable cause for conducting a search and negligence are similar rules governed by like considerations and subject to the same, clearly erroneous standard of review. The differences between constitutional rules and others, and between substantial basis review and clear error review, are significant and vital and should not be ignored in the interests of simplicity and uniformity. But the fact that probable cause is a constitutionally mandated standard, and that findings of probable cause must have a substantial basis to be affirmed, will rarely shape the outcome in a case like this one where there was a warrant. For in cases where a warrant has issued, the importance of probable cause has been substantially eclipsed by the "good faith" exception to the fourth amendment. United States v. Leon, 468 U.S. 897, 104 L. Ed. 2d 3405, 82 L. Ed. 2d 677 (1984). Unless the officer executing the warrant knows or reasonably should have known that it was procured improperly, his "good faith" reliance on it will be sustained. In this case, even if there had been no probable cause (no matter what the standard of review) to issue a warrant for the search which turned up McKinney, the guns, and the cocaine, the result of our review would have been the same. There is no evidence to suggest anything but that the warrant which issued in this case was signed by a detached and neutral magistrate, facially valid, and acted upon by the police in good faith.

A final observation. It is unfortunate, since in this case all three members of the court reach the same result, that so much time, energy and money have been devoted to discussing whether or not a new standard of review not yet mandated by the Supreme Court should be applied in this type of case. I would have preferred to leave that question to a case in which it made a difference in the result.