United States v. Aldo Mazzone, Robert Quagliato, Steven Kandis, and Tony Slutsky

POSNER, Circuit Judge. The four appellants were convicted of drug crimes, received sentences ranging from probation and a fine, for Mazzone, to ten years in prison and a fine, for Kandis, and appeal on two grounds — illegally seized evidence and improper argument to the jury by the prosecutor. The first relates to the seizure, without a warrant, of drugs from two vehicles, a van driven by Mazzone and a car driven by Slutsky. Although all four appellants complain about the seizure, only Mazzone and Slutsky have standing to do so; no rights of Kandis and Quagliato were violated by searches of vehicles in which they had no proprietary interest. The pertinent facts are few and simple. Pallone (a defendant, but not an appellant) was seen to hand Mazzone, through an open window in Mazzone’s van, a box wrapped in tape. The observing agents had sufficient reasons, unnecessary to dilate on here, to think that the box contained illegal drugs. Mazzone drove off in the van, followed by the agents. They stopped the van, arrested Mazzone, opened the box, and removed the drugs, which were used as evidence in the trial. Later the same day another group of agents saw Pallone driving a car in which Slutsky was a passenger. Pallone got out of the car, removed a white bag from the trunk, and handed it to Slutsky, who took it to a car parked nearby, put it in the trunk, and then drove off, followed by agents. The agents had every reason to believe that the bag contained illegal drugs. The agents flagged down Slutsky, arrested him, took the car keys out of the ignition, and opened the trunk. In it they found and opened the bag, in which was a box which they also opened. The box contained drugs, and the drugs were used in evidence in the trial. Although the agents had probable cause to believe that the two vehicles contained contraband, this does not conclude the case, since, as interpreted, the Fourth Amendment creates a presumption that a search not made pursuant to a warrant is unreasonable. See, e.g., Mincey v. Arizona, 437 U.S. 385, 390, 98 S.Ct. 2408, 2412, 57 L.Ed.2d 290 (1978). The most commonly stated rationale is that the interposition of a judicial officer between law enforcement agents and their quarry is a more effective protection against unreasonable searches than ex post remedies such as a motion to suppress illegally seized evidence or a tort suit for a violation of the Fourth Amendment. See, e.g., Johnson v. United States, 333 U.S. 10, 13-14, 68 S.Ct. 367, 368-69, 92 L.Ed. 436 (1948) (Jackson, J.). Although the framers of the Fourth Amendment were more fearful that the warrant would protect the police from the citizen’s tort suit through operation of the doctrine of official immunity than hopeful that the warrant would protect the citizen against the police, see Taylor, Two Studies in Constitutional Interpretation 23-43 (1969), and although the effective neutrality and independence of magistrates in ex parte proceedings for the issuance of search warrants may be doubted, there is a practical reason for requiring warrants where feasible: it forces the police to make a record before the search, rather than allowing them to conduct the search without prior investigation in the expectation that if the search is fruitful a rationalization for it will not be difficult to construct, working backwards. See, e.g., United States v. Mar *760 tinez-Fuerte, 428 U.S. 543, 565, 96 S.Ct. 3074, 3086, 49 L.Ed.2d 1116 (1976). But in the qualifying words “where feasible,” we approach the crux of the present case. If the police stop a car that they have probable cause to believe contains contraband, they cannot be made to let the car go on its way while they go get a search warrant; by the time they get it any contraband in the car will have been removed. It might seem that they could impound the car without searching it, while they seek a warrant. But to make sure that any valuable property in the car is secured against theft or charges of theft, they must conduct an inventory search when they seize a car; and since the privacy of the car’s interior is thereby compromised, it has not seemed worthwhile to require that a warrant be obtained. Moreover, all this assumes that the vehicle has been impounded, but impounding may be impractical in particular circumstances, as well as unnecessary if all that the police want to do is search the vehicle. See Arkansas v. Sanders, 442 U.S. 753, 765 n. 14, 99 S.Ct. 2586, 2594 n. 14, 61 L.Ed.2d 235 (1979). But when the search of a vehicle reveals a sealed container, it can be argued that the police should be forbidden to open it till they get a warrant. The inventory search can go forward but sealed containers found in that search should be left sealed until the magistrate decides that there is probable cause to believe that they contain contraband, or evidence of crime, or leads to such evidence. The police can take the sealed container to the police station if they want, and thus avoid the bother of impounding the car. See United States v. Esle, 743 F.2d 1465, 1468 (11th Cir.1984). Whatever its force, this argument has been rejected by the Supreme Court, whose present view is that if the police have probable cause to believe that there is contraband or other lawfully seizable material anywhere in the car they can search for it even if it is in a sealed container, or in a closed or even locked compartment such as the glove compartment or the trunk. See United States v. Ross, 456 U.S. 798, 821, 825, 102 S.Ct. 2157, 2171, 2173, 72 L.Ed.2d 572 (1982); United States v. Johns, — U.S.