LAW 646

Professor Karjala


2-1/2 Hours


Monday, December 10, 2007                                                                                              8:30 a.m.


            This examination consists of a three essay questions that you are to answer in bluebooks or by computer. The total exam time is 2-1/2 hours. The problems will be weighted 45%-20%-35% in determining the grade on the exam, but you may allocate your time to the problems as you choose. The weights were determined very roughly by the number of issues I found in the problems in drafting model answers. If you wish to allocate your time according to the problem weights, you should devote 67-1/2 minutes to Problem 1, 30 minutes to Problem 2, and 52-1/2 minutes to Problem 3. However, I believe that you should have enough time to deal with all of the problems adequately without formally allocating such blocks of minutes to each one.

            You are permitted to have with you and to use during the examination your textbook, the statutory supplement or other printed version of the Copyright Act (an edited version of the Act accompanies this exam), any class handout materials (including materials printed from the Course Materials web page for the class), your class notes, and any other written materials made by you or your study group (not photocopied from printed materials).

            Base your answers SOLELY on the materials that were assigned on the syllabus or discussed in class. Please remember that your answers should RAISE issues, not eliminate them. Even if you are quite sure that your resolution of one issue eliminates the need to consider others, where significant issues arise with another resolution, try to deal with those issues on the assumption that the first issue is resolved contrary to your own approach to it. This is especially the case if there are facts in the problem that are relevant only to these subsidiary issues and that you have not made use of in resolving the primary issue. This is an EXAM, not real life! Do not give abstract lectures on copyright law (and, in particular, do not give back to me lengthy quotes concerning what you think I said in class), but in applying copyright law to the facts of a given problem, be sure you make full use of the facts that are given. If you have not used all the facts, you are probably missing an issue that you are expected to consider.

            If you write your answers in longhand, please write legibly, double space, and ON ONE SIDE OF A BLUEBOOK PAGE ONLY. If you type or use a computer, try to type enough of the letters of each word correctly to make the word decipherable. Your score will be based on what is actually on the paper, not my guess as to what you might really have meant.


Problem 1 (45%)

            Schools and universities generally have been facing a growing problem of plagiarism off the internet. Seeing an opportunity to make some money by filling a niche, defendant D sets up a website called “”. High schools and universities can join D’s network for an annual fee   per institution. Teachers and professors from member institutions can submit for analysis electronic copies of class or term papers that have been handed in by students, for a per-paper charge. D then, using his special software, analyzes each paper submitted and compares it to D’s growing database of papers as well as material found at other web sites. Any paper submitted for checking is automatically included in D’s database, whether or not the student has authorized such retention. D’s software is highly creative in its use of algorithms to detect similarities between papers, even where there has been substantial alteration by paraphrasing or similar efforts at disguise. D’s database is not available to anyone other than employees of D, meaning that no one, including institutional members, can download any of the papers stored in the database. However, whenever similarities are found between a submitted paper and one found in the database, the similar portions are printed out for comparison purposes in D’s report.

            P is a high school student who has just learned that her paper was submitted to (the report on her paper was “clean,” meaning that D found no evidence of plagiarism) and that a copy of her paper remains in D’s database. P duly registers a copyright in her paper and brings a copyright infringement action against D. D moves for partial summary judgment, arguing that, even if P is the copyright owner and entitled to bring the action, and even if D’s storing P’s paper in D’s database constitutes prima facie copyright infringement (that is, putting aside any claims that the institutions were somehow empowered to license D’s use), D is not liable because including P’s paper in the database is a fair use.

            You are the judge. Decide the motion for partial summary judgment.

Model Answer

            This case is another example of the copyright quandaries raised by digital technologies. In the analog world, if every institution required students to submit two copies of papers, and one of the copies was submitted to an antiplagiarism company that used old-fashioned technology (e.g., eyeballs) to compare newly submitted work to copies retained from earlier submissions by others, no copyright problem would arise at all, because none of the exclusive rights of copyright are breached. Digital technology makes copying easier but also more rampant, in that it is difficult to take advantage of the features of digital representations of works without making more “copies”. Even if D receives its copies of student papers on floppy disks or similar media, D must make a “new” copy to get the new paper into its database. Here, we assume that copies come to D by electronic mail or directly over the internet, but the exact method makes no difference to the analysis because the cases quite uniformly hold that even storing a digital work in RAM constitutes the making of a “copy” (see text p. 88). There can be no doubt that D makes at least one copy in the course of creating his database, so P has a prima facie case of copyright infringement, namely, that D has infringed her exclusive right to reproduce the work under section 106(1). We therefore turn to the fair use defense raised by D’s motion.

