Israel Travel Advisory Service, Inc., Celia Shar, and Marilyn Ziemke v. Israel Identity Tours, Inc., Larry Ritter, and Marlene Ritter

EASTERBROOK, Circuit Judge. Bar mitzvah tours of Israel. That is the market defined for the antitrust claim in this case. It is an absurd market definition. For a market is defined to aid in identifying any ability to raise price by curtailing output. Ball Memorial Hospital, Inc. v. Mutual Hospital Insurance, Inc., 784 F.2d 1325, *1253 1335-87 (7th Cir.1986). These litigants do not own or control the assets that produce output in a travel market: fleets of airplanes, chains of hotels, Israel and its history. Anyone can buy tickets, rent rooms, and conduct tours. If these firms were to vanish tomorrow production would be unaffected. Their principal assets — skills as marketers and guides — would survive. People rather than firms hold expertise (economists call it “human capital” for good reason), and they can change affiliation. Output often grows in the process as competition intensifies. That is not only an economic theory but also the genesis of this case. Israel Travel Advisory Service (itas) specializes in bar and bat mitz-vah tours of Israel. Families tour historical sites, with a celebration arranged at a place of religious significance. Israel Identity Tours (iit), founded by disaffected employees of itas, began vigorous competition, itas remains the larger firm and has increased its sales consistently — though itas accuses iit of monopolization! — but begrudges iit even a single customer. The competition, like the litigation, is marred by charges of deception and defamation. We have a spite match, and as in many such contests the parties have raised every imaginable claim and others that exceed the bounds of imagination. To make this opinion tractable, we simplify greatly. Celia Shar and Marilyn Ziemke founded itas in 1970. Parents who want to arrange religious celebrations for their children in Israel as part of a package tour are a select market, and itas (like its rivals) is an in-home operation. Shar and Ziemke recruited experienced guides in Israel; they relied on a combination of ads and word-of-mouth to obtain clients in the United States. The best promoters are persons well known and respected in the Jewish community. llene Wallerstein represented itas in Chicago, and by all accounts she was successful in obtaining business. Eventually she had a falling-out with Shar and Ziemke. Wallerstein quit in 1990 and set up a rival in Illinois; Karen Kaplan, itas’s sales representative in Boston, also quit. Ami Ben Geller, one of itas’s former tour leaders in Israel, persuaded his friend Larry Ritter to set up a bar mitzvah travel service in New Jersey. Geller agreed to be the guide; Transglobal Travel, itas’s former land facilitator in Israel, also switched allegiances. The iit in the caption of this case is Larry Ritter’s operation; Waller-stein’s firm, incorporated in Illinois, used the identical name (which to reduce confusion we call HT-Illinois). Wallerstein, Kaplan, and Ritter cooperated, and their ads implied that the two corporations are a single firm. One of the issues in the case is how close that cooperation became. llene Wallerstein began to slander itas, telling potential customers that it was on the verge of bankruptcy — a serious charge, because travelers do not want to hand over thousands of dollars to a firm that may furnish a chose in action rather than travel. Being stranded with worthless vouchers is no one’s idea of a good trip, itas sued iit-Illinois and Wallerstein, who settled. But if iit was a joint venture or conspirator with HT-Illinois, then iit is jointly liable for Wal-lerstein’s words, itas has numerous other gripes. Ads for iit trumpeted 33 years’ experience; itas says that this is a lie because iit had just commenced business, while iit says that the number referred to Geller’s long experience as a guide, although one of the ads did not name Geller. iit implied that Joel Leibowitz was affiliated with the firm, which itas says he was not. itas also wants to hold iit liable for disparaging comments Leibowitz made, on the theory that he had apparent if not actual authority to represent iit. When itas began to offer a “free” tour of Israel to the bar or bat mitzvah celebrant, iit responded by advertising “free” tours of its own; itas believes that these ads and corresponding oral statements to potential customers were false, or at least misleading, because iit offered only “free” land arrangements, while itas included air fare too. (We put “free” in quotations because