What is a "preferred provider organization" (PPO)? Our litigants, two intermediaries between hospitals and insurers, disagree about the answer. Plaintiff, which we call First Health, believes that it is misleading, and actionable under &#167 43(a)(1)(B) of the Lanham Act, 15 U.S.C. &#167 1125(a)(1)(B), to apply the label "PPO" or anything similar (such as "preferred provider network") to any business structure other than one in which the insurer compels its clients to use the "preferred" suppliers of medical care. Defendant, which we call UP&UP, labels First Health's business model a "directed PPO," to distinguish it from the "non-directed PPO" or "silent PPO" system that it manages. UP&UP negotiates price reductions with hospitals (often in exchange for a capital contribution as well as a promise of extra business) and signs up insurers at these favorable prices. Insurers are supposed to offer their clients some incentive to use the participating hospitals, for example by reducing copayments, but are not obliged to direct the insureds exclusively to the hospitals participating in the plan. Hospitals typically contract with First Health for a 40% discount from list price, while they allow UP&UP only 20% off, even though UP&UP offers direct financial support that First Health does not. Although the discrepancy in discount rates implies that hospitals well understand the difference in business-generation potential between the systems, First Health insists that it is deceptive for UP&UP to refer to its business as a form of "PPO." First Health also contends that UP&UP hoodwinked hospitals by promising that insurers would induce patients to use participating hospitals.

After the district court granted summary judgment to UP&UP under &#167 43(a)(1)(B) and some related theories, 95 F. Supp. 2d 845 (N.D. Ill. 2000); 2000 U.S. Dist. Lexis 5964 (N.D. Ill. Apr. 10, 2000), but held that one claim (for trademark infringement) requires additional factual development, the parties entered into an agreement under which First Health dismissed its trademark claim and the district court entered a judgment. The parties' agreement, which the district court incorporated into its judgment, provides that First Health will be free to reinstate the trademark claim, but only if we reverse at least in part with respect to the involuntarily dismissed theories. Both parties then briefed the appeal as if it had been taken from a judgment terminating the whole case. They did not mention the reservation of right to reinstate a claim, or address its jurisdictional significance, until this court noted the problem on its own and required the parties to file supplemental memoranda before argument.

A series of decisions, including ITOFCA, Inc. v. MegaTrans Logistics, Inc., 235 F.3d 360 (7th Cir. 2000); JTC Petroleum Co. v. Piasa Motor Fuels, Inc., 190 F.3d 775 (7th Cir. 1999); and Horwitz v. Alloy Automotive Co., 957 F.2d 1431 (7th Cir. 1992), show that the parties' strategy jeopardizes appellate jurisdiction. All of these cases hold that the dismissal of one claim or theory without prejudice, with a right to reactivate that claim after an appeal on the remaining theories, makes the judgment non-final. The parties' dispute has not been fully resolved, and the remaining elements are apt to come back on a second appeal. Sometimes a partial final judgment with respect to particular claims is authorized by Fed. R. Civ. P. 54(b), but when the requirements of that rule are not satisfied, the parties and district judge are not at liberty to home-brew their own approach to obtaining appellate review while portions of the case remain unresolved. Neither side suggested that Rule 54(b) could have been used in this case; there is too much overlap between the trademark-infringement and false-advertising theories. Another method of obtaining interlocutory appeal, the certification of an order under 28 U.S.C. &#167 1292(b), was tried by the district court, but a motions panel declined to accept the appeal. It is especially inappropriate for the parties and district judge to try a back-door approach to interlocutory review after we have turned them away at the front door.

Nonetheless, the parties insist, the right to reinstate the dismissed theories here differs from the stratagems disapproved in ITOFCA, JTC Petroleum, and Horwitz, because reinstatement is contingent on a remand of at least one theory. If no appeal had been taken, then the litigation would have been over; in this sense the judgment must be final, the parties insist. Moreover, if we affirm the case stays over, and there can be no succession of appeals. This shows that First Health has taken a larger gamble than the appellants did in our prior cases, but does the difference matter to jurisdiction? Sometimes it does. Suppose, for example, that a plaintiff is unable to obtain discovery on which its case depends. Rather than go through the formality of a trial, the parties may stipulate that the defendant would prevail on the existing record. Then the court would enter a judgment for the defendant based on the stipulation, allowing the plaintiff to appeal. Decisions such as Martin v. Franklin Capital Corp., 251 F.3d 1284, 1288-89 (10th Cir. 2001), hold that the judgment is final under these circumstances, even though a reversal on the discovery matter would reactivate the main dispute and likely lead to another appeal down the road. Cf. Downey v. State Farm Fire & Casualty Co., 266 F.3d 675, ___ _ ___ (7th Cir.2001).