            The introductory sentence of section 107 sets out “criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research” as potential fair uses. Here, it is not clear whether D’s storage of P’s paper in his database fits within any of these categories. The most likely candidate is “research,” as D is assisting academic institutions uncover incidents of unethical behavior among their students. In any event, qualifying in this list does not insure that fair use will be found, nor does failure to qualify rule out fair use (as shown most clearly by the Betamax case, where the Supreme Court found fair use in the copying of entire television shows for later viewing). The Supreme Court has said that we must consider all four of the section 107 factors, so we turn to that analysis.

            Factor 1 looks to the purpose and character of the use, including whether the use is of a “commercial nature” or for nonprofit educational purposes. We know that D is trying to make money from his discovery of this niche market, and conceivably at least some of the member institutions are for-profit schools. However, the overall goal of the member institutions is education, and they seek to do that through the entirely laudable approach of requiring students to do their own work rather than copy that of others (even with permission). Moreover, while D operates his business for profit, D does not profit directly from the sale (or other exploitation, such as public performance) of P’s or anyone else’s papers. The niche that D is trying to fill would not work to the extent that D’s database is incomplete. It is essential to the entire operation that essentially all papers available to be plagiarized by students be also contained among the materials that D searches. The facts say that D also searched other web sites, and none of them are complaining at the moment so we put aside any potential copyright infringements as to them. There is no evidence that P’s paper would otherwise be included in D’s searches, so to exclude it would diminish the value of the entire project. Indeed, if D infringes as to P, D infringes as to nearly all of the other student papers that have been included without student permission. The entire operation would collapse. Therefore, while D’s operation must be deemed “commercial,” it is largely balanced by the overall educational goal. Therefore, it only counts weakly against fair use under Factor 1, and there are other elements to be considered under this Factor that turn it in D’s favor.

            The Supreme Court in Campbell instructs us that we must also look under Factor 1 to whether the use is “transformative.” At first glance, we see D making a verbatim copy of P’s entire paper, which hardly seems transformative. However, Perfect 10 v. (Supplement 54) tells us that we should take a more expansive interpretation of the term “transformative.” In that case, the court concluded that making exact, but low resolution, copies of photographic images was “highly transformative” by placing the images in a context that served a different purpose from the original, namely, that of searching for images on the internet. Here, P’s paper was not copied for any of its substantive content, expressive or otherwise. Only a computer would ever “read” D’s copy of P’s paper, unless it turned out that a more recent submission contained suspicious similarities, and even then the only parts that would be read by a human would be in the context of determining whether the new submission was copied from P’s paper. Finding the answer to that latter question is a matter of high social importance. The copying required by digital technology is precisely what permits this new method of detecting plagiarism to work. In this respect, our case is similar to the Google Book Project. While no decision has issued in that case, the argument for fair use in that case is strong, because no single book publisher or copyright owner can possibly put in one searchable database all of the books that exist, because no single copyright owner owns all of the rights (or even a significant portion of them). If Google cannot do it under fair use, nobody else can, either, and that means the world is permanently deprived of a potentially valuable new research tool. No one would seriously argue that detecting plagiarism is quite the same level of social importance as being able to search all books, but the analogy is apt: If D (or someone like D) cannot do this, nobody can, and it simply will not get done.

            One distinction between our case and Google Books is that Google allows copyright owners to opt out of the system. D has not, apparently, given that power to the authors of the papers D includes in the database. If the case were closer, this might be a point on which we should await a full-fledged trial to determine how many school paper authors would opt out of the system if given a chance. One senses that precisely those student authors who would allow others to copy their work (perhaps in return for money or other benefit) are the ones who would opt out, making the system considerably less valuable, and this point could perhaps be more fully developed (or refuted) at trial. On balance, however, Factor 1 still comes out in favor of fair use.