neither firm offered anything for free; money from the sale of travel arrangements to the celebrants’ parents and siblings covered the costs.) Then there is a claim that Marlene Ritter, Larry’s wife, told one potential customer that itas and iit are “the same thing.” All of this, according to itas, violated the common law of defamation and unfair competition, consumer protection laws in Illinois and New *1254 Jersey, the Sherman Antitrust Act, the Lan-ham Trademark Act, and the Racketeer Influenced and Corrupt Organizations Act (Rico). For its part, iit filed a counterclaim seeking relief on nine theories. The district court carved several claims and parties from the case by partial summary judgment. What was left was tried to a jury — although the judge resolved some additional claims in mid-trial under Fed. R.Civ.P. 50. The jury concluded that iit (but not Larry Ritter) had slandered itas. It awarded $75,000 in compensatory and $102,-945 in punitive damages for this tort. The jury rejected every other claim presented to it, and the judge rejected all efforts to upset the verdict. 1994 U.S.Dist. Lexis 751. Now we have cross-appeals, iit believes that it should have prevailed on the defamation theory and counterclaims, while itas thinks that it is entitled to a larger recovery, and from the Ritters personally as well as from their corporation. We start with iit’s argument that it should be itas’s judgment creditor rather than its judgment debtor. Wallerstein made the defamatory remarks that led to this verdict, iit could be responsible for them only if the two iits were joint adventurers. Firms do not act by themselves; people are responsible for their deeds, so iit could be responsible. only if Larry or Marlene Ritter made common cause with Wallerstein. The district judge granted summary judgment in Marlene’s favor, and the jury ruled in Larry’s. Yet if neither Ritter was responsible, how can iit, whose liability is vicarious, be responsible? How can a corporation be ordered to pay punitive damages when no natural person has committed a tort? Perhaps iit’s liability depends exclusively on Leibowitz’s statements, but these would be slender support for a verdict of the size the jury returned. Perhaps the jury’s verdicts are simply inconsistent. As it happens, iit did not point out the inconsistency to the district judge, and it has not argued to us that the jury’s absolution of Larry Ritter calls for a new trial — the right way to deal with such problems, for neither of the inconsistent verdicts has priority. American Casualty Co. v. B. Cianciolo, Inc., 987 F.2d 1302, 1305 (7th Cir.1993). We do not invent arguments for the parties in civil litigation and therefore pass to the contentions iit presses. What iit does argue is that it lacked an adequate opportunity to develop its defense. It wanted to take depositions of Shar and Ziemke, itas’s principals. The district court refused to compel Shar and Ziemke to attend the depositions iit noticed, a step that iit believes entitles it to another trial. We don’t see why. Harmless errors do not vitiate verdicts, and iit does not explain how the lack of discovery injured it. It does not say that any surprising evidence came out at trial or that depositions would have assisted in the exploration of a material issue or influenced trial strategy. At all events, the district court did not err. The discovery cutoff was March 15, 1993. Having twice extended this date, the district judge told the parties that she would not do so a third time. On March 3, 1993, iit mailed four notices of deposition; itas received them on March 8. They called for the deposition of Shar and Ziemke in Chicago on March 11 and 12; iit wanted to take depositions of other witnesses in New Jersey on the same dates. Before mailing these notices, iit had received itas’s notices of third-party depositions for upstate New York on March 11 and 12. It cannot have come as much surprise to iit that the judge refused to compel itas’s lead counsel to rearrange his schedule — or to be in three places at once — when iit had waited until the end of discovery to schedule the depositions of its adversaries, itas had offered to make Shar and Ziemke available earlier, but iit was not interested. The district court did not commit clear error; to the contrary, her decision was clearly correct. Gamesmanship had greater costs for iit. Throughout discovery iit was less than forthcoming in producing documents itas sought. Three times the district judge granted itas’s motions to compel. The judge characterized iit’s conduct as “continued testing of the