To decide whether we have appellate jurisdiction given the events as they stood in the district court, we would have to decide whether the principle that allows a dispositive issue to come up, when the plaintiff is willing to stake the entire case on its resolution, extends to multiple claims for relief. But events are no longer what they were. First Health offered in its memorandum addressing jurisdiction to dismiss its trademark claim, if we first determined that the existing judgment is non-final. At oral argument the court made it clear that First Health would not be given that opportunity; the court is not willing to make a decision about appellate jurisdiction only to have the effect of that decision manipulated after the event by the parties. It would become an advisory opinion. There are only two options, we informed counsel: if the judgment is final, we resolve the appeal on the merits; if the judgment is not final, we dismiss the appeal, which stays dismissed until the litigation is wrapped up in the district court. At that point First Health elected to dismiss its trademark claims unconditionally, so that they cannot be reinstated no matter what happens on this appeal. JTC Petroleum held that such an election, by averting any possibility of a future appeal on a different claim, permits the appeal to proceed. So on the authority of JTC Petroleum we shall resolve this appeal, without deciding whether the appeal would have been dismissed had First Health retained its right to reinstate an additional claim following any remand.

First Health's claims arise out of UP&UP's description of its system as creating a "preferred" provider organization or network. Applying the PPO label to a system that lacks a list of hospitals that the insureds must use is misleading, according to First Health--and, although no hospital, insurer, or patient has joined this suit, First Health believes that its competitive losses to UP&UP enable it to obtain relief under &#167 43(a)(1)(B) of the Lanham Act, as a business harmed by UP&UP's false advertising. Section 43(a)(1) provides in part:

Any person who, on or in connection with any goods or services, or any container for goods, uses in commerce any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact, which . . . (B) in commercial advertising or promotion, misrepresents the nature, characteristics, qualities, or geographic origin of his or her or another person's goods, services, or commercial activities, shall be liable in a civil action by any person who believes that he or she is or is likely to be damaged by such act.

First Health contends that use of the term "PPO" in connection with UP&UP's system is a "false or misleading description of fact" that appears "in commercial advertising or promotion" and "misrepresents the nature . . . of [UP&UP's] services". Moreover, First Health believes, UP&UP has further deceived hospitals by promising that insurers will provide additional business by inducing patients to use their services. This is false, or at least misleading, First Health submits, because UP&UP does not tell the hospitals that it views the discounts themselves as inducements to use the hospitals, even if the financial benefits of these inducements are not passed on to patients (as by reductions in copayments), and even if patients do not learn about the discounts until they have used a particular hospital for the first time. The prospect of repeat business after a given patient knows about the discount is not enough to satisfy UP&UP's contractual promises to hospitals, First Health insists. Once again it claims injury as a competitor, this time redressable under the common law or state consumer-fraud statutes.

One preliminary question is how the negotiations between UP&UP and particular hospitals--negotiations handled in private, among business executives and lawyers-- could be called "commercial advertising or promotion," an essential ingredient of any claim under &#167 43(a)(1)(B). The district court treated "commercial advertising or promotion" as a synonym for all commercial speech that the first amendment allows the federal government to regulate. See 1998 U.S. Dist. Lexis 3183 (N.D. Ill. Mar. 12, 1998). In reaching this conclusion it relied on a line of decisions by other district courts that follow a multi-factor approach for determining the scope of &#167 43(a). See, e.g., Gordon & Breach Science Publishers S.A. v. American Institute of Physics, 859 F. Supp. 1521 (S.D.N.Y. 1994) (the originator of this four-part test). We have serious doubts about the wisdom of displacing the statutory text in favor of a judicial rewrite with no roots in the language Congress enacted. Treating &#167 43(a)(1) as filling all ground that the Constitution allows to the legislature is especially implausible, for when the Lanham Act was adopted there were no constitutional limits on the regulation of commercial speech. See Valentine v. Chrestensen, 316 U.S. 52 (1942), overruled by Virginia Board of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748 (1976). Any plan to occupy all of the constitutionally permitted space would have used the phrase "commercial speech," rather than "advertising or promotion" in &#167 43(a)(1)(B), and the variety of other scope rules found elsewhere in the statute. The language and structure of &#167 43(a)(1)(B) do not suggest a line between commercial and political speech, with the former completely covered and the latter not. They suggest a line between varieties of commercial speech.