-, 105 S.Ct. 881, 885-86, 83 L.Ed.2d 890 (1985). If the Court held this view without qualification, the present case would be very easy to decide. But for the Court to have adopted the position in its full sweep it would have had to overrule United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538 (1977), and Arkansas v. Sanders, supra, which thus far it has been unwilling to do. In Chadwick federal agents followed a footlocker that they suspected contained contraband, and when the footlocker was put in the trunk of the defendant’s ear they stepped in and seized the footlocker and searched it. The Court viewed the site of the seizure — the trunk of the car — as a matter of happenstance. The purpose of having a vehicle exception to the requirement of a search warrant is to make searches of vehicles possible, but the agents didn’t want to search Chadwick’s car. Their focus was not on Chadwick at all; they didn’t even know whether he was a drug trafficker and knew what was in the footlocker. See 393 F.Supp. 763, 768 (D.Mass.1975). He might have been a complete dupe, in which event there would be no reason at all to think that his car contained contraband elsewhere than in the footlocker. And when the agents discovered the footlocker in the car they could easily have taken it to the police station and held it there until they could get a warrant to search inside it. They would not have had to impound the car. Sanders is the same case, only it involved a suitcase rather than a footlocker and a cab rather than a private car. In Johns, where the Supreme Court tried to reconcile Chadwick with the principle of Ross that allows the police to search anywhere inside a vehicle if they have probable cause to believe that it contains contraband (Johns makes no mention of Sanders, but Sanders is the same case as Chadwick, as we have said), the Court suggested that it matters whether the police are interested in a specific container or *761 in the contents of the vehicle generally. If the latter they can (provided of course that they have probable cause) conduct a thorough search and open anything inside the vehicle that might contain what they are looking for. But if they are just interested in a specific container, they have got to get a warrant to open it. See 105 S.Ct. at 884. After all, if police had stopped Mazzone and Slutsky on the sidewalk, each with a container in his hand which the police reasonably believed contained illegal drugs, the police could not have opened the containers on the spot and searched them (assuming that such a search would not be a permissible incident to the arrests, a possible theory in this case, too, with regard to Mazzone, see New York v. Belton, 453 U.S. 454, 101 S.Ct. 2860, 69 L.Ed.2d 768 (1981), but not one that we shall have to consider). They would have had to arrest Mazzone and Slutsky and take the containers down to the station and apply for a warrant to open them. If the fact that the containers are found- in a vehicle is incidental, the police cannot use the fortuity of finding the containers there to justify dispensing with a search warrant. To put the same thought in slightly different words, if (as the facts of Chadwick and Sanders suggest) the police have probable cause only to believe that the container contains contraband they cannot, by waiting until the container happens to be placed in a vehicle, dispense with the usual requirement (strikingly illustrated by Walter v. United States, 447 U.S. 649, 100 S.Ct. 2395, 65 L.Ed.2d 410 (1980)) that a sealed container may not be opened without its owner’s permission or a search warrant. This is a fine distinction — perhaps (the facts of Johns suggest) even finer than the rule stated in the Court’s opinion. As the agents approached the defendants’ truck they detected an odor of marijuana emanating from it, and as they came closer they saw packages in the back of the truck of the sort they knew to be used for packing marijuana. Putting two and two together they concluded that the odor was coming from the packages and they seized and later opened them. So far as appears they didn’t search the truck. Thus it seems that their interest was in the packages, not in the vehicle; that as in Chadwick it was happenstance that the packages were found in a vehicle. It is true that the odor of marijuana gave them probable cause to search the entire vehicle, but only until they saw the packages, at which point it became