bounds of permissible discovery,” 1993 U.S.Dist. Lexis 7433, *2, followed by conduct beyond permissible bounds— flouting the court’s orders. The court directed iit to produce certain documents, which counsel said they could not find, only to be *1255 embarrassed by Larry Ritter’s admissions (a) that counsel had never asked him to locate the documents, and (b) that he had destroyed pertinent documents, including correspondence with Wallerstein, after itas had filed motions to compel. The judge also ordered iit to produce all financial records by February 13, 1993, on pain of default judgment. On March 3, 1993, itas took the deposition of iit’s accountant, who revealed that certain financial records had been prepared and given to iit. (In fairness to iit’s current lawyers, we add that iit changed counsel after these episodes.) When itas found that these had not been produced, it moved for sanctions. Instead of following through on her threat to write finis to the whole case, as she could have done, see Fed. R.Civ.P. 37(b)(2)(C), the judge chose a milder sanction: she dismissed iit’s counterclaims. 1993 U.S.Dist. Lexis 4435. On the counterclaims iit was the plaintiff, and a plaintiff who does not cooperate in discovery must expect to lose. Otis v. Chicago, 29 F.3d 1159, 1168 (7th Cir.1994) (en banc); Newman v. Metropolitan Pier & Exposition Authority, 962 F.2d 589 (7th Cir.1992); Metropolitan Life Insurance Co. v. Cammon, 929 F.2d 1220 (7th Cir.1991). Our review is deferential, and the judge did not abuse her discretion. The judge gave ample warning, and the sanction was proportional to the wrong. See United States v. Golden Elevator, Inc., 27 F.3d 301 (7th Cir.1994); Ball v. Chicago, 2 F.3d 752 (7th Cir.1993). iit proffered excuses, but the judge did not have to accept them. Ordered to disclose “all corporate records,” iit did not turn over its checking account records. Why, nothing in the order mentioned checks, it contends. Come, come. A lawyer who makes such an argument can’t expect anything else he says in the litigation to be believed. We have dropped a discussion of the counterclaims into the midst of assessing iit’s request for a new trial because Ritter’s destruction of documents influences another of iit’s contentions: that the evidence does not show that Ritter and Wallerstein were in league against itas. Destruction of evidence can support a contrary inference, and there was more. Larry Ritter told his friends that the only reason he was in the travel business was to destroy itas. Wanting harm, even bankruptcy, to come to one’s business rivals is not actionable; hatred is a spur to competition, which serves consumers’ interests. Entrepreneurs are privileged to compete because any effort to separate pure from impure motives would in the end undercut the power of rivalry to promote consumers’ welfare. See Oliver Wendell Holmes, Privilege, Malice & Intent, 8 Harv.L.Rev. 1 (1894), reprinted in 3 Collected Works of Justice Holmes 371, 373-74 (Sheldon M. Novick ed. 1995). Although the motive for going into business is not actionable, the mind may illuminate the means, may disambiguate borderline practices, itas’s theory is that Ritter engaged in cooperative advertising with Wal-lerstein and referred potential clients to her for advice, knowing that she would impugn itas’s solvency, to the advantage of both iit firms. Several customers testified to both the referrals and the defamation. Perhaps they were stretching matters; or perhaps, as iit argued, itas really was in financial straits, but the jury could credit these witnesses and accept the inferences itas pressed. Pace iit, that we review without deference the district judge’s decision refusing to grant judgment as a matter of law does not imply that we review the evidence de novo. The question for a court of appeals is whether a reasonable juror could find the evidence sufficient under a preponderance standard. Mayer v. Gary Partners & Co., 29 F.3d 330 (7th Cir.1994). The answer is yes. As for damages: general damages were permissible, and the award of $75,000 was proper under the circumstances. A reasonable jury could conclude that the defamatory statements misled some travelers, inducing them to use iit when, with correct information, they would have used itas. Quantifying the loss was exceedingly difficult; how does itas prove a counterfactual proposition about the behavior of persons who bought iit’s services? itas was able to prove that lies had been told, but the extent of their effect was bound to be problematic. That’s why general damages are available in the law of defamation. See