Summary
holding that t-shirt design was not protectable because it expressed only the same idea as similar t-shirts
Summary of this case from Trek Leasing, Inc. v. United StatesOpinion
No. 98-1408.
Heard September 15, 1998.
Decided October 9, 1998.
Appeal from the United States District Court for the District of Massachusetts, Hon. Robert E. Keeton, U.S. District Judge.
James G. Goggin with whom Jacqueline W. Rider and Verrill Dana, LLP were on brief for appellant.
Andrew D. Epstein with whom Barker, Epstein Loscocco was on brief for appellees.
Before Selya, Circuit Judge, Aldrich, Senior Circuit Judge, and Boudin, Circuit Judge.
Denise Matthews, a fabric designer and wholesale distributor of "soft-good" souvenirs, makes and sells a children's t-shirt with a design entitled, "Someone Went to Boston and got me this shirt because they love me Very much." This phrase appears on a t-shirt surrounded by small drawings of a fish, a sailboat, a lobster, Faneuil Hall, and scattered hearts. Matthews registered this "Someone . . ." t-shirt design with the U.S. Copyright Office, receiving a certificate of registration effective August 9, 1995.
In February 1997, Alan Freedman, a competing souvenir producer, began offering a children's t-shirt with the slogan "Someone Who Loves Me Went to Boston and Got Me this Shirt." This phrase appears on the Freedman t-shirt and is surrounded by small drawings of a fish, a sailboat, a lobster, a swan boat, ducklings, and a smiling sun. Freedman and Matthews both attend the Boston Gift Show where Freedman has seen Matthews' "Someone . . ." t-shirt displayed.
Matthews filed suit against Freedman for copyright infringement under 17 U.S.C. § 501 et seq. in August 1997. The district court held a non-jury trial on January 21, 1998, and issued a memorandum and order dated February 5, 1998, granting judgment in favor of Freedman on the ground that Freedman's t-shirt did not infringe upon the protectible elements of Matthews' design. Thereafter, the district court also awarded attorney's fees to Freedman in the amount of $25,000. This appeal by Matthews followed.
To make out a copyright infringement, a plaintiff must show ownership of a valid copyright — not questioned here — and the "copying of constituent elements of the work that are original." Feist Publications v. Rural Tel. Serv. Co., 499 U.S. 340, 361 (1991). As to this latter element, sometimes the dispute concerns the factual question of whether the defendant saw and copied the plaintiff's work; but in other cases, as in this one, the dispute centers around more subtle, related questions: what aspects of the plaintiff's work are protectible under the copyright laws and whether whatever copying took place appropriated those elements.
A major limitation on what is protectible under the copyright laws is capsulized in the notion that copyright protects the original expressions of ideas but not the ideas expressed. See Feist, 499 U.S. at 345-51; CMM Cable Rep, Inc. v. Ocean Coast Properties, Inc., 97 F.3d 1504, 1516 (1st Cir. 1996). The point is that the underlying idea (e.g., the travails of two star-crossed lovers), even if original, cannot be removed from the public realm; but its expression in the form of a play script (such as William Shakespeare's Romeo and Juliet) can be protected. Needless to say, the line is a blurry one.
If the idea-expression dichotomy favors copyright-case defendants, a somewhat different notion in copyright law favors the plaintiffs. It is well-settled that where the plaintiff has copyrighted an original expression, an infringement can be found to exist even where the defendant does not copy the original expression in a literal manner (word for word or image by image). It is enough to make out infringement — assuming actual copying — if the alleged infringing work is "substantially similar" to the protectible expression in the copywritten work. CMM Cable, 97 F.3d at 1513. In other words, if Romeo and Juliet were under copyright, the infringer could not escape solely by changing the names of every character and altering a few words in every line.