highly likely that the only contraband in the truck was in the packages. Thus Johns confines Chadwick and Sanders on a very short tether indeed; and perhaps this was inevitable. For imagine the Court’s having held in Johns that the agents could not search the packages without a warrant once they were sure the packages were where the marijuana was, but could search the packages without a warrant if the agents were less observant or knowledgeable and hence did not know where in the vehicle the marijuana was. Lethargy and stupidity would be rewarded by such a rule. It seems, then, that if there is any reason for searching beyond the container that is the primary focus of the agents’ interest, they can search it without a warrant even if (in this case as in Johns) they don’t bother to search anywhere else in the vehicle. By this standard, the appellants must lose. The agents had reason to believe that Mazzone and Slutsky were engaged in illegal drug trafficking, and once they saw Pallone hand what they had reason to think were drugs to Mazzone in the van and later to Slutsky who put them in the trunk of his car, they also had reason to believe that a search of the vehicles might turn up additional drugs, or drug paraphernalia, or proceeds of other drug sales. Mazzone might have picked up drugs from someone else besides Pallone; Slutsky might have had other drugs in his trunk. This is speculative but the fact that a particular package provides the most important reason for the search of a vehicle does not make it any the less a search of the vehicle, provided that there is some reason to search the rest of the vehicle, and there was here. It would not have been just a fishing expedition, as a search of the *762 car in Chadwick or even more clearly the cab in Sanders would have been. Indeed, since quite apart from the particular containers that were the focus of the agents’ interest the agents had reason to believe that Mazzone and Slutsky were active members of a substantial drug ring, it was rather a dereliction of duty for the agents, knowing what they knew, not to search Mazzone’s van and Slutsky’s car but merely to seize the packages. Although we have found no case exactly like this one, we have found several where observing a suspicious container being placed in a vehicle was a link in establishing probable cause to search the container without a warrant. See United States v. Marin, 761 F.2d 426, 432 (7th Cir.1985); United States v. Wiecking, 757 F.2d 969 (9th Cir.1983); United States v. Shepherd, 714 F.2d 316, 317 (4th Cir.1983). The courts in those cases did not even ask whether there was probable cause to search elsewhere than in the container that the police had seen brought into the vehicle. The rule we distill from these cases, from Johns, and from the need to provide the police with (if possible) simple and readily applicable guidelines to lawful searches of vehicles, is that if there is some reason to believe that a full search may turn up more than is just in the sealed container, the police can make the full search and as part of it open and search the container itself. The rule was flunked in Chadwick and Sanders but is passed in this case as in all the other pertinent cases since Ross that we have found. We are not troubled by the thought that police might deliberately wait to seize a package that they had probable cause to believe contained contraband until it was placed in a vehicle, in order that they could search the package without a warrant. That would be Chadwick or Sanders —casés where the legitimate interest of the police was at all times confined to the package, and its mobile site was adventitious. But here there was good reason not to arrest Pallone en route to Mazzone’s van, and not to arrest Pallone and Slutsky until they saw what Slutsky did with the drugs. If the agents had arrested Pallone before he made the deliveries they would have nabbed Pallone, whom they got anyway, but they would not have gotten Mazzone or Slutsky. The circumstances rebut any inference that the agents delayed just in order to be able to open the containers without a search warrant. The other issue in these appeals involves remarks by the prosecutor in the rebuttal portion of his closing argument. He said such things as, “I know as good defense lawyers it is their job to defend their clients, no matter how guilty they may perceive them____” And: “I took an oath to uphold the laws of the United States, that was one of the most important things I ever did, and, of course, if the government’s case here today is a lie, then I haven’t been upholding that oath.” And: “as an Assistant United States Attorney I am proud to be in this case. ... I think ... these agents did exactly what they were supposed to have done and they have helped rid the community of these dope peddling defendants.” He described some defense documents as “forged-up documents,” and repeatedly described the closing argument by the defendants’ lawyers as “malarky” and “hogwash.” The district judge sustained several