            Factor 2 looks to the nature of the copyright-protected work. P’s paper here was done as part of fulfilling academic requirements of D’s school. The work is unpublished, which counts against a finding of fair use, but more analysis is required. Our record does not show the subject matter of the paper, but most student papers are not written for their creative expression but rather to show the students’ abilities to research a topic, analyze the research, and present it in a logically coherent form. While it is again a question of fact just what percentage of student papers in D’s database are “creative” as opposed to “factual” or “scientific,” so a trial could be useful in sorting this out, this court is sufficiently familiar with student papers in general that it can say with confidence that the vast majority will never again see the light of day, let alone publication or even further distribution beyond a very limited circle of family and friends. The incentive of copyright is not a factor in their creation, as they are required by the educational institutions. Factor 2 weighs slightly in favor of fair use.

            Factor 3 looks to the amount and substantiality of the portions used. Here, P’s entire paper was copied, so it appears at first glance that Factor 3 is against fair use. However, Campbell says we measure substantiality under Factor 3 against the legitimate use determined under Factor 1. Unless the entire paper is included in the database, the chances of detecting partial plagiarism are diminished. The very goal of the system requires that papers be included in their entirety. In this case, because nobody reads the copies of the work made by D, P cannot really care whether all or only a part was copied. (If D’s copies were actually read by others, they would operate as a substitute for copies authorized by D, so copying the whole work would be a stronger factor against fair use.) The Betamax case, and the more recent Google v. Amazon case, among others, show that even copying an entire work can sometimes be fair use. Factor 3 weighs mildly against fair use.

            Factor 4 is sometimes said to be the most important of the four factors and looks to the effect of the use on the potential market for or value of the work. Here there is no evidence that P’s work had any market value. Again, a trial could be useful in determining this with more confidence, but looking at student papers in general we can say that only a tiny percentage, at most, have any potential market, at least as literary works, outside potential sale in a plagiarism market. The latter is not a market that any society would wish to promote. Moreover, D’s use does not interfere in that market, at least in the sense that someone would “buy” a license to plagiarize from D’s copy instead of from P. D’s copy is not available for copying by anyone, for any purpose, so it simply does not compete in this, or any other, market that P might have for her work. Of course, D’s maintaining a copy of P’s work in D’s database may undercut P’s hypothetical efforts to license her paper for plagiarism by others (because of the fear that they will be detected by D’s software). Like biting criticism through parody, however, copyright does not protect against market suppression of this type. The same thing would happen if D’s plagiarism analysis could be carried out in an analog world, where there would be no copyright infringement at all, so no fair use defense would be necessary.

            Finally, however, we must consider the market for the very use that D is making of P’s paper. In the Texaco and Princeton University Press cases, the courts considered failure to license the specific use as to which fair use was being claimed as a harm to the copyright owner’s market. This argument is circular, because there is no need for D to seek a license if D’s use is fair. In any event, it was important in both of the cited cases that mechanisms were in place for users to seek and obtain a license for the use in question (course packs and copies for research use) at reasonable cost. Here, we are aware of no similar mechanism that would reduce or eliminate the market failure due to the high transaction costs of seeking each student’s permission. Many would like refuse (especially if they see the possibility of licensing for copying by other students) and many others would simply not respond. In particular, the students who allow their papers to be used by others would be precisely the students most likely to opt out of D’s system, as discussed above. There seems to be a market failure here that was not present in the Texaco or PUP cases.

            The analogy to parody is important more generally. Parody also involves, at least in general, a market failure, and the Campbell court expressly recognized that fact (without sending the case back for retrial on this point). Few enjoy the biting criticism that is implicit in parody and few would therefore be willing to license parody. Thus, without fair use, the new (and assumedly valuable) work will not get made. The facts of the case before us are perhaps not quite so potent, but the transaction costs remain high because of the very large number of student papers that are written at thousands of educational institutions every year. As discussed above, unless nearly all such papers can be included in some single searchable database, the socially worthwhile goal of detecting plagiarism will likely go unfulfilled.

