329 F.3d 923 | 7th Cir. | 2003
Stephen L. BUCKLEW, Plaintiff-Appellee, Cross-Appellant,
v.
HAWKINS, ASH, BAPTIE & CO., LLP, and HAB, Inc., Defendants-Appellants, Cross-Appellees.
No. 02-2244.
No. 02-2299.
United States Court of Appeals, Seventh Circuit.
Argued April 4, 2003.
Decided May 27, 2003.
COPYRIGHT MATERIAL OMITTED Harry E. Van Camp (argued), Hayes, Van Camp & Schwartz, WI, for plaintiff-appellee.
Donald W. Rupert (argued), Mayer, Brown, Rowe & Maw, Chicago, IL, for defendnats-appellants.
Before POSNER, EASTERBROOK, and ROVNER, Circuit Judges.
POSNER, Circuit Judge.
Stephen Bucklew brought this suit in a federal district court in Wisconsin against two affiliated companies that we'll call "HAB." The suit charges copyright infringement, fraud, conversion, and violation of RICO. The district judge dismissed all but the copyright claim, which was tried to a jury, resulting in a verdict for Bucklew of $660,000, which the judge cut down to $395,000. HAB appeals; Bucklew cross-appeals, challenging the dismissal of his noncopyright claims.
Local housing authorities that want grants from the federal department of Housing and Urban Development have to complete forms prescribed by HUD. The forms require not only specific basic data such as salaries and other categories of proposed expenditure by the grant applicant but also simple arithmetical transformations of the data, such as adding the numbers in particular cells in the table of basic data. Bucklew developed and copyrighted software, intended to be used in conjunction with standard spreadsheet applications such as Lotus 1-2-3 and Excel, for doing these transformations and displaying them in tables. The essential transformation involves entering the basic data in an electronic copy of the HUD-prescribed form and applying to those data an algorithm that picks out the relevant cells and performs the relevant operation (namely addition) on them and displays the results in tabular form. Standard spreadsheet applications make this transformation relatively easy to program because they include functions such as DSUM, a simple command for adding up the numbers in the cells identified by a criterion specified by the programmer and displaying the results of the addition. Bucklew does not claim copyright in either the spreadsheet applications or DSUM. But there is more, though not a great deal more, to his product than these programs. Decisions have to be made regarding choice and size of font, the size of cells and columns, whether and where to use color, the wording of labels and headings (other than those prescribed by the HUD forms), and whether to use boldface or italics for column headings. These decisions were made by Bucklew. The ones we have named all involve the appearance of the forms, but software that is read only by the computer and not by its human user is also copyrightable, and Bucklew does claim copyright in the "construction" as well as display of his forms; but it is unexplained what he means by this or whether it has been infringed.
The trial focused on one of the four forms (as the parties refer to the conversion of a prescribed HUD form into an electronic form that computes and displays the arithmetic manipulations that HUD requires) copyrighted by Bucklew that he claimed were copied by HAB. This form is a transformation of HUD form 52566, which requires salary data; the other three forms require data for other categories of expenditure. HAB grudgingly concedes that the evidence compels an inference that it copied Bucklew's form 52566 to create its own form 52566. Some similarities between a copyrighted work and a work alleged to infringe it are consistent with an inference of independent creation, and in that case evidence that the alleged infringer had access to (that is, saw or at least could have seen) the copyrighted work is indispensable. Ty, Inc. v. GMA Accessories, Inc., 132 F.3d 1167, 1169-70 (7th Cir.1997). But when the similarities concern details of such an arbitrary character that the probability that the infringer had duplicated them independently is remote, an inference of copying may be drawn without any additional evidence. Id.; Selle v. Gibb, 741 F.2d 896, 901, 903-05 (7th Cir.1984); Gaste v. Kaiserman, 863 F.2d 1061, 1067-68 (2d Cir.1988); Bouchat v. Baltimore Ravens, Inc., 241 F.3d 350, 355-56 (4th Cir.2001). These cases say that access can be inferred from a sufficiently striking similarity between the two works, and that is true; but as the Fifth Circuit suggested in Ferguson v. National Broadcasting Co., 584 F.2d 111, 113 (5th Cir.1978), it is more straightforward to say that in some cases proof of access isn't required.