Advertising is a form of promotion to anonymous recipients, as distinguished from face-to-face communication. In normal usage, an advertisement read by millions (or even thousands in a trade magazine) is advertising, while a person- to-person pitch by an account executive is not. So we have held in a series of disputes that require a definition of "advertising injury" under insurance policies. See, e.g., Zurich Insurance Co. v. Amcor Sunclipse North America, 241 F.3d 605 (7th Cir. 2001); Western States Insurance Co. v. Wisconsin Wholesale Tire, Inc., 184 F.3d 699 (7th Cir. 1999). Zurich put it this way: "Advertising is a subset of persuasion and refers to dissemination of prefabricated promotional material." 241 F.3d at 607. Now this is as a matter of ordinary usage; perhaps "commercial advertising or promotion" in &#167 43(a)(1)(B) is a term of art. Yet First Health has not offered any material from the time of the Lanham Act's adoption (either legislative history or other examples of usage in the legal community) implying that the statutory phrase "commercial advertising or promotion" means anything other than what a user of normal English would assume. To repeat an illustration from Zurich: First Health's lawyers were engaged in persuasion when they filed their briefs and presented oral argument to this court, but they were not engaged in "commercial advertising or promotion" in any useful sense of that phrase. Their written and oral advocacy did not become "commercial advertising or promotion" just because the first amendment allows this court to regulate the practice of law (as by sanctioning frivolous arguments or misstatements about the contents of the record). We held in Zurich that statements by a firm's sales force could not be called "advertising"; likewise, it is hard to see how statements of UP&UP's executives and lawyers, made over a conference table in an effort to negotiate a contract, or included in the contract's language, could be called "commercial advertising or promotion".

Because the district court treated "commercial advertising or promotion" as synonymous with all commercial speech, it did not conduct the inquiry necessary to determine whether any of UP&UP's sales and contracting endeavors entailed promotional material disseminated to anonymous recipients. We shall therefore assume that &#167 43(a)(1)(B) applies to some of UP&UP's promotional materials. Still, this does not much help First Health, because like the district court we conclude that First Health has not established that any of UP&UP's statements was false or even misleading. What First Health wants--a judicial decree that only businesses using a single model can employ the term "PPO" or the phrase "preferred provider network"--is nothing less than an order establishing property rights in the language. Words are not born with meanings. They acquire meaning with use, and as use changes so does meaning. It may well be that all of the initial PPOs were "directed." But for many years not only UP&UP but also other similar intermediaries have been offering what they call silent or non-directed PPOs. Both have become standard usages. To see this, one need go no farther than the proceedings of the Office of Personnel Management, which conducted an inquiry into the kinds of medical plans that would be offered to federal employees as fringe benefits. One of the ensuing reports, Office of the Inspector General, Office of Personnel Management, Rep. No. 99-00-97-054, Report on the Use of Silent PPOs in the Federal Employees Health Benefits Program 20-23 (1998), not only uses the phrase "silent PPO" exactly as UP&UP does but also concludes that the arrangement can produce benefits for consumers, just as directed PPOs do. Legislative committees have used the term just as UP&UP and OPM do. E.g., S. Rep. No. 257, 105th Cong., 2d Sess. 3 (1998). We cannot imagine a court issuing an injunction that would prohibit UP&UP from using "PPO" or "silent PPO" in the same way units of the national government use that phrase. First Health should have turned to an advertising agency, not to a court, in search of a response to its business rivals. If directed PPOs are "better" than silent PPOs, then First Health should be able to explain this to potential customers--for recall that both First Health and UP&UP deal with business executives and lawyers, not with patients who may miss the distinction conveyed by "directed" versus "non-directed."