Prosser & Keeton on *1256 Torts 791 & n. 80, 843 (5th ed. 1984). iit stresses that all of the witnesses at trial eventually got their facts straight and used itas’s services, but this says more about the way parties find witnesses in litigation than about the effects of the slander. The people most likely to report to itas that they had been told unpleasant things by iit are the ones who doubted iit’s statements and sought more information; the ones taken in would not have come to itas’s attention. The evidence did enough to demonstrate malice that punitive damages were available. The ratio of punitive to actual damages is not excessive, and iit’s slim net worth did not preclude the award. A large net worth is a bad reason to award punitive damages, Zazú Designs v. L’Oréal, S.A., 979 F.2d 499, 508-09 (7th Cir.1992), and a small net worth is a bad reason to let the wrongdoer off the hook. itas believes that it is entitled to more than the jury awarded. It invokes two statutes, Rico and the Sherman Act, that provide for treble damages and attorneys’ fees. The district judge granted summary judgment to iit on the Rico claim, and it threw out the antitrust claim at the close of itas’s case. The latter decision is obviously correct. We have explained already what is wrong with itas’s market definition — and, if “bar mitzvah tours of Israel by families who live in New Jersey or near Chicago or Boston” (the full “definition”) were indeed a market, then the dominant firm would be itas itself, a monopolist by its own admission until 1990. iit’s entry upset the larger firm, but the antitrust laws encourage such upsets. It is an abuse of the antitrust laws to invoke them to stifle nascent competition. Stamatakis Industries, Inc. v. King, 965 F.2d 469 (7th Cir.1992). Although itas believes that iit’s “free” land tours for celebrants amounted to predatory pricing because iit lost money, it has not begun to satisfy the requirements of a predatory-pricing case. Chief among these is the need to show that the “victor” in the price war can recoup its losses by charging supracompetitive prices. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., — U.S. -, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993); Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986); A.A. Poultry Farms, Inc. v. Rose Acre Farms, Inc., 881 F.2d 1396 (7th Cir.1989). Unless the “predator” can recoup, consumers are net beneficiaries, iit has not achieved a dominant position in the market (itas remains much larger), and because rivals could freely enter if iit drove out itas, recoupment would be impossible. So if iit indeed charged less than its costs, there is no reason to get excited; iit was giving money away to consumers rather than injuring them by curtailing output and charging monopoly prices. Cf. Spectrum Sports, Inc. v. McQuillan, — U.S. -, 113 S.Ct. 884, 122 L.Ed.2d 247 (1993) (“dangerous probability of success” is an essential element in a § 2 case). More likely it was discounting to get established in the market, also a lawful step, see Phillip Areeda & Donald F. Turner, 3 Antitrust Law ¶ 716 (1978), but we need not pursue the subject. Some antitrust offenses do not depend on proof of market power. Price fixing and market allocation, for example, are illegal per se whether or not the firms have any hope of success. Palmer v. BRG of Georgia, Inc., 498 U.S. 46, 111 S.Ct. 401, 112 L.Ed.2d 349 (1990); FTC v. Superior Court Trial Lawyers, 493 U.S. 411, 110 S.Ct. 768, 107 L.Ed.2d 851 (1990); Blackburn v. Sweeney, 53 F.3d 825 (7th Cir.1995); 15 U.S.C. § 1. Let us suppose that iit and HT-Illinois conspired to fix prices. How does this injure itas? If the iit firms raised their price (the concern with price fixing), then itas would have been the beneficiary and could raise its own prices. Acts harmful to business rivals generally help consumers; this fact is the basis of the antitrust injury rule of Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977). A merger unlawful because of its tendency to increase prices aids rivals; conversely, a claim by the rival of the merged firms that the merger has intensified competition shows how the rival suffers injury, but not how consumers lose. Such a rival has actual injury but does not suffer from the effect that makes the transaction unlawful under the antitrust laws and therefore, Brunswick held, cannot invoke the antitrust laws. See also Cargill, Inc. v. Monfort of Colorado, *1257 Inc., 479 U.S. 104, 107 S.Ct. 484, 93 L.Ed.2d 427 (1986); Chicago Professional Sports Limited Partnership v. National Basketball Ass’n, 961 