objections to the argument and told the jury to pay no heed to the objectionable remarks. The government argues that the prosecutor’s comments were proper because invited by the defense lawyers, who in their five hours of closing argument accused the government of perjury, fraud, intimidation, and other improprieties in prosecuting their clients. We reject the argument. The federal government’s lawyers may not fight fire with fire. If defense counsel exceed proper bounds in their closing arguments, the prosecutor can object; he can, if need be, ask that counsel be held in contempt for improper argument or questions (we sustained such a contempt judgment in United States v. Lowery, 733 F.2d 441 (7th Cir.1984)); but he cannot *763 respond in kind and violate ethical standards himself. We deprecate the prosecutor’s improper remarks in closing argument and the government’s bold effort not merely to excuse but to justify those remarks in its brief in this court. But it is a separate question whether the government’s misconduct justifies our reversing the appellants’ convictions. It is at this point and this point only that the doctrine of “invited response” becomes relevant. Properly understood that doctrine does not condone the prosecutor’s descending to the level of defense counsel or enact the proposition that two wrongs make a right; it merely recognizes that the impact on the defendant from the prosecutor’s misbehavior may be less if the defendant’s counsel aroused the jury against the prosecutor. See United States v. Young, — U.S. -, 105 S.Ct. 1038, 1044-45, 84 L.Ed.2d 1 (1985). The prosecutor’s misconduct may just have offset the defense counsel’s misconduct, thus producing no net effect on the jury’s deliberations. In such a case ordering the conviction set aside would be punitive rather than remedial — a windfall to a defendant who hadn’t been hurt by the prosecutor’s misconduct (except insofar as it may have neutralized his own counsel’s misconduct), just as an award of punitive damages in a tort case is a windfall to the plaintiff because it is superadded to the plaintiff’s actual loss from the defendant’s misconduct. The windfall here would be a modest one since if an error is truly harmless the next trial is likely to end the same way as the first, with a conviction; but there is always some chance of an acquittal and the prosecution has no remedy even if the acquittal is irrational. But if modest, the windfall is not costless. The lawyers for both sides must be paid, and the scarce resources of the judicial system expended, in a second trial. Notice that most of the costs do not even fall on the malefactor, the misbehaving prosecutor. It is therefore no surprise that we have no general power to set aside convictions merely to punish prosecutors. It is better to punish the prosecutor directly; there is no lack of direct sanctions for a lawyer’s misconduct, of which improper advocacy is a well-recognized species. See United States v. Young, supra, 105 S.Ct. at 1043-44. We must therefore decide whether the error of the prosecutor’s ways was harmless, or as the cases usually say, did not prejudice the defendants’ right to a fair trial. See United States v. Brack, 747 F.2d 1142, 1152-53 (7th Cir.1984); United States v. Bagaric, 706 F.2d 42, 58-59 (2d Cir.1983); United States v. Haskins, 737 F.2d 844, 850 (10th Cir.1984); United States v. Falk, 605 F.2d 1005, 1014 (7th Cir.1979) (concurring opinion, but joined in by another judge, and thus the majority opinion on this point). It may seem odd to run the issue of harmlessness in with the issue of the fairness of the trial in this way, but it is not once the precise nature of the alleged misconduct is understood. Misconduct by the prosecutor that is promptly and vigorously corrected by the judge is not reversible error; misconduct that poisons the trial is. The difference is simply the likely impact of the misconduct on the jury. If the impact seems to have been nil (“harmless”), that is just another way of saying that the trial was not poisoned, due process was not denied, reversible error was not committed. That is the case here. The district court concluded that the prosecutor’s misconduct had not affected the trial and his conclusion deserves to be given substantial weight. After all, he was able to observe the jury and thus gauge the probable impact on it of the prosecutor’s remarks and of his own admonitions, and we are not. The judge was not evaluating his own conduct (which is awfully hard to do objectively) but the prosecutor’s, since the judge had sustained defense counsel’s objections to the prosecutor’s improper statements and had told the jury to ignore them. The case is thus like United States v. Bruscino, 687 F.2d 938, 941 (7th Cir.1982) (en banc), which held that great deference was due the district judge’s judgment on wheth *764 er the jury was poisoned by the accidental introduction (through no fault of the judge) into the jury room of documents that had not been admitted into evidence. The presumption in