No business is entitled to a trial after which judge and jury will determine how language ought to be used, as if usage were a question of law or logic. It is enough to guard against misleading expressions that play on how language is used. Perhaps First Health might have some room for maneuver if it could show that the linguistic community of hospital executives and lawyers differs from that of Congress and the OPM--though even so it is doubtful that the Lanham Act may be employed to ensure that language, used in accord with normal rules of grammar and diction, cannot be misinterpreted. See Mead Johnson & Co. v. Abbott Laboratories, 201 F.3d 883 (7th Cir. 2000). But this is a subject we need not explore, because First Health lacks evidence that hospitals were deceived by what UP&UP said. It offered "expert" evidence that hospitals were bound to be confused by the difference between directed and silent PPOs, but why ask an "expert" when the hospitals can and do speak for themselves? That First Health receives on average twice the discount offered to UP&UP shows that hospitals understand the difference between the two plans.

Asked at oral argument to identify its best evidence that hospitals were deceived, First Health identified the letters some hospitals sent to UP&UP terminating their contracts. We have reviewed these letters and found no evidence of confusion at the time the contracts were signed. None of the hospitals professes a failure to appreciate the difference between directed and silent PPOs. Some of the letters show a good understanding of the difference, but disappointment with the economic consequences of a silent PPO arrangement. (E.g., "In response to the growing concerns and identification of 'Silent-PPOs', Columbia Healthcare Corporation, South Texas Division is amending all managed care contracts to include appropriate language.") Other letters express unhappiness about the performance of insurers with which UP&UP contracted to supply patients to the hospitals. For example, the termination letter from Upstate Carolina Medical Center complained that some insurers, instead of providing patients with incentives to use the Medical Center, were penalizing them for doing so and refusing to pay in full on the ground that the patients had gone "out of network." None of this helps First Health, let alone establishes enough confusion to go forward. Some confusion and misunderstanding are inevitable in any business relation. The Lanham Act deals only with confusion that exceeds the norm in the human condition, see Reed-Union Corp. v. Turtle Wax, Inc., 77 F.3d 909 (7th Cir. 1996), and First Health has not made any effort to measure the normal level of misunderstanding in this business.

Lack of actual confusion is irrelevant, First Health insists, because (a) UP&UP's statements were "actually false" rather than just misleading, and (b) it seeks prospective relief in addition to damages. For reasons we have already given the use of "PPO" by a non-directed network is not "actually false." But First Health's argument also is incorrect as a matter of law. See B. Sanfield, Inc. v. Finlay Fine Jewelry Corp., 258 F.3d 578 (7th Cir. 2001). Section 43(a) (1)(B) offers relief only to one who "is or is likely to be damaged by" the misrepresentation. Proof of likely confusion is essential to show injury-- actual or potential--without which there is not even a case or controversy.

The district judge concluded, and we agree, that as far as this record shows First Health never lost the business of a single hospital because UP&UP's use of "PPO" confused that hospital. First Health did lose to UP&UP the business of the Government Employees Hospital Association, Inc. (GEHA) network, but it does not attribute this to any confusion on GEHA's part. First Health concedes that GEHA understood completely the difference between First Health's and UP&UP's business models. Instead First Health says that it lost GEHA's business because UP&UP had grown, as a result of its misuse of "PPO," to a size where GEHA found it a useful business partner. GEHA then called for bids on terms that were better suited to UP&UP than to First Health (which declined to submit a bid on the terms GEHA prescribed). That theory of liability is divorced from any effort to prevent misleading advertising; it amounts to an effort to deprive hospitals, insurers, and consumers of the efficiencies made available by economies of scale.

No rule of either state or federal law entitles a competitor to damages from its rival because of scattered contractual disputes between the rival and its customers (for example, the complaint made by Columbia Healthcare that UP&UP had not done enough to ensure that insurers would steer patients to its facilities). If UP&UP does not keep its promises, it will lose business and may have to pay damages. Suits by rivals are not necessary for deterrence but could be used to harass a competitor and stifle the competitive process. More, it is integral to First Health's theory that problems such as those at Columbia Healthcare are uncommon; UP&UP has been growing, not shrinking as trading partners bail out. (Growth is, after all, the method by which UP&UP snared GEHA's business.) This entire suit strikes us as one designed to hamstring a competitor whose success reflects its ability to please its trading partners. If the vocabulary of a business such as UP&UP is to be revised, that is a job for legislatures and regulatory agencies, rather than for judges and juries in suits under the Lanham Act and state consumer-fraud statutes.

First Health's remaining contentions, and theories of liability, do not require separate discussion.