The idea/expression dichotomy is fundamental to copyright law, and can be stated with relative ease:
That simple statement, however, hides an awful lot of conceptual and analytical depth: where is the line drawn between "idea" and "expression"? As an example, we can easily conclude that the notion of a boy wizard attending a school for wizards and witches is not protected by copyright, but somewhere along the line between that skeletal concept and its formulation in Harry Potter and the Philosopher's Stone, it transforms from an unprotected "idea" to protected "expression". How do we articulate where that line exists?
Academics and judges have long struggled with the dichotomy, which is particularly important in the entertainment industries, since imitation can sometimes be more valuable than innovation - if I have an idea for a show about vampires, at what point, if ever, do I need worry about treading on the copyrights of True Blood, The Vampire Diaries, the Twilight books or movies, Angel or any of more than the half-dozen Dracula movies which have been made? Three of the best academic papers on trying to articulate the line between idea and expression are Allen Rosen's “Reconsidering the Idea/Expression Dichotomy” ((1992) 26 UBCL Rev 263), Carys Craig's“Resisting Sweat and Refusing Feist: Rethinking Originality After CCH” ((2007) 40 UBCL Rev 69) and Abraham Drassinower's "A Rights-Based View of the Idea/Expression Dichotomy in Copyright Law" ((2003) 16 Canadian Journal of Law and Jurisprudence) (my own paper which touches on the idea/expression dichotomy, "'I’ve Got This Great Idea for a Show…' – Copyright Protection for Television Show and Motion Picture Concepts and Proposals" (2004) 17 Intellectual Property Journal 189, is available for download here).
Learned Hand, when considering the idea/expression dichotomy in Nichols v Universal Pictures Corp., 45 F2d 119 (2d Cir 1930), opined that "nobody has ever been able to fix that boundary, and nobody ever can". And yet courts are constantly called upon to try to identify what constitutes unprotected idea and what constitutes protected expression. Two recent United States appellate court decisions provide further insight into how courts will (and should) attempt to meet the challenge.
In the 9th Circuit Court of Appeals decision in Mattel v MGA Entertainment (09-55673) (text of decision) (hat tip: Sookman), the dispute centred around a claim of copyright infringement asserted in respect of dolls: did the Bratz doll line infringe copyright held by Mattel in the iconic Barbie doll (an overview of the dispute, though from a couple of years ago, can be found in this BusinessWeek article). The 9th Circuit articulated its analysis, before concluding that there was no copyright infringement, as follows [emphasis added]:
Mattel, of course, argues that MGA went beyond this by copying Bryant’s unique expression of bratty dolls, not just the idea. To distinguish between permissible lifting of ideas and impermissible copying of expression, we have developed a two-part “extrinsic/intrinsic” test. See Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1442 (9th Cir. 1994). At the initial “extrinsic” stage, we examine the similarities between the copyrighted and challenged works and then determine whether the similar elements are protectable or unprotectable. See id. at 1442-43. For example, ideas, scenes a faire (standard features) and unoriginal components aren’t protectable. Id. at 1143-45. When the unprotectable elements are “filtered” out, what’s left is an author’s particular expression of an idea, which most definitely is protectable. Id.
Given that others may freely copy a work’s ideas (and other unprotectable elements), we start by determining the breadth of the possible expression of those ideas. If there’s a wide range of expression (for example, there are gazillions of ways to make an aliens-attack movie), then copyright protection is “broad” and a work will infringe if it’s “substantially similar” to the copyrighted work. See id. at 1439, 1146-47. If there’s only a narrow range of expression (for example, there are only so many ways to paint a red bouncy ball on blank canvas), then copyright protection is “thin” and a work must be “virtually identical” to infringe. See id.; Satava v. Lowry, 323 F.3d 805, 812 (9th Cir. 2003) (glass-in-glass jellyfish sculpture only entitled to thin protection against virtually identical copying due to the narrow range of expression). The standard for infringement—substantially similar or virtually identical—determined at the “extrinsic” stage is applied at the “intrinsic” stage. See Apple Computer, 35 F.3d at 1443. There we ask, most often of juries, whether an ordinary reasonable observer would consider the copyrighted and challenged works substantially similar (or virtually identical). See id. at 1442. If the answer is yes, then the challenged work is infringing.