It is in order to avoid having to prove access that mapmakers will sometimes include a fictitious geographical feature in their maps; if that feature (what is called in the trade a "copyright trap") is duplicated in someone else's map, the inference of copying is compelling. Rockford Map Publishers, Inc. v. Directory Service Co. of Colorado, Inc., 768 F.2d 145, 147 (7th Cir.1985); Melville B. Nimmer & David Nimmer, Nimmer on Copyright, vol. 4, § 13.03[C], pp. 13-75 to 13-77 (2002); see also General Drafting Co. v. Andrews, 37 F.2d 54, 56 (2d Cir.1930) (errors as distinct from deliberately set traps). And that is the case with regard to form 52566. Bucklew's version contains an arbitrary pattern of boldfacing of cells; HAB's duplicates it exactly. Bucklew's form also contains an "output range," an intermediate table that he carried over from a previous spreadsheet program but that has no function in his current program—or in HAB's. Had HAB written its program from scratch, it would have had no reason to include an output range—yet it did. And in another intermediate table, one that contains the identical data as the form 52566 itself (the starting point for the transformation), HAB's version uses the identical headings as Bucklew's corresponding intermediate table, rather than the headings in the form 52566 with which HAB had begun. (For unexplained reasons, that form had different headings from the form from which Bucklew had begun.) And HAB's program like Bucklew's used a font (Swiss font) that was not available in the version of the Windows operating system that HAB used to develop its software. This is further evidence that rather than creating its own form using the software available to it on Windows, HAB had copied Bucklew's form.
The evidence of copying with regard to the other three forms is scanty. Indeed, they were barely mentioned at trial. HAB complains that they were smuggled into the trial and that the jury should not have been permitted to consider them. Before trial the district judge, in conformity with his established practice (though we cannot find it stated in writing in the rules of his court or anywhere else), explained to the parties that any objections to exhibits, other than objections based on relevance, had to be made before trial and any exhibit not objected to before trial would become a part of the trial record. The exception for relevance was presumably intended for the situation in which, because of the course the trial might take, a document that had seemed relevant to the issues in the case before trial no longer was. HAB did not object before trial to the introduction into evidence of any of the four forms of Bucklew's that were alleged to be infringed. But it argues that it assumed that the forms would be discussed at trial rather than just being handed to the jury, along with the other exhibits that had not been ruled inadmissible, when the jury retired to deliberate.
There was no basis for the assumption. Admissible documents are evidence, just like testimony. They do not require sponsorship. The procedure followed by the district judge is common and sensible. It saves trial time. The Federal Rules of Civil Procedure encourage judges to resolve evidentiary disputes so far as possible before trial in order to shorten trials and minimize interruptions, Fed.R.Civ.P. 16, 26(a)(3)—an especially important consideration in jury trials. Evidently Bucklew was content, having established that HAB had copied his form 52566 to make its form 52566, to let the jury make up its own mind about the other forms. He testified that he believed his other three forms had been copied as well, and his expert's reports, which though hearsay were admitted into evidence without objection, said the same thing though without elaboration. Had HAB wanted to challenge the inference of copying with evidence of its own, there was nothing to stop it from doing so. Maybe it held back because it thought that the jury would be unlikely to find infringement of the other three forms when the plaintiff had presented so little evidence about them and there was no evidence that HAB had had access to Bucklew's versions of those forms, though this is irrelevant because Bucklew's claim is that HAB's versions of those forms were copied from Bucklew's form 52566, to which HAB did have access.
In this court, HAB's principal argument about the admissibility of the three forms is that they confused the jury. Just as a document that seemed admissible before trial might be objected to at trial as lacking relevance, so it might be objected to as confusing because it had not been the subject of testimony and without testimony could not be comprehended by a lay person. But there was testimony about the forms—elicited indeed by HAB itself, on cross-examination—enough to justify the judge in overruling the objection.