F.2d 667, 670 (7th Cir.1992); Indiana Grocery, Inc. v. Super Valu Stores, Inc., 864 F.2d 1409 (7th Cir.1989); Zinser v. Rose, 868 F.2d 938 (7th Cir.1989). We know from Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990), that the antitrust injury doctrine applies with full vigor to the per se offenses, itas cannot show how price fixing by the two iits injured both producers and consumers. As we held in Stamatakis, the antitrust injury doctrine KOs claims based on injury caused by defecting employees who set up rivals. This antitrust claim was stillborn, and the district judge’s only error was waiting as long as she did to scotch it. The Rico claim is more substantial. The district judge granted summary judgment against itas because the predicate offenses, mail fraud, involved consumers rather than itas itself. The judge believed that indirect injury is insufficient after Holmes v. SIPC, 503 U.S. 258, 112 S.Ct. 1311, 117 L.Ed.2d 532 (1992). 1993 WL 155503, 1993 U.S.Dist. Lexis 6141, rehearing denied, 1993 WL 239051, 1993 U.S.Dist. Lexis 8744. Yet the Supreme Court observed, 503 U.S. at 272-73 n. 20, 112 S.Ct. at 1320 n. 20, that it did not mean to preclude all possibility of recovery for injury that was transmitted indirectly; the holding of Holmes is no more than that common law ideas about proximate causation inform the understanding of Rioo. Holmes explicitly refrained from importing a theory akin to the antitrust-injury doctrine into the law of rioo. Id. at 269 n. 15, 112 S.Ct. at 1318 n. 15. Sometimes a recovery for a derivative injury would undermine rioo’s own purposes, or those of other statutes; the Court gave as one illustration our holding in Mid-State Fertilizer Co. v. Exchange National Bank, 877 F.2d 1333 (7th Cir.1989), that shareholders cannot use Rico to recover for injury done to their corporation. Such a process would create unnecessary complexities in apportioning recoveries, and at the same time would undermine the interest of the corporation’s creditors in collecting their debts. (The shareholders wanted to bypass the corporation precisely because it was insolvent, and its creditors would have reaped most of the benefit.) Holmes dealt with a similar claim, by one of a corporation’s creditors rather than an equity investor. A fraudulent scheme injured the corporation and its clients, causing the Securities Investor Protection Corp. to advance money to protect the clients. The corporation had other debts — and other sources of financial distress. The Supreme Court concluded that the clients and the injured corporation were the proper plaintiffs; whatever the corporation obtained would inure to the benefit of its creditors, including sipc, according to their priorities in bankruptcy. Our case does not present a similar apportionment problem, and there is no risk of multiple recoveries. The “direct” victims — as the district court understood things, the readers of iit’s ads — got what they paid for and are not complaining. Holmes called for the use of common-law analogies under 18 U.S.C. § 1964(c), and there is no doubt that a producer injured by a campaign of misinformation directed at its customers suffers an injury compensable under the law of torts; it is not cut off by the proximate-causation and foreseeability requirements. Prosser & Keeton on Torts at 1013-20. After giving thoughtful treatment to Holmes, the fourth circuit held in Mid Atlantic Telecom, Inc. v. Long Distance Services, Inc., 18 F.3d 260 (4th Cir.1994), that Rico similarly allows suits when the predicate offenses influence customers and, derivatively, injure business rivals. A conflict among the circuits on such a question would be regrettable, and we therefore accept Mid Atlantic. (The district court’s alternate holding that itas cannot show how it relied on the misrepresentations made to its potential customers just restates the conclusion that derivative injury is never cognizable under Rico; it does not require separate analysis.) Like a troll under the bridge, another problem lies in wait for a litigant that has come this far. An entrepreneur can recover under the common law of unfair competition when a rival tells fibs to potential customers, because the rule against fraud has been crafted for the benefit of both seller *1258 and customer. Can one say the same about the federal mail fraud statute, 18 U.S.C. § 1341, the only criminal predicate act in iit’s supposed “pattern of racketeering”? Only one court of appeals has addressed this question, and it answered “no.” Lancaster Community Hospital v. Antelope Valley Hospital District, 940 F.2d 