The substantial similarity rubric is sometimes used in this way — to define infringement — and sometimes used in a related but distinguishable sense, i.e., as an evidentiary inference. Thus, a substantial similarity between two works may help plaintiff establish that in fact defendant borrowed protectible expression from the plaintiff's work (as opposed to creating the alleged infringing work out of the defendant's own imagination — in which case even virtual identity would not be unlawful). See Feist, 499 U.S. at 346; see also 3 Nimmer on Copyright, § 13.03[A], at 13-27 to 13-29 (1997); cf. Eden Toys, Inc. v. Marshall Field Co., 675 F.2d 498, 501 (2d Cir. 1982).
A potential tension is now obvious. An alleged infringing work taken as a whole may seem "substantially similar" to the copyrighted work taken as whole, but the impression of similarity may rest heavily upon similarities in the two works as to elements that are not copyrightable — because those elements are the underlying ideas, or expressions that are not original with the plaintiff, or for some like reason. See Feist, 499 U.S. at 358-60; CMM Cable, 97 F.3d at 1515. Courts have used various formulas to isolate the protectible expression in the copyrighted work to determine whether the alleged infringing work is "substantially similar" to that protectible expression.
See, e.g., CMM Cable, 97 F.3d at 1514-15 ("dissection analysis"); Whelan Assoc. v. Jaslow Dental Lab., 797 F.2d 1222 (3d Cir. 1986), cert. denied, 479 U.S. 1301 (1987) (iterative test); Sid Marty Krofft Television Prod., Inc. v. McDonald's Corp., 562 F.2d 1157, 1163 (9th Cir. 1977) ("abstractions test").
That is just what the district court did in the present case. It declined implicitly to give Matthews any protection for the idea that underlies the phrasing on the t-shirt — that the t-shirt's purchase represents someone's love of the donee — or the unoriginal notions of having a legend on a t-shirt, using childish lettering, or placing on the t-shirt symbolic emblems reminiscent of a location (here, Boston). The district court then turned its attention to the remaining elements and found that, as to these elements, the differences between the two works precluded a finding of infringement.
In this regard, the district court pointed to differences in the particular icons selected, their arrangement, the colors used, and the lettering. Examination of Matthews' and Freedman's t-shirts bears out the district court's conclusion; no one would think the lettering, colors or particular icons had been literally copied from one t-shirt to the other. For example, the icons are not all of the same subjects and even where both t-shirts depict a lobster or fish, the drawings are completely different.
Because Matthews does not directly challenge the district court's assessment on an element-by-element basis, we pause only to address the most obvious possible claim: that the expression of the "someone" theme in the two t-shirts is "substantially similar." Obviously, there is a good deal of similarity in the substance of the two competing phrases: "someone went to Boston and got me this shirt because they love me very much" and "someone who loves me went to Boston and got me this shirt." But most of the similarity rests on the fact that both phrases are seeking to express essentially the same idea.
Admittedly, there is some similarity in the two expressions of the idea, but this is largely because both are expressing the same idea briefly and in straightforward terms and the available variations in wording are quite limited. Even if the sentiment were original with Matthews — which is by no means clear — it would virtually give Matthews a monopoly on the underlying idea if everyone else were forbidden from using a differently worded short sentence to express the same sentiment. Cf. CMM Cable, 97 F.3d at 1519-20.
In all events, Matthews' attack on the district court's decision is not based on any refined comparison of individual elements. Rather, Matthews says broadly that the test for copyright infringement is whether the "ordinary observer" would find a substantial similarity between the two works and would be disposed to overlook the disparities as minor. Matthews then argues that the district court, at least if its judgment is reviewed de novo (as Matthews urges), erred because the two t-shirts are to the average observer substantially similar "in design and overall aesthetic appeal" and because "both exhibits have similar 'Someone Who Loves Me Went To Boston And Got Me This Shirt' language and display boats, fish, and a lobster in a childlike fashion."
Case law supports the "ordinary observer" test and the disregard of minor differences in favor of major similarity. See Concrete Mach. Co. v. Classic Lawn Ornaments, Inc., 843 F.2d 600, 608-11 (1st Cir. 1988). As for the proper standard of review, there is some difference among the circuits where the question of substantial similarity is based solely on a visual comparison of two items easily available to the reviewing court. But even if we assumed arguendo a de novo standard of review, the protectible elements here are not substantially similar for reasons already indicated.