HAB remains free to argue that there was a failure of proof with respect to the three forms. Bucklew, however, testified without contradiction to certain similarities that indicated copying (though, to repeat, from Bucklew's form 52566, not from Bucklew's versions of the three other forms), notably a missing vertical line where a line would otherwise be expected, explicable only as a trace of the deletion of the unneeded output range that HAB had copied from Bucklew's form. (A cell to the left of the deleted area had a box around it; apparently when an HAB programmer deleted material to the right of that cell, he stripped the box around the cell of its right-hand side vertical border.) What is true is that there are more differences between HAB's version of the other three forms and form 52566, and this may be relevant to infringement, as we are about to see; but it does not rebut an inference of copying. To return to the map example, the presence of a fictitious geographical feature in the alleged infringer's map might be conclusive proof of copying even though the copier had changed the colors or other features of the map, so that at a glance the two maps appeared quite different.
With respect to all four forms HAB argues that even if there was copying there was no infringement. It appeals primarily to the copyright doctrines of merger and scènes à faire. The first of these confusing labels refers to the situation in which there is only one feasible way of expressing an idea, so that if the expression were copyrightable it would mean that the idea was copyrightable, and ideas are not copyrightable. E.g., Ets-Hokin v. Skyy Spirits, Inc., 225 F.3d 1068, 1082 (9th Cir.2000); Kregos v. Associated Press, 937 F.2d 700, 705 (2d Cir.1991); Kern River Gas Transmission Co. v. Coastal Corp., 899 F.2d 1458, 1459, 1463-64 (5th Cir.1990). The standard citation remains Baker v. Selden, 101 U.S. 99, 25 L.Ed. 841 (1879). Selden had published (and copy-righted) a book describing a bookkeeping system that he had invented, and he illustrated the book with blank bookkeeping forms. Baker copied the forms, rearranging columns and using different headings, and sold them to people who wanted to use Selden's system. This was held not to be copyright infringement, even though Baker had copied part of a copyrighted work, since otherwise Selden would have had a monopoly over his bookkeeping system (which was an idea, and hence not copyrightable) that he could have exploited by insisting that anyone wanting to use the system buy the forms necessary for using it from him. If Bucklew were claiming copyright in the tabular presentation of the summary data required by HUD, this case would be governed by Baker v. Selden. But he is not. He is claiming copyright in tables configured in an optional way, tables that are the product of format choices that are not unavoidable, for which indeed there were an immense number of alternative combinations any one of which HAB was free to use in lieu of Bucklew's. See, e.g., Mason v. Montgomery Data, Inc., 967 F.2d 135, 138-41 (5th Cir.1992); Kregos v. Associated Press, supra, 937 F.2d at 706-07.
The fact that Bucklew's formatting choices do not reflect a high degree of originality is irrelevant. When as in this case a work in which copyright is claimed is based on work in the public domain, the only "originality" required for the new work to be copyrightable (the very term is a misnomer) is enough expressive variation from public-domain or other existing works to enable the new work to be readily distinguished from its predecessors. Alfred Bell & Co. v. Catalda Fine Arts, 191 F.2d 99, 102-03 (2d Cir.1951). "Originality in this context means little more than a prohibition of actual copying." Id. at 103 (internal quotation marks deleted); Three Boys Music Corp. v. Bolton, 212 F.3d 477, 489 (9th Cir.2000). That undemanding requirement is satisfied in this case; any more demanding requirement would be burdensome to enforce and would involve judges in making aesthetic judgments, which few judges are competent to make.
The doctrine of scènes à faire (another confusing label, literally "scenes for action," which the Oxford English Dictionary tells us is a theatrical term meaning "the most important scene in a play or opera, made inevitable by the action which leads up to it" — which is not the legal doctrine at all) teaches, sensibly enough, that a copyright owner can't prove infringement by pointing to features of his work that are found in the defendant's work as well but that are so rudimentary, commonplace, standard, or unavoidable that they do not serve to distinguish one work within a class of works from another. Reed-Union Corp. v. Turtle Wax, Inc., 77 F.3d 909, 913-14 (7th Cir.1996); Atari, Inc. v. North American Philips Consumer Electronics Corp., 672 F.2d 607, 616-17 (7th Cir.1982); Williams v. Crichton, 84 F.3d 581, 587-89 (2d Cir.1996). Every expressive work can be decomposed into elements not themselves copyrightable— the cars in a car chase, the kiss in a love scene, the dive bombers in a movie about Pearl Harbor, or for that matter the letters of the alphabet in any written work. The presence of such elements obviously does not forfeit copyright protection of the work as a whole, but infringement cannot be found on the basis of such elements alone; it is the combination of elements, or particular novel twists given to them, that supply the minimal originality required for copyright protection.