397, 405-06 (9th Cir.1991). Lancaster Community Hospital held that business rivals may not use Rico to complain about injuries derivatively caused by mail frauds perpetrated against customers, because only the customers are the beneficiaries of the statutory protection. We are no more inclined to create a conflict on this question than we were on the Mid Atlantic issue — especially not after Jepson, Inc. v. Makita Corp., 34 F.3d 1321, 1327 (7th Cir.1994), indicated tentative agreement with Lancaster Community Hospital. Tentative now becomes definitive. Lancaster Community Hospital relied in part on McNally v. United States, 483 U.S. 350, 107 S.Ct. 2875, 97 L.Ed.2d 292 (1987), which jettisoned the intangible rights doctrine in mail fraud cases. The ninth circuit thought that preservation of market share is an intangible right outside the scope of § 1341. Congress has reinstated the intangible rights approach, 18 U.S.C. § 1346, but we do not think that this changes anything. Lost business opportunities, the subject of both Lancaster Community Hospital and this case, are property as § 1341 used that term before the amendment. See Carpenter v. United States, 484 U.S. 19, 108 S.Ct. 316, 98 L.Ed.2d 275 (1987). The essential point, rather, is that the victims of the fraud are the object of solicitude; § 1341 does not establish a regimen of truth-telling without regard to details like who is losing out and why. See United States v. Walters, 997 F.2d 1219 (7th Cir.1993). Such a limitation is common in tort law. A plaintiff claiming injury by the defendant’s violation of a statute must show not only that the defendant violated the law but also that the plaintiff is among the persons protected by the law. Prosser & Keeton on Torts at 225-26. Thus violation of a law requiring animals to be kept in pens on shipboard does not lead to liability if the animals are washed overboard; the function of the law was to curb disease, not to reduce the effects of storms. Gorris v. Scott, 9 Ex. 125 (1874). In federal administrative law, this principle is known as the “zone of interests” requirement. Director, OWCP v. Newport News Shipbuilding & Dry Dock Co., — U.S. -, -, 115 S.Ct. 1278, 1283-84, 131 L.Ed.2d 160 (1995); Clarke v. Securities Industry Ass’n, 479 U.S. 388, 395-96, 107 S.Ct. 750, 754-55, 93 L.Ed.2d 757 (1987); Association of Data Processing Service Organizations, Inc. v. Camp, 397 U.S. 150, 153, 90 S.Ct. 827, 829-30, 25 L.Ed.2d 184 (1970). See also Holmes, 503 U.S. at 287-88, 112 S.Ct. at 1328 (Scalia, J., concurring). It carries over to criminal law. For example, in determining the “loss” from iit’s ad campaign under the Sentencing Guidelines (if indeed iit committed mail fraud, a subject we do not broach), a court would not inquire how much itas lost; it would ask only what the travelers lost. U.S.S.G. § 2F1.1 Application Note 7(a), (c); see also § 3D1.2 Application Note 2; United States v. Marlatt, 24 F.3d 1005 (7th Cir.1994). So too with civil RICO now that Holmes has adopted common-law approaches to decide when Rico authorizes recovery for derivative injuries. Section 1341 does not protect vendors to persons who may be deceived, and firms suffering derivative injury from business torts therefore must continue to rely on the common law and the Lanham Act rather than resorting to RICO. itas won under the common law, and it lost under the Lanham Act. The jury decided that neither iit nor Larry Ritter had violated that statute. The verdict drains all significance from several of itas’s arguments. Consider, for example, the contention that Marlene Ritter should have been among the defendants at trial. Only one accusation has been leveled against her: that she told a potential customer that itas and iit are the same thing. The customer related calling the iit business number (which rings in the Ritter home) and talking with a woman. itas’s argument depends on the fact that Marlene Ritter is the most logical woman. The district court granted summary judgment against itas, ruling that it had not excluded the possibility that some other woman took the call. 1993 WL 155503, *5, *1259 1993 U.S.Dist. Lexis 6141, *16-17. Yet litigants need not surmount such a hurdle. If a reasonable juror could infer that Marlene Ritter spoke with the customer, then summary judgment was inappropriate, itas could meet this burden by showing that iit lacked other female call-takers. But so what? The person who was told that itas and iit are the same entity took a tour with itas. Proof of this incident at trial therefore could not have affected the quantum of damages, and itas does not argue that it needs Marlene Ritter on the judgment to improve its chances of collection. As things developed at trial — including the fact that the jury heard via deposition the tale of the customer who believes that she spoke with Marlene Ritter — the order removing her from the case was inconsequential. So too with claims itas presents under the Illinois Consumer Fraud Act, 816 ILCS 505/1 to 505/lla, the Illinois Deceptive Trade Practices Act, 815 ILCS 510/1 to 510/7, and the New Jersey Consumer Fraud Act, N.J.Rev.Stat. 56:8-1 to 56:8-20. The district court removed these from the case by summary judgment. 