            The court concludes that D’s use of P’s paper by copying it into a searchable database that is not accessible by anyone except D, who uses it solely for comparison to other papers and not for its substantive or expressive content and in a way that does not compete with any potential market P may have for her paper is a fair use. D’s motion for partial summary judgment is granted. Because this issue is dispositive of the case, the action is dismissed.

Problem 2 (20%)

            In Torres-Negrón v. J&N Records LLC, the BNA summary of which was assigned for reading in the class, the First Circuit concluded that registration of copyright in a work requires the deposit with the Copyright Office of a copy that is made with direct reference to the original. It refused to allow the plaintiff to maintain his copyright infringement lawsuit where the copy that had been submitted for registration was reconstructed from memory. Yet, it is universally accepted in copyright that a work that is once fixed does not lose its copyright when the original material object embodying the work is destroyed. Note also section 101: “A work is ‘created’ when it is fixed in a copy . . . for the first time.” Assuming this latter rule to be correct, namely, that a work once fixed does not lose copyright protection even when the material object in which the work is embodied is destroyed or disappears, evaluate whether the First Circuit’s decision in Torres-Negrón is correct. Consider the following example: Art Professor draws a picture (“Picture”) on the blackboard to illustrate some technical point for her class. The drawing on the blackboard is erased that evening when the custodial service does its nightly cleaning. (Assume that this constitutes “fixation” of the Picture on the blackboard.) Student, relying on memory, reproduces expressive (i.e., protected) elements of Picture when he gets home and begins selling copies. Art Professor wishes to enjoin Student. Can Art Professor reconstruct from memory a copy of Picture for registration, which is a prerequisite to bringing suit against Student?

Model Answer

            The Picture was fixed on the blackboard in class, meaning that it was “created.” Section 102(a) says that copyright subsists in original works of authorship that are fixed. Copyright therefore subsists from the moment of creation. Nothing in the statute says that copyright is lost if the only “copy” embodying the work is destroyed. (Indeed, this might violate the Berne Convention, but we did not study the Berne Convention much in class so we put this to the side.) We are in fact told to assume this rule. Therefore, Art Professor has a copyright in the Picture – the intangible pictorial work that she depicted in fixed form on the blackboard. To say that Art Professor cannot reproduce a copy from memory for registration purposes says, in effect, that we recognize her copyright but refuse to enforce it. It is unlikely that Congress could have intended such a result.

            Admittedly, the First Circuit approach has certain advantages, mainly in eliminating disputes over exactly what the “work” in question really is, which can sometimes be important in determining whether someone is infringing. But denying registration is tantamount to denying copyright protection completely, without any investigation of how close the reconstructed copy is to the now-destroyed copy. If we take the position that Art Professor has done, at this stage, pretty much everything she could do to register her copyright and allow the suit to go forward, Student would still be able to defend on the ground that his drawing is not substantially similar to the Picture that was depicted on the blackboard. Other students were presumably in the class at the time and could give evidence as to whether Art Professor’s registration copy was, in fact, a good reconstruction of the original. If the case is close as to whether Student took protected expression rather than just general ideas or techniques, the court should side with the student, as Art Professor was in the best position to protect the original copy if she had really thought about maintaining her copyright rights. But in our problem we are assuming that Student has included protected elements of Picture in Student’s drawing. In short, allowing the registration (and therefore the litigation to go forward) is messier but fairer. Rules versus standards yet again. I would say the First Circuit has to be wrong, but reasonable minds may differ.

Problem 3 (35%)

            City commissions Sculptor to design and build a large piece of sculpture to decorate the entrance to a park. Sculptor creates the piece, entitled “Nature,” and transfers the physical sculpture and the copyright in the work to City. Nature is “signed” with Sculptor’s mark prominently displayed. According to Sculptor, the piece was designed to blend in with the foliage and other human-made objects in its vicinity (sidewalks, fences, gates, etc.) to convey an overall harmonious and natural feeling. After its unveiling, Nature was the subject of commentary in the local press for a week or so, with both positive and negative comments in roughly equal amounts. Sculptor has been paid everything that City promised to pay under the commission agreement.