... When works of art share an idea, they’ll often be “similar” in the layman’s sense of the term. For example, the stuffed, cuddly dinosaurs at issue in Aliotti v. R. Dakin & Company, 831 F.2d at 901, were similar in that they were all stuffed, cuddly dinosaurs—but that’s not the sort of similarity we look for in copyright law. “Substantial similarity” for copyright infringement requires a similarity of expression, not ideas. See id. The key question always is: Are the works substantially similar beyond the fact that they depict the same idea? MGA’s Bratz dolls can’t be considered substantially similar to Bryant’s preliminary sketches simply because the dolls and sketches depict young, stylish girls with big heads and an attitude.
The 11th Circuit Court of Appeals also recently considered the idea/expression dichotomy, in Baby Buddies, Inc. v. Toys “R” Us, Inc. (08-17021 / 03-01377 CV-T-17-MSS) (text of decision) (hat tip: Exclusive Rights), a case involving two similar "bear-themed" baby pacifiers (the Exclusive Rights post has a photo of the two pacifiers side-by-side for ease of comparison).
Equally fundamentally, “[i]n no case does copyright protection for an original work of authorship extend to any idea.” 17 U.S.C. § 102(b); see also Harper & Row Publishers, 15 Inc. v. Nation Enters., 471 U.S. 539, 556, 105 S. Ct. 2218, 2228, 85 L. Ed. 2d 588 (1985) (“No author may copyright his ideas or the facts he narrates.”). Accordingly, we must apply the substantial similarity test to only those elements of the copyrighted work that are actually subject to copyright protection—that is, elements of original expression in the copyrighted work. ...
In deciding whether the protected elements of two works are substantially similar, we compare the various components of the two works, but are mindful that “lists of similarities are ‘inherently subjective and unreliable, particularly where the lists contain random similarities, and many such similarities could be found in very dissimilar works.’” Corwin, 475 F.3d at 1251 (quoting Herzog, 193 F.3d at 1257). At the same time, “[a]nalyzing relatively amorphous characteristics of the [work] as a whole (such as the ‘mood’ or ‘combination of elements’) creates a danger of unwittingly extending copyright protection to unoriginal aspects of the work.” Leigh, 212 F.3d at 1215. At the most narrow, focused level, two works will almost always be distinguishable, and at the broadest level of abstraction they will almost always appear identical. Thus, although we identify and compare the protected expressive features of the two works, we do so not simply to count the number of similarities and differences, but rather to determine whether the work’s protected expression has been copied. ...
We begin with the plastic teddy bears, which are the most prominent features of both pacifier holders. As mentioned, both bears are formed from white, sculpted plastic. They also share the same basic anatomical features common to all teddy bears—a head, two ears, two eyes, a nose, a mouth, a torso, two upper paws (representing the fore paws of a real bear), and two lower paws (representing the hind paws on a real bear). But every sculpture of a teddy 17 bear shares these features simply because these features are what defines a teddy bear. To protect this basic combination of features would in effect give Baby Buddies exclusive rights over the very idea of a plastic sculpted teddy bear, which is expressly precluded under the copyright laws.
From there, the decision continues on to examine such elements of the bears as the positioning of the paws, the shape of the ears and the belly of the bears (pot belly vs a flatter torso - I swear I am not making this up...).
In addition to providing further examples of the nature of the analysis which US courts undertake, tThe decisions are notable from a Canadian perspective because of the highly formulaic manner in which the analysis proceeds: identify the elements of both works, separate out the protectable from the non-protectable elements, then compare the protectable elements to determine whether copying of a substantial part has occurred. While that step-by-step approach is nascent in some Canadian cases (see my paper, linked to above, for its application in Canadian entertainment copyright cases), Canadian courts should regularize their analysis in a similar fashion so as to encourage the predictability of the infringement analysis.