1993 WL 155503, *3-4, 1993 U.S.Dist. Lexis 6141, *11-13, 1993 WL 239051, *5-9,1993 U.S.Dist. Lexis 8744, *19-28,1994 WL 30984, *4-5,1994 U.S.Dist. Lexis 751, *11. itas says that this was error. Who cares? itas lost under the Lanham Act. The jury therefore must have believed that iit did not make any false statement likely to cause confusion among consumers. See 15 U.S.C. § 1125(a)(1). (Though commercial defamation can violate the Lanham Act, see § 1125(a)(1)(B), we reiterate the point that iit has not asked for a new trial on account of inconsistent verdicts.) What would rephrasing the claims under the state laws have added? Throughout its lengthy briefs, itas never tackled this question. At oral argument itas argued that it is a little easier to prevail under the Illinois Consumer Fraud Act than under the Lanham Act, but it did not supply details. The place for such arguments is in the briefs; it is too late to advance them for the first time on one’s feet, after the other side’s submissions are in. We therefore have nothing to say about the question whether the district court should have allowed claims based on these three statutes to go to the jury. This analysis assumes, however, that the verdict on the Lanham Act stands, itas wants it set aside for two reasons. First, it believes that the district judge erred in preventing Toby Katz from testifying. Second, itas contends that the judge should have permitted Marilyn Ziemke to relate what potential customers told her on the telephone. Katz appeared in itas’s witness list in the final pretrial order. According to itas’s list, Katz was one of its employees or agents, itas did not provide Katz’s telephone number or address, and iit did not attempt to contact her to learn what she might say. (The telephone number was in an exhibit.) Come the trial, iit asked the judge to forbid Katz from testifying because the omission of her address violated N.D.Ill. Rule 5.00. The judge agreed and kept Katz off the stand. 1994 WL 30984, *3, 1994 U.S.Dist. Lexis 751, *7. This mechanical application of Local Rule 5.00 is untenable under Grove Fresh Distributors, Inc. v. New England Apple Products Co., 969 F.2d 552, 558-60 (7th Cir.1992). Like many of our cases, Grove Fresh Distributors invokes the principle that a sanction should be proportional to the wrong. See also, e.g., Okaw Drainage District v. National Distillers & Chemical Corp., 882 F.2d 1241, 1248 (7th Cir.1989). Omission of an unimportant detail, like a misspelling of the witness’s name, should not lead to the exclusion of vital evidence. The district court observed that Grove Fresh Distributors involved a rebuttal witness, while Katz was to testify in itas’s case-in-chief. True, but irrelevant. The principle of proportionality is what matters, and it is applicable to all witnesses — and all sanctions, iit had an opportunity to take Katz’s deposition but declined; it is therefore difficult to see why Katz should have been prevented from testifying. Although iit observes that itas’s incorrect assertion that Katz was still itas’s employee dissuaded iit’s lawyers from calling her phone number, the foregone opportunity to take her deposition cannot be so easily explained away. *1260 itas believes that an error in preventing a witness from testifying entitles it to a new trial. Yet there is a harmless-error rule to reckon with. Whether the error was harmless depends on what Katz would have added to the information placed before the jury, itas assures us that Katz’s testimony was vital. Apparently itas expects us to take that representation on faith, because itas did not make an offer of proof. Too bad for itas; the omission is fatal. “Error may not be predicated upon a ruling which admits or excludes evidence unless a substantial right of the party is affected, and ... [i]n case the ruling is one excluding evidence, the substance of the evidence was made known to the court by offer or was apparent from the context”. Fed.R.Evid. 103(a)(2). Beyond