circumstantial evidence establishing that the defendant had access to the original work and that the two works are substantially similar. Other indications of copying, such as the existence of common errors, have also been accepted as evidence of infringement. 319 The copying of the copyrighted work must be copying of protected expression and not just ideas;320 likewise, the similarity between the two works must be similarity of protected elements (the expression), not unprotected elements (the facts, ideas, etc.). The portion taken must also be more than de minimis. The similarity between the two works need not be literal (¿.e., phrases, sentences or paragraphs need not be copied verbatim); substantial similarity may be found even if none of the words or brush strokes or musical notes are identical.321 Various tests have been developed to determine 319 See, e.g., Rockford Map Publishers, Inc. v. Directory Serv. Co., 224 U.S.P.Q. 851 (C.D. III. 1984), aff'd, 768 F.2d 145 (7th Cir. 1985), cert. denied, 474 U.S. 1061 (1986); Sub-Contractors Register, Inc. v. McGovern's Contractors & Builders Manual, Inc., 69 F. Supp. 507, 509 (S.D.N.Y. 1946). It is common for publishers of directories and other compilations to deliberately insert mistakes into the work (such as periodically adding a fictitious name, address and phone number in a telephone directory) to detect and help establish copying. See 2 H. ABRAMS, THE LAW OF COPYRIGHT § 14.02[B][3][c], at 14-19 to 20 (1993). 320 This should be implied in the requirement that there be copying of the copyrighted work. Ideas and facts, of course, are not copyrightable. In the case of compilations, such as databases, if enough facts are copied, the copyrighted expression (the selection, arrangement or coordination of the facts) may be copied and infringement may be found. See CONTU Final Report at 42 ("The use of one item retrieved from such a work be it an address, a chemical formula, or a citation to an article would not... conceivably constitute infringement of copyright. The retrieval and reduplication of any substantial portion of a data base, whether or not the individual data are in the public domain, would likely constitute a duplication of the copyrighted element of a data base and would be an infringement."). 321 - - See Donald v. Zack Meyer's T.V. Sales & Service, 426 F.2d 1027, 1030 (5th Cir. 1970) ("paraphrasing is equivalent to outright copying"), cert. denied, 400 U.S. 992 (1971); Davis v. E.I. DuPont de Nemours & Co., 240 F. Supp. 612, 621 (S.D.N.Y. 1965) ("paraphrasing is tantamount to copying in copyright law"); see generally 3 NIMMER ON COPYRIGHT § 13.03[A] at 13-28 to 13-58 (1993). whether there has been sufficient non-literal copying to constitute substantial similarity between a copyrighted work and an allegedly infringing work.322 Judge Learned Hand articulated the well-known "abstractions test," where the expression and the idea are, in essence, treated as ends of a continuum, with infringement found if the allegedly infringing work crosses the line delineating the two.323 Such a line, as Judge Hand recognized, is not fixed in stone; indeed, as he put it, its location must "inevitably be ad boc.... 324 The "pattern" test has also been suggested, where infringement is found if the "pattern" of the work is taken (in a play, for instance, the "sequence of events, and the development of the interplay of characters")." - The "subtractive" test which dissects the copyrighted work, disregards the noncopyrightable elements, and compares only the copyrightable elements of the copyrighted work to the allegedly infringing work -- has been the traditional method for determining substantial Nimmer identifies two bases upon which courts impose liability for less than 100 percent verbatim copying: (1) "fragmented literal similarity" (where words, lines or paragraphs are copied virtually word-for-word, although not necessarily verbatim) and (2) "comprehensive nonliteral similarity" (where the "fundamental essence or structure" of a work is copied); see also P. GOLDSTEIN, COPYRIGHT $7.2.1 at 13-17 (1989). Goldstein identifies three types of similarity: (1) where the infringing work "tracks" the original work "in every detail," (2) "striking similarity" (where a brief portion of both works is "so idiosyncratic in its treatment as to preclude coincidence") and (3) similarities that "lie beneath the surface" of the works ("[i]ncident and characterization in literature, composition and form in art, and rhythm, harmony and musical phrases in musical composition"). Id. at 13 (citations omitted). 322 For analyses of the various tests that have been used, see 3 NIMER ON COPYRIGHT § 13.03[A] at 13-28 to -58 (1993); M. LEAFFER, UNDERSTANDING COPYRIGHT LAW §§ 9.5 - 9.7 at 268-76 (1989). 323 324 See Nichols v. Universal Pictures, Corp., 45 F.2d 119, 121 (2d Cir. 1930). See Peter Pan Fabrics Inc. v. Martin Weiner Corp., 274 F.2d 487 (2d Cir. 1960). 325 See Z. Chaffee, Reflections on the Law of Copyright: 1, 45 COLUMBIA L. REV. 503, 513 (1945). similarity.326 Following the 1970 Ninth Circuit decision in Roth Greeting Cards v. United Card Co.,327 the "totality" test became popular for determining substantial similarity. The totality test compares works using a "total concept and feel" standard to determine whether they are substantially similar. Although chiefly used by the Ninth Circuit in the 1970s and 1980s,328 the test was used by other circuits as well. 329 The Ninth Circuit further defined an "extrinsic/intrinsic" test in proof of substantial similarity in Sid & Marty Krofft Television Productions, Inc. v. McDonald's Corp.330 The intrinsic portion of the test measures whether an observer "would find the total concept and feel of the works" to be substantially similar. 