A more fundamental limitation on the scope of copyright infringement is, as we have already noted, that copyright protects only expression and not what is expressed, the "idea," which in the case of fiction is more likely to be a plot or situation than, as in Baker v. Selden, an algorithm or other abstraction. Suppose one wanted to prove that the Tom Hanks-Meg Ryan movie You've Got Mail infringed the Jimmy Stewart-Margaret Sullavan movie The Shop Around the Corner. You could point to the fact that in both movies the hero falls in love with the heroine on the basis of correspondence (letters in the earlier movie, email in the later) without having met her and without realizing that it is someone he knows—and dislikes (and the feeling is reciprocated)—though everything comes right in the end. But it would not be evidence of infringement that both movies were love stories, or that they were comedies, or that the lead actors were a man and a woman. These are generic characteristics, obtained by abstracting the commonest features of the specific work; they do not indicate uniqueness. They are at the opposite extreme from the arbitrary details, such as the pattern of boldfacing in Bucklew's form 52566, that contribute such originality as a new expressive work may have. The format choices that Bucklew made were not generic. It is not as if everyone who writes programs of this sort uses Swiss font or displays an output range or uses a particular pattern of boldfacing.
Nor is it important that after copying Bucklew's form 52566 and using it as the template for the other three forms, HAB introduced a number of variations in its version of form 52566. Variants that result from tinkering with a copied form are derivative works from that form, and it is a copyright infringement to make or sell a derivative work without a license from the owner of the copyright on the work from which the derivative work is derived. Pickett v. Prince, 207 F.3d 402, 405-407 (7th Cir.2000); Nimmer & Nimmer, supra, vol. 1, § 3.06, pp. 3-34.24 to 3-34.26; Mark A. Lemley, "The Economics of Improvement in Intellectual Property Law," 75 Tex. L.Rev. 989, 992 (1997). In some cases a work, however, though derivative in a literal sense, is so utterly transformed as to bear no traces of the original; and then there is no infringement. Castle Rock Entertainment, Inc. v. Carol Publishing Group, Inc., 150 F.3d 132, 143 n. 9 (2d Cir.1998). Such cases are in fact common, though they are rarely litigated. Suppose one copied a long passage from a copyrighted work and then edited it to produce a paraphrase so loose that it would not be similar enough to the original to constitute an infringement. The fact that the paraphrase had been "derived" in a genetic sense from a copyrighted original would not make it infringing. Pickett v. Prince, supra, 207 F.3d at 407 ("works only loosely connected with some ancestral work claimed to be their original"). Similarly, if the original expression added by the unauthorized preparer of a derivative work is clearly detachable from the original work itself, so that no confusion, or disruption of the copyright owner's plans for the exploitation of his work, would be created by allowing the unauthorized preparer to copyright his original expression, the unauthorized preparer might be allowed to do so, Williams v. Broadus, 60 U.S.P.Q.2d 1051, 1055 (S.D.N.Y.2001); Paul Goldstein, Copyright, vol. 1, § 2.16, p. 2:209 and n. 11 (2d ed. 2003), though this principle may be limited to compilations, where "the infringing portion would be easily severable and the scope of the compilation author's own work ... would be easily ascertainable." Anderson v. Stallone, 11 U.S.P.Q.2d 1161, 1167-69 (C.D.Cal.1989). But HAB does not argue that its alterations of the copied form went so far as to bring it within the scope of either of these qualifications of the general rule concerning derivative works.