            Five years after the unveiling, members of the city council of City decide that the entrance to the park in simply not colorful enough. They seek proposals from various artists on redesigning the entrance. Sculptor sends a note to the council saying that Sculptor objected to any modification of Nature or any of the surroundings in which Nature was designed to fit. Artist submits a proposal to the council to paint the individual pieces of Nature in a variety of bright primary colors. Artist also proposes major revisions to the surrounding foliage, even replacing some nearby trees with colorful striped and polka-dotted poles. Artist’s proposal is accepted on condition that it will be legal for City to carry out Artist’s plan.

            City has come to you for a legal opinion on whether City may legally carry our Artist’s plan and an analysis of the arguments Sculptor is likely to make in what is sure to be a legal challenge by Sculptor if Artist’s plan is adopted. City (with Artist’s support) insists that all of Artist’s plan must be adopted or none of it will “work” artistically. How do you advise City? In the course of initial discussions with City, someone also suggested possibly making an exact replica of Nature and painting it in accordance with Artist’s plan, while moving the original Nature to another part of the park. Would this approach make City’s case any stronger?

Model Answer

            It seems clear that Nature is a work of visual art (WVA). The definition in section 101 defines a WVA as “a painting, drawing, print, or sculpture, existing in a single copy” or, for limited edition works fewer than 200 signed and numbered copies. Nature is a sculpture, and it even bears Sculptor’s identifying mark (if that is necessary for works existing in a single copy – a matter that is not entirely clear from the definition). There are no facts indicating that Nature is a work made for hire (WMFH) or that it fits within any of the other exceptions to the definition of a WVA.

            Although Sculptor has transferred to City all of Sculptor’s section 106 rights, Sculptor retains the rights afforded to authors under section 106A. Nothing in the facts suggests that any waiver has been given with respect to these changes.

            Section 106A(a)(1)(A) gives the author the right to claim authorship of the work. This right is not relevant here, because we (City) are perfectly willing to continue attributing the work to Sculptor if Sculptor so wishes. Section (a)(1)(B) gives the right not to have an author’s name associated with a WVA he or she did not create. If either the original or the “exact” replica is painted, it becomes a derivative work that was not created by Sculptor. Consequently, in either case, Sculptor’s mark will have to be removed or covered if Sculptor so demands. 106A(a)(2) gives in this case the same right, with respect to the “original,” as in (a)(1)(B), provided the painting of Nature is prejudicial to Sculptor’s honor or reputation. Even if it is so prejudicial, City would simply have to remove or cover the mark identifying Sculptor as the author of Nature.

            106A(a)(3) gives Sculptor the right to prevent any intentional “distortion, mutilation, or other modification” of Nature that would be prejudicial to Sculptor’s honor or reputation, and presumably a right to damages if Nature is so intentionally distorted, mutilated, or otherwise modified. Painting the original Nature seems to fit squarely within the term “other modification,” even if it is not a distortion or mutilation (which it might be in the eyes of some). Would such a modification prejudicial to Sculptor’s honor or reputation? It is difficult to say without having more facts. However, Sculptor will be able to muster a strong argument by simply pointing to the bright primary colors that now garishly call attention the piece, in place of the “harmonious and natural” feeling Sculptor designed. There have been no cases in the U.S. as to what constitutes prejudice to honor or reputation in the context of WVA rights. I would not recommend going ahead and painting the original and then having to defend against a parade of art critics that Sculptor can surely line up to testify how much the paint job had ruined the piece and damaged Sculptor’s name and reputation in the artistic community.