331 The extrinsic portion of the test, meanwhile, is an objective analysis of similarity based on "specific criteria that can be listed and analyzed. 332 Thus, this test requires substantial similarity 326 See Universal Athletic Sales Co. v. Salkeld, 511 F.2d 904, 908-09 (3d Cir.), cert. denied, 423 U.S. 863 (1975) (subtracting all but the "stick figures" from chart as non-protectible subject matter); Alexander v. Haley, 460 F. Supp 40, 46 (S.D.N.Y. 1978) (finding "alleged infringements display no similarity at all in terms of expression or language, but show at most some similarity of theme or setting. These items, the skeleton of creative work rather than the flesh, are not protected by the copyright laws."). 327 328 See 429 F.2d 1106 (9th Cir. 1970). See, e.g., Sid & Marty Krofft Television Prods., Inc. v. McDonald's Corp., 562 F.2d 1157 (9th Cir. 1977); McCulloch v. Albert E. Price, Inc., 823 F.2d 316 (9th Cir. 1987). 329 See, e.g., Reyher v. Children's Television Workshop, 533 F.2d 87 (2d Cir. 1976); Atari, Inc. v. North American Philips Consumer Electronics Corp., 672 F.2d 607 (7th Cir.), cert. denied, 459 U.S. 880 (1982); Atari Games Corp. v. Oman, 888 F.2d 878 (D.C. Cir. 1989); Whelan Assocs., Inc. v. Jaslow Dental Lab., Inc., 797 F.2d 1222 (3d Cir. 1986), cert. denied, 479 U.S. 877 (1987). 331 See Pasillas v. McDonald's Corp., 927 F.2d 440, 442 (9th Cir. 1991). 332 See Brown Bag Software v. Symantec Corp., 960 F.2d 1465, 1475 (9th Cir. "not only of the general ideas but of the expressions of those ideas as well. "333 More recently, however, both the Ninth and Second Circuits have moved away from the totality test, particularly with respect to computer applications. In Data East USÁ, Inc. v. Epyx, Inc.,334 the Ninth Circuit rediscovered "analytic dissection of similarities" in the substantial similarity determination of video games. 335 Similarly, the Second Circuit, in Computer Associates International, Inc. v. Altai, Inc.,336 fashioned an "abstraction-filtration-comparison test" for a computer program that combined Judge Learned Hand's "abstraction" test (to separate ideas from expression) and "filtration" reminiscent of traditional "subtraction' analysis in distinguishing protectible from non-protectible material. 337 In addition to the evolution of substantial similarity tests, there is disagreement as to the appropriate "audience" for determining substantial similarity. The "ordinary observer test" alluded to in Arnstein v. Porter 338 and followed in a number of Second Circuit decisions 339 considers the question of substantial similarity from the - 333 334 335 Krofft, supra note 330, at 1164. 862 F.2d 204 (9th Cir. 1988). See also Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1445 (9th Cir. 1994) (approving of district court's use of analytical dissection and agreeing with other courts' use of the "same analysis although articulated differently"). 336 982 F.2d 693 (2d Cir. 1992). See Autoskill Inc. v. National Educational Support Systems, Inc., 994 F.2d 1476, 1490-91 (10th Cir. 1993). 337 Other circuits have applied this test. See Engineering Dynamics, Inc. v. Structural Software, Inc., 26 F.3d 1335, 1343 (5th Cir. 1994); Gates Rubber Co. v. Bando Chemical Indus., Ltd., 9 F.3d 823, 834 (10th Cir. 1993). 338 154 F.2d 464 (2d Cir. 1946). 339 See, e.g., Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487 (2d Cir. 1960); Ideal Toy Corp. v. Fab-Lu Ltd., 360 F.2d 1021 (2d Cir. 1966); Eden Toys, Inc. v. Marshall Field & Co., 675 F.2d 498 (2d Cir. 1982). 341 viewpoint of the "average lay observer."340 The Fourth Circuit, however, set forth a modified test in Dawson v. Hinshaw Music Inc., requiring the ordinary observer to be the "intended" audience for the particular work. Relying on decisions by both the Ninth and Seventh Circuits, the court in Dawson stated: [i]f the lay public fairly represents the intended 342 The challenge of this test, especially in more advanced technologies, is determining when, if ever, a work is not directed to an audience possessing specialized expertise, and at what point a work once intended for a specialized audience becomes accepted by the general public. The ability to manipulate works in digital form raises an issue with respect to infringement of the reproduction and derivative works rights. A copyrighted photograph, for instance, can be manipulated in the user's computer in such a way that the resulting work is not substantially similar to 342 905 F.2d 731 (4th Cir. 1990). See Aliotti v. R. Dakin & Co, 831 F.2d 898, 902 (9th Cir. 1987) (holding that perceptions of children must be considered in substantial similarity analysis because they are intended market for product); Atari, Inc. v. North American Philips Consumer Electronics Corp., 672 F.2d 607, 619 (7th Cir.), cert. denied, 459 U.S. 880 (1982) (holding that "[v]ideo games, unlike an artist's painting,... appeal to an audience that is fairly undiscriminating insofar as their concern about more subtle differences in artistic expression"). |