Compare Folio Impressions, Inc. v. Byer California, 937 F.2d 759, 766 (2d Cir. 1991) (applying de novo review) with Wildlife Express Corp. v. Carol Wright Sales, Inc., 18 F.3d 502, 506 n. 1 (7th Cir. 1994) (adhering to a "clearly erroneous" standard). The First Circuit has not taken a clear position. See O'Neill v. Dell Publ'g Co., Inc., 630 F.2d 685, 687 (1st Cir. 1980).
Indeed, Matthews' own argument (just quoted) is practically explicit in its refusal to distinguish what is protectible from what is not. Lack of originality to one side, the notion of displaying "boats, fish and a lobster in childlike fashion" is not in itself an expression; and although a particular drawing might be, the two t-shirts in this case do not display similar icons. For reasons already indicated, the "someone" phrasing in the two t-shirts reflects the same sentiment, but the sentiment is not protectible, and Freedman's expression of it was different from Matthews', given the limited options available.
We do not mean to suggest that every element in copyrighted work must be considered apart from all others; a collection of protectible elements may seem substantially similar primarily through the overall impression created by those protectible elements taken together. Cf. Knitwaves, Inc. v. Lollytogs Ltd., 71 F.3d 996, 1003-04 (2d Cir. 1995). Even entirely unoriginal expressions taken from others might sometimes be combined in a way that gave rise to a new protectible expression (imagine a collage of newspaper headlines juxtaposed in some highly creative and original fashion). Cf. Feist, 499 U.S. at 358-59. But this case falls into neither of these categories.
Turning to the issue of attorney's fees, Matthews urges us to reverse as an abuse of discretion the district court's award of $25,000 in attorney's fees. She says that because the district court found that both parties made "non-meritorious" arguments, the Supreme Court decision in Fogerty v. Fantasy, Inc., 510 U.S. 517, 534 (1994), dictated an even-handed treatment of attorney's fees, namely, none to either side. Matthews also claims that the district court failed to consider all of the Fogerty factors and that even if an award was proper, the fees awarded here were excessive.
The copyright statute departs from the usual American rule and says that the district court "may" award attorney's fees to the prevailing party. 17 U.S.C. § 505. In Fogerty, the Supreme Court rejected the view that fees should be awarded more readily to prevailing plaintiffs than to prevailing defendants, 510 U.S. at 533; and it said that the weakness of arguments made was a relevant consideration. Id. at 534 n. 19. But the Court did not say that fees should be withheld from the prevailing party simply because both sides made some arguments that did not persuade.
Here, the district court ruled that the Matthews' claim was not frivolous and was not brought in bad faith, but that Matthews "did try to extend her copyright protection far beyond what is allowed by law." This last conclusion is a judgment call and, on the present facts, well within the district court's discretion. A plaintiff's decision to bring a weak, if non-frivolous, case and to argue for an unreasonable extension of copyright protection are relevant concerns. Fogerty, 510 U.S. at 526-27; Edwards v. Red Farm Studio Co., 109 F.3d 80, 82-83 (1st Cir. 1997). Depending on other circumstances, a district court could conclude that the losing party should pay even if all of the arguments it made were reasonable.
Apart from the weakness of the arguments on both sides, Matthews says only that her company is not a "corporate giant"; but it may well be larger than Freedman's, and anyway Matthews makes no effort to develop the details, although free to offer such evidence in the district court. We conclude that the district court gave a rational explanation for its decision to award fees to Freedman, and that Matthews has failed to show an abuse of discretion or any other error. This is equally so as to the amount awarded — $25,000 out of the $60,000 originally sought.
Freedman has sought a further award of attorney's fees to cover this appeal. In our view, Matthews has raised some issues worth airing; the arguments on appeal mirror those made in the district court; and we think that $25,000 is an adequate sum for Freedman for the litigation as a whole, considering that the district court conducted an abbreviated bench trial using stipulated facts and affidavits. Accordingly, in the exercise of our own discretion, we decline to make any further award to Freedman.
Affirmed.