The analysis to this point sufficiently establishes that HAB's version of form 52566 infringed Bucklew's version. But that leaves the other three forms, where there are fewer traces of borrowing from Bucklew's form. The principal traces are the captions "input range" and "criteria range," the use of "crit1," "crit2," etc. to designate specific criteria, and the designation of each category of data as a "type." These are good examples of scènes à faire—standard expressions, like language itself, without which the would-be author of an expressive work would be speechless. The only other residue of a borrowed-from-Bucklew feature in the three forms was that telltale missing vertical line—the tipoff that HAB had indeed copied Bucklew's form 52566 in creating its own version of the other three forms and had then deleted the superfluous output range. But a nonexistent line has no expressive content, except possibly in the most advanced modern art. (Cf. Georges Perec, La Disparation (1969)—a 300-page novel written without the letter e.) The only element of an expressive work that is infringed by copying is an expressive element; what remain from the copying in the case of the three forms under consideration either were scènes à faire or weren't expressive elements at all.
So the jury verdict on infringement can stand only with respect to the first of the four allegedly infringing forms. This has obvious implications for the damages award, but before getting to those let us consider HAB's challenges to that award that are independent of any doubts about liability. After the judge reduced the jury's award by granting a remittitur that Bucklew accepted, the award consisted of four items: $100,000 for Bucklew's lost profits; $125,000 for HAB's profits from the infringing forms; $70,000 for a savings in time that HAB obtained as a result of the infringement; and $100,000 for profits obtained by HAB on separate products by virtue of its being able to offer its customers "one-stop shopping," i.e., a complete line of HUD financial software, including the copied forms.
A copyright owner can sue for his losses or for the infringer's profits, but not for the sum of the two amounts. 17 U.S.C. § 504(b); Deltak, Inc. v. Advanced Systems, Inc., 767 F.2d 357, 363 (7th Cir.1985); Taylor v. Meirick, 712 F.2d 1112, 1120 (7th Cir.1983); Hamil America, Inc. v. GFI, 193 F.3d 92, 108 n. 7 (2d Cir.1999); Nimmer & Nimmer, supra, vol. 4, § 14.01[A], pp. 14-5 to 14-6. That would be double counting. The profits that HAB obtained from selling forms copied from Bucklew came at his expense; his loss was HAB's gain. If the infringer is a more efficient producer, his gain may exceed the copyright owner's loss. If he is not more efficient, his gain is likely to be less than the copyright owner's loss because competition will tend to force price down to cost, minimizing the infringer's profits but depriving the copyright owner of the supra-competitive return that he enjoyed before he faced competition from the seller of an exact substitute for his product. The copyright owner is allowed to waive damages (lost profits) and sue for the infringer's gain. Copyright infringement unlike patent infringement is an intentional tort, and by forcing the infringer to disgorge his profit should it exceed the copyright owner's loss the law discourages infringement and encourages the would-be infringer to transact with the copyright owner rather than "steal" the copyrighted work. Taylor v. Meirick, supra, 712 F.2d at 1120. But there is no basis in the law for requiring the infringer to give up more than his gain when it exceeds the copyright owners' loss. Such a requirement would add a punitive as distinct from a restitutionary element to copyright damages, and while the copyright statute does authorize statutory damages unrelated to losses or gains, see 17 U.S.C. § 504(c), these were not sought here and—a point to which we shall return —the statute contains no provision for punitive damages. But, astonishingly, HAB has failed to complain about double counting, and so it has booted away a winning issue.
HAB argues that Bucklew's lost profits were due not to the infringement but simply to the fact that HAB was competing with him. Remember that Bucklew had no copyright in the database programs that he used (Lotus and Excel) or in the spreadsheet formula (DSUM) that transformed the raw data into the data required by HUD. HAB could have designed its own, noninfringing program that would have duplicated the essential—and noncopyrighted —features of Bucklew's. Since it would have been producing a close substitute, it would have taken sales from Bucklew, reducing Bucklew's profits.
The essential distinction at which this argument gestures is between infringer's profits that are due to the infringer's being a more efficient producer than the plaintiff, which the copyright owner can sue for in order to discourage infringement, and infringer's profits that are due to features of his work that do not infringe; those profits belong to him and not the copyright owner. But there is a difference between an infringer's adding expressive features to what he has copied, and a copier who adds no expressive features but argues that since the copied work contains mainly public-domain material, damages should be limited to the perhaps trivial commercial value attributable to the expressive features that the author of that work added to the public-domain materials with which he began. The argued distinction is artificial. The value of an expressive work is the result of the operation of the author's creativity on public-domain materials, ideas, and so forth; partitioning the two sources of the value, the public domain and the author's creative transformation of it, would not be practical. The apportionment that the law permits is not between the creative and the derived features of the plaintiff's work, but between the infringing and the noninfringing features of the defendant's work.