itas’s assertion that Katz would have supplied “critical” testimony relevant to damages, we know nothing and therefore are in no position to conclude that the exclusion affected a substantial right. The party aggrieved by exclusion must supply details; itas has volunteered only rhetoric. Ziemke’s proposed testimony is less of a mystery; itas made a formal offer outlining the excluded testimony. Ziemke took many of the calls made to itas’s telephone number and wanted to testify that after iit went into business she began to receive inquiries such as: “They’re offering the same thing you’re offering. What’s the big deal?” and “Doesn’t iit have the same free bar mitzvah?” Ziemke added: “Practically every phone call that itas got in which people were comparing [itas] to iit, every one of those people asked “what’s the big deal? They have the same thing you have.’” The district judge excluded this testimony as hearsay, a step that she conceded in post-trial proceedings had been a blunder. 1994 WL 30984, *2,1994 U.S.Dist. Lexis 751, *4. itas offered this testimony not for its truth — to show that iit was offering the same “free” package as itas — but to show that customers were confused. For that purpose it was not hearsay. Fed.R.Evid. 801(c). Evidence about what customers say to vendors is a staple of trademark and business tort litigation. International Kennel Club of Chicago v. Mighty Star, Inc., 846 F.2d 1079, 1089-90 (7th Cir.1988). Nonetheless, the district judge denied itas’s post-trial motion, explaining that the inquiries that Ziemke would have related at trial did not show consumer confusion. Instead, the judge believed, they showed customers making inquiries to obtain information. Informed by hindsight, we think that the evidence should have been admitted. The line the district judge drew is difficult to maintain. Perhaps the principal inference from the inquiries is that competition had begun. No longer the sole packager of bar mitzvah tours of Israel, itas started to receive inquiries by customers asking about the differences between available options. On this understanding, there is no problem under the Lanham Act. There is only competition, to be encouraged. Another possible inference, however, is that some of these customers were calling because they had been misled by the implication of the iit ads that the two firms’ “free” packages were identical. Ziemke explained the difference to those who called; but some travelers would have signed with iit without calling. Drawing inferences from ambiguous events is the province of the trier of fact, here a jury. That said, however, we do not think that the error was sufficiently grave to require a new trial. The jury was floating in evidence about iit’s representations. Jurors knew the difference between itas’s package and iit’s and were well able to determine whether consumers would have been confused. Ziemke’s testimony would have added evidence in the lists on one side, but it was so much to be expected — many calls of the kind she described in the offer of proof were inevitable even if iit’s ads had included elaborate and complete narrations — that the exclusion does not require the entire trial to be redone. Not when it is so hard to see what damages the Lanham Act (or for that matter the state consumer-protection laws) could have added to the award itas received for defamation. Attorneys’ fees maybe, but they are hardly automatic, itas thinks that it could have collected extra money for corrective advertisements, but this misunderstands the nature of such claims. See Zazú Designs, 979 F.2d at 506-07. Altogether, the probable effect on the award was sufficiently *1261 small, and the probable effect on the outcome sufficiently problematic, that the error was harmless. Only one issue remains: costs. The district judge ordered each side to bear its own. itas believes that this was an abuse of discretion because it prevailed. So it did, but only in part. It lost outright against Larry Ritter. Marlene Ritter won in advance of trial. The Ritters beat back itas’s demands and therefore have a better claim for costs— although the Ritters lost their counterclaims. itas peppered the judge with woebegone theories, such as the antitrust claim, and pointless ones, such as the demands under the state consumer-deception laws. These took up time and iit’s money without producing anything in return. Under the circumstances, declining to shift costs was a wise decision, certainly not an abuse of discretion. Landau & Cleary, Ltd. v. Hribar Trucking, Inc., 807 F.2d 91, 94 (7th Cir.1986). Affirmed.