            The idea of replacing the original Nature with an exact replica, however, has some merit, from City’s point of view. Sculptor’s work will not be modified in any way, and City now owns the copyright and therefore has the right to reproduce the work in copies. As discussed above, the replica would likely have to leave off the mark identifying Sculptor as author, but that must be done, anyway, if Nature is to be painted. As we discussed in class, the Visual Artists Rights Act (VARA) applies to physical copies of works, not to the intangible works themselves. Although this is not explicit in the statute, most of the provisions of VARA only make sense if we interpret the term “work” in 106A in this way (which is different from the way the term is used in most, if not all, of the rest of the Copyright Act). If it applied to the intangible work, then any run of posters or magazine photos would take the work out of the WVA category. No picture in any museum would be protected from physical damage or modification if copies are sold in the gift shop. So, because City has the right under copyright law to build a replica, and because the replica will not be numbered, signed, or marked by the author, it would seem to be fair game for the painting City wishes to commission. Moreover, neither City nor Artist is, presumably, an “author” with respect to the replica, because, as an exact copy, it lacks originality. This means the replica is also disqualified as a WVA under section 101. (If courts were to adopt the Karjala “originality in the brushstrokes” approach to copyright originality, whoever builds the exact replica would, in fact, be an author, but then that person, as author of a WVA, would be the one entitled to decide whether to permit modifications.)

            The next question is whether moving the original Nature to another part of the park would violate any of Sculptor’s VARA rights. Sculptor claims that the piece was designed to blend in with the foliage and other human-made objects in its vicinity. In other words, he will make the claim that the piece is “site specific.” The Phillips v. Pembroke court (Supp. p.29) holds that VARA does not protect the placement of site specific sculpture, and that is the only jurisprudence we have on the issue. The policy reasoning of the court is convincing. If site-specific art were protected by VARA, it would wreak havoc in our expectations of property law and ownership: Once a piece is in place, it could not be moved without the author’s permission, giving the author tremendous leverage over property owners who seek to modify their properties. If Congress intended this result, it would have been more explicit. Moreover, the definition of WVA in section 101 speaks of “painting, drawing, print, or sculpture.” This does not easily include large-scale integrated art, especially integrated art that is also site-specific. While Phillips would not prevent Sculptor from making this argument outside the First Circuit, so no conclusive opinion is possible elsewhere, it is likely that courts will follow it elsewhere. Moreover, if it comes to litigation, City could at least argue that Sculptor’s claim to site-specificity is unfounded. To the extent that Nature was designed to blend in with the foliage, for example, what happens when real nature takes over and changes the surroundings?

            If, contrary to the above analysis, VARA does protect site-specific art, the next issue is whether moving Nature to another part of the part is either a “distortion, mutiliation, or other modification”that is detrimental to Sculptor’s honor or reputation or, if Nature is a “work of recognized statute,” whether moving it would be a destruction of the work. It would seem that taking Nature out of its intended context would at least me a “modification” of the work, so the issue is whether it would be detrimental to Sculptor’s honor or reputation. Again, Sculptor is likely to be able to muster expert testimony to the effect that the new location (wherever it is) does not give the same harmonious feel that the original location gave, so people seeing the work in the new context would think less of Sculptor than they otherwise would. It would not be fun attempting to refute such testimony.

            Does moving Nature imply that the work has been “destroyed”? We have no cases on what constitutes destruction under VARA. Sculptor would argue that the work is nothing without its intended surroundings and so has been destroyed. City would argue that if it is moved once it can be moved again back to its original context, so merely moving Nature does not destroy it. I would say that City has the best of this argument, but reasonable people could differ. None of this matters, however, unless Nature is a “work of recognized statute.” The courts have set little in the way of objective standards for what this important term in VARA means. The Martin case (p. 630) says that articles and letters from newspapers and magazines complimenting the piece was sufficient to uphold determination of “recognized stature,” while saying that live testimony would usually be required. Here there was some discussion in the local press about the work, both positive and negative, and Sculptor will surely be able to find “experts” who will testify as to Nature’s “recognized stature.” There is a good chance Sculptor will prevail on this issue (but it is irrelevant unless moving Nature amounts to a “destruction”).

            In sum, whether we can adopt artist’s plan (by recreating a new copy for painting and replacing the original Nature with it) depends almost wholly on whether our circuit will follow the Phillips case. If VARA does not protect site-specific art, simply moving the original without making any changes will not invoke any VARA rights that Sculptor owns, and Sculptor has no VARA rights in the new copy because VARA applies to physical copies of works, not the abstract “work” protected by copyright.