Once the plaintiff proves his losses, or the defendant's profits, from the defendant's sale of an infringing work, the burden shifts to the defendant to apportion the profits or losses between the infringing and noninfringing features of the defendant's work. 17 U.S.C. § 504(b); Ty, Inc. v. Publications Int'l Ltd., 292 F.3d 512, 523-24 (7th Cir.2002), and cases cited there. HAB failed to do that.
The item of damages for time savings is not duplicative of the other damages that were awarded, though it was improper on another ground. The savings are in the time that HAB takes to service its customers, the local housing authorities, and do not show up in the profits that HAB makes from the sale of the infringing products and so the loss that Bucklew suffered from the infringement. But HAB argues compellingly that any savings in time could not have reflected the copyrighted features of Bucklew's forms, because those features— the formatting choices—do not reduce the amount of time that is required to process data, using the spreadsheets with their built-in DSUM function. There is an analogy to the principle of "antitrust injury." Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 486-88, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977); Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 337, 339, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990); Dial A Car, Inc. v. Transportation, Inc., 82 F.3d 484, 486-87 (D.C.Cir.1996). If two competing firms merge in violation of antitrust law, but the merger's effect on the market price from eliminating competition between the merging firms is dominated by its effect in reducing the costs of the firm resulting from the merger below those of the two predecessor firms, and in consequence the firm reduces its price below the old competitive price, the remaining firms in the market will have to lower their price in order to remain competitive and so they are injured by the merger. But the injury is not actionable; it is not "antitrust injury" (that is, injury for which antitrust law provides a remedy) because it is not the sort of injury that antitrust law is concerned with preventing or remedying. Likewise here, HAB may have obtained cost savings by copying Bucklew's form, but insofar as those cost savings reflected not the appropriation of the expressive features of Bucklew's form but the appropriation merely of the idea of using spreadsheet programs with the DSUM function to expedite the preparation of grant applications to HUD, they were not the fruit of conduct for which copyright law is intended to provide a remedy.
What is true, and is the basis of Bucklew's claim of conversion, of which more shortly, is that HAB did not buy Bucklew's form; it borrowed and, without authorization, copied it. The unauthorized copying was an infringement, but the only damages to which it can give rise are the price of the form, not the value of the ideas in the form because the ideas are not protected by copyright law.
There was also a failure of proof with regard to the "one-stop shopping" damages. HAB argues that damages can never be obtained that are based on the profits that the infringer made from the sale of a product that did not infringe the copyright. That is incorrect. Business Trends Analysts, Inc. v. Freedonia Group, Inc., 887 F.2d 399, 404, 407 (2d Cir.1989); Frank Music Corp. v. Metro-Goldwyn-Mayer, Inc., 772 F.2d 505, 516-17 (9th Cir.1985). Remember that the purpose of allowing suit for the infringer's lost profits is to make infringement worthless to the infringer. This will sometimes require tracing those profits into another product, as where it is bundled with the infringing product. Suppose a defendant had copied a copyrighted book verbatim and he then offered to sell copies of the book for nothing to anyone who would pay him $25 for a bookmark that had cost him 10 cents and had a market value of 50 cents. To hold in such a case that the defendant's profits from infringement were zero would approve a formula for evading copyright law. Cf. Jefferson Parish Hospital District No. 2 v. Hyde, 466 U.S. 2, 13 n. 19, 104 S.Ct. 1551, 80 L.Ed.2d 2 (1984); United States v. Microsoft Corp., 253 F.3d 34, 86 (D.C.Cir.2001) (en banc) (per curiam); Howard P. Marvel, "The Resale Price Maintenance Controversy: Beyond The Conventional Wisdom,"63 Antitrust L.J. 59, 63-64 (1994).