our system, artificial as it may sometimes be, is that curative instructions cure, that admonitions to the jury are taken seriously; and if the trial judge believes they were taken seriously as he did here, ordinarily we will have little basis for disbelieving him. We are not quite so naive as to believe that telling jurors not to think about something will cause them to forget it, but we trust juries to behave responsibly and to put aside as considerations bearing on their judgment matters that the judge tells them to put aside. At least within limits; for in a close case, egregious prosecutorial misconduct might require us to overturn the trial judge’s finding that it had not prejudiced the jury’s consideration of the case. Cf. Joseph v. Brierton, 739 F.2d 1244, 1247-48 (7th Cir. 1984). This, however, was not a close case. The evidence against the appellants was overwhelming; it included substantial eyewitness evidence by federal agents as well as physical evidence (the drugs that were seized). The prosecutor’s remarks, those we quoted above and a few more like them (but somewhat less objectionable), must be placed in the context of a 45-minute rebuttal, the rest of which was unobjectionable, following five hours of vigorous closing argument by the defendants’ counsel. It is almost inconceivable that if the prosecutor had refrained from making the remarks that he did, the appellants would have been acquitted. It is more likely that the judge’s admonitions reduced the stature of the assistant U.S. attorney in the jury’s eyes than that the remarks that were admonished swayed the jury. That does not excuse the government’s conduct, of course. It makes it, if anything, less excusable; in Talleyrand’s phrase, it was worse than a crime, it was a blunder. So while we reprimand the government for the prosecutor’s conduct and for its attempt to defend that conduct before us, we do not find reversible error: a conclusion reinforced by the very large number of cases that refuse to reverse convictions for prosecutorial excesses of the kind committed here. See, e.g., United States v. Howard, 774 F.2d 838, 847-49 (7th Cir. 1985) (prosecutor said, “My job is to bring out the truth,” etc.); United States v. West, 670 F.2d 675, 689 (7th Cir.1982) (“The government doesn’t need cases”); United States ex rel. Clark v. Fike, 538 F.2d 750, 758 (7th Cir.1976) (“I [the prosecutor] work for you as well as for the defendant”); United States v. Strmel, 744 F.2d 1086, 1089 (5th Cir.1984) (“thank God for Customs and the Drug Enforcement Administration____ Thank God for their diligence and their expertise”); United States v. Shackelford, 709 F.2d 911, 913 (5th Cir. 1983) (per curiam) (“You [the jury] were to be tricked [by defense counsel]”); United States v. Esposito, 523 F.2d 242, 251 (7th Cir.1975) (prosecutor described defendant as “the lowest form of life in the country”). The strongest case for the appellants, United States v. Garza, 608 F.2d 659 (5th Cir.1979), is distinguishable. The improper remarks were more egregious, and also more prejudicial since it was a close case. See id. at 661-62, 665-66. It is true that the court found plain error, see id. at 665-66, which is more than just not harmless error, see United States v. Silverstein, 732 F.2d 1338, 1349 (7th Cir.1984). Defense counsel had not objected to the prosecutor’s comments, which was why the issue was cast in terms of plain error; but, probably because defense counsel did not object, the trial judge did not admonish the jury to ignore the improper comments — a highly prejudicial omission with no counterpart in the present case. No two cases in this area are alike but it seems unlikely that any court would reverse on the facts presented here. Our point, however, is not that the defendants were denied a fair trial yet that somehow doesn’t matter because they probably would have been convicted anyway; the point is that prosecutorial misconduct that is unlikely to affect the jury’s deliberations is not a denial of a fair trial, and hence is not reversible error. *765 While we are admonishing counsel, we take the opportunity to admonish the appellants’ counsel to pay greater heed in the future to the rules and precepts of this court regarding the briefing of appeals. Apparently frustrated by the 50-page limitation on a main brief, the appellants’ counsel placed a portion of their statement of facts in a separately paginated appendix. They also presented their regular statement of facts in the form of a summary of each witness’s testimony. If they had presented the facts in narrative form as we require they probably would not have needed to file a brief in excess of 50 pages; but if they did need more than 50 pages they should have requested leave to file an oversized brief, as our rules require, rather than engaging in subterfuge. See United States v. Devine, 768 F.2d 210 (7th Cir. 1985) (en banc) (per curiam); Prudential Ins. Co. v. Sipula, 776 F.2d 157, 161 n. 1 (7th Cir.1985). AFFIRMED. SWYGERT, Senior Circuit Judge, concurs in the result.