But Bucklew's evidence that HAB shifted profits from the infringing forms to a separate, noninfringing financial software package ("HMS for Windows") was too speculative to sustain an award of damages. His theory is that the prospect of one-stop shopping attracted customers who might otherwise have gone to a competitor, and so increased HAB's revenue and presumably its profits. One of HAB's employees testified that the infringing forms would indeed help with the sale of HMS for Windows, but no evidence was presented that would have enabled the market value of this "help" to be gauged. The testimony by an expert witness for the plaintiff that 10 percent of the profits on HAB's sales of HMS for Windows were due to the buyers' being able to buy the infringing forms from HAB had no factual basis whatsoever, and we repeat previous reminders to the bench and bar of this circuit that proof of damages requires— proof. CSC Holdings, Inc. v. Redisi, 309 F.3d 988, 995 (7th Cir.2002); MindGames, Inc. v. Western Publishing Co., 218 F.3d 652, 658 (7th Cir.2000); Harbor House Condominium Ass'n v. Massachusetts Bay Ins. Co., 915 F.2d 316, 318-19 (7th Cir.1990); see also Hillside Enterprises v. Carlisle Corp., 69 F.3d 1410, 1414 (8th Cir.1995).
Besides all these problems with the computation of damages, it will be necessary on remand to apportion damages between the infringing form and what we have determined to be the three noninfringing forms.
The cross-appeal need not detain us for long. Bucklew seeks punitive damages for fraud and conversion under Wisconsin law. He argues that HAB obtained a computer disk containing his copyrighted program by falsely representing that it wanted merely to evaluate the program with a view toward possibly licensing it from Bucklew, and having thus obtained the program by false pretenses then copied it, with the consequences described earlier. We may assume that this is a good claim under Wisconsin law. But Bucklew is not asking to have the disk returned, or for damages equal to the value of the disk plus punitive damages proportioned to that value. The compensatory damages that he seeks for the fraud and conversion are identical to the damages that it seeks for copyright infringement, so that its request for punitive damages is in fact a request for punitive damages for copyright infringement. The copyright statute does not authorize such damages, as we have noted, and the statute's preemption clause forbids states to add sanctions for a wrongful act that is identical to a violation of the statute. See 17 U.S.C. § 301(a); Baltimore Orioles, Inc. v. Major League Baseball Players Ass'n, 805 F.2d 663, 674, 676-77 (7th Cir.1986), and with specific reference to punitive damages Budget Cinema, Inc. v. Watertower Associates, 81 F.3d 729, 733 (7th Cir.1996); Hays v. Sony Corp. of America, 847 F.2d 412, 415 (7th Cir.1988); On Davis v. The Gap, Inc., 246 F.3d 152, 172 (2d Cir.2001); Nimmer & Nimmer, supra, vol. 4, § 14.02[B], p. 14-28.
The RICO claim is another loser. RICO provides a remedy against conducting the activities of an enterprise through a pattern of racketeering activity, and the definition of racketeering includes criminal copyright infringement. But apart from whether Bucklew could show that HAB has engaged in a pattern of such crimes, the RICO claim fails because the enterprise alleged to have been conducted through a pattern of racketeering activity (defendant HAB, Inc.) is a wholly owned subsidiary of the alleged racketeer (the other defendant, Hawkins, Ash, Baptie & Co.). The claim is that the parent stole the software and gave it to the subsidiary to market. A parent and its wholly owned subsidiaries no more have sufficient distinctness to trigger RICO liability than to trigger liability for conspiring in violation of the Sherman Act, see Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 777, 104 S.Ct. 2731, 81 L.Ed.2d 628 (1984), unless the enterprise's decision to operate through subsidiaries rather than divisions somehow facilitated its unlawful activity, which has not been shown here. Bachman v. Bear, Stearns & Co., 178 F.3d 930, 932 (7th Cir.1999); Emery v. American General Finance, Inc., 134 F.3d 1321, 1324 (7th Cir.1998); Fitzgerald v. Chrysler Corp., 116 F.3d 225, 226-28 (7th Cir.1997); Bessette v. Avco Financial Services, Inc., 230 F.3d 439, 448-49 (1st Cir.2000); Brannon v. Boatmen's First National Bank of Oklahoma, 153 F.3d 1144, 1147-49 (10th Cir.1998).
To summarize, the judgment on liability is affirmed in part and reversed in part, and the case is remanded for a redetermination of the damages to which the plaintiff is entitled.
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.
