Harvard Law Review Harvard Law Review Harvard Law Review

Harvard Law Review Forum

More Than a Thousand Words in Response to Rebecca Tushnet

Responding to Rebecca Tushnet, Worth a Thousand Words: Images of Copyright, 125 Harv. L. Rev. 683 (2011)

I have come to the world of law from the other side, the world of pictures, and much of my time spent in law’s precincts has been spent as a “translator” between the realm of words and the realm of pictures, so I read Professor Tushnet’s analysis1 with great interest.2 She reviews a variety of ways that pictures are discussed in legal opinions, providing a sort of ethnographic account of how law pictures pictures, and then situates copyright in terms of that discussion. Her argument is that pictures are different from words and that the law treats them differently, tending to view them as either “opaque” (uninterpretable) or “transparent” (their obvious meaning being their recognizable subject).3 She maintains that both views are simplistic, and neither yields a coherent approach to how copyright ought to apply to pictures. (I gather that her recommended “greater epistemic humility”4 is an appeal for a more nuanced understanding of what a picture is, although she does not explicitly offer an affirmative view of what pictures can contribute beyond being a part of the landscape of copyrightable material.) Tushnet concludes that because “[c]opyright is literal”5 (her first words), the best way to clear the fog of confusion is to retreat to a simple, clear right of reproduction, thereby eliminating all the problems raised by the notion of “substantial similarity,” the amount of a copyrighted work that can be used, and so on.6 This proposal has the satisfying result of situating copyright battles in the market for works where the real issues are plagiarism and counterfeit copies, not sampling, quotation, or reuse in either scholarship or transformative work by others. I fully support both of these prescriptions. The contrast that Tushnet uses to differentiate pictures and words poses some problems, however; in view of the challenges posed by the Internet, where all expression is representation made of the same material and everything is first a picture or a sound behind a picture, the different status of words and pictures requires further sorting. So while I like Tushnet’s proposal, I want to complicate the conversation because copyright law is now at the center of very large debates about the Internet and our public lives.

The ethnographic account Tushnet uses, of law’s problems with handling pictures, does not reflect on reasons why it might be important to sort out how the law views pictures other than to bring order to a disorderly branch of the law. And in the course of her discussion, text is left unperturbed, a stable field against which we can see how pictures are unruly, hard to contain. Debates about intellectual property can impact our freedoms, our rights, our obligations, our environment, and the future of our culture. It is popular in some industries to concretize intellectual property as simple property that can be owned like a house or a car, making unauthorized use of it stealing. This way of conceptualizing intellectual property does not work in the digital world. Some may feel fear and dismay at the breaking up of an old order; others may see promise and renewal in reconsidering our legal norms.


Pictures are different from words.7 They are perceptually immediate, they can be vivid, and under some circumstances they can be confused with reality itself. No matter whether the pictures reflect the visible world or make ideas visible or visualize what cannot be seen without special instrumentation, or even convey quantitative data, they are always associational. That is, their elements are displayed in space; we read the relationships between the parts within a framing edge as they are visually bound together and related through many possible qualities (not grammatical order). They acquire meaning from our associations to them, drawn from our perceptual knowledge, experience, cultural setting. We experience them as communication assigning content not just to subject matter but to size, color, closeness or distance in the visual field, and so on. Elements of pictures can look like the thing they represent, perhaps the way we might visually encounter the real thing (picture Las Meninas by Velí¡zquez),8 or they can refer to other pictures of the same thing (think of Picasso’s many variations on the theme of Velí¡zquez’s painting9) or to other realities altogether, conjured up through skillful deployment of the same optical cues that we use to parse our everyday visual worlds (think about work like Dalí­’s paintings10 and films like Avatar11). Just as elaborate scientific instrumentation is letting us see things we have never seen before (a new and beautiful view of the weather over time courtesy of NASA data, for instance12), digital technology is now cheap enough so that people can make and exchange still and moving pictures in unbelievable numbers. (YouTube has forty-eight hours of video uploaded every minute; more video is uploaded in a month than the major U.S. television networks created in sixty years; seventy percent of its traffic comes from outside the United States.13 Flickr, devoted to still photography, houses over six billion of the world’s photos.14) The combination of sophisticated tools priced for general consumption and virtually free publication via the Internet has fundamentally altered communication.

Pictures, once simply entities made and looked at, have become something very different as the public “speaks,” deploying them as part of debate on issues, the very kind of speech protected by the First Amendment. When Lieutenant John Pike used pepper spray on seated nonviolent demonstrators at the University of California-Davis, videos made by observers of his act went viral immediately. Very quickly his image was isolated from the photographic frame and then mashed up (collaged) into pictures of all kinds.15 What seems to have started with Pike’s spraying the signing of the Declaration of Independence in John Trumbull’s painting16 has become a flood of Pike’s spraying all of our culture. While no one picture says it all, it is easy upon reviewing many pages of these pictures in an image search that yields 188,000 results to read that people collectively feel that his act was a profound betrayal of American cultural norms. It is no wonder that content industries feel threatened, not just because people recirculate what they regard as their property, but also because they are no longer able to exercise the kind of control over the marketplace that they imagine they used to have. Others are not so sure they like debate by the public and are quite sure they do not like to lose control over what the public knows.17 Disputes in copyright have become part of conducting business as patents are stockpiled and deployed to assert broad forms of commercial protection that can suppress innovation or just make trouble for competitors.18

The world of pictures as evoked by the language that Tushnet quotes is “irrational” but powerful;19 pictures persuade “without overt appeals to rhetoric”;20 they are dangerous “because they seem so real . . . they make people feel rather than think.”21 The immediacy of pictures (their all-at-once-ness) without constraints of distance and time22 is a problem.23 Tushnet shows us that meaning in pictures is context-bound and that this aspect complicates any assessment of pictures in terms of intellectual property: to cite a picture, the whole thing needs to appear, so determining whether it was used fairly rests largely on a judgment about whether its meaning has shifted. And, of course, assertions about the meaning of pictures in a legal discussion are accomplished through a “translation” into words.


Are words really so stable in their meaning? Are they not also conditional on context? Words in texts remain abstract at the level of their representation and unfold over the sequences of phrases, sentences, and paragraphs, which are built step by step in reading. The relationship between an alphabet and sounds, letters and words, is purely symbolic and a matter of custom. (Think of English words that are pronounced the same but spelled differently, like the various forms of to, too, and two, or words that look as if they should sound alike but do not. Rough and dough, on the other hand, differ only in one letter but by history and convention they are pronounced differently.) As we deploy language in either speech or writing, we come immediately to the problem of implicature and how both author and reader understand, correctly or not, messages implied by statements.24

The words of law, by contrast, are the means through which, according to Tushnet, the law does its job of “evidencing truth – in the form of pure reason and logic.”25 I do not know whether she believes this statement or whether it is simply a convenient foil against which to cast the language about pictures that she discovers in the record of judicial opinions. However, reason and logic are only part of the legal word-picture. Not only does legal training teach budding lawyers how to read texts in law-specific ways that become markers of professionalism,26 but in law, words are also the medium used to negotiate between and among competing interests and desires, with discordant goals, so that the words’ inherent function is not purely rational but also rhetorical. Indeed, the legal process encourages people to make legal language unclear by challenging and upsetting received meanings. This pesky problem of unstable meaning has led one scholar to imagine that when artificial intelligence is applied to law, one benefit will be that “words will always have the same meaning.”27 But this dream that everyone will understand the same thing when the law “speaks” ignores the obvious: we come from different cultures (even if we share a common language); some of us know more than one language and so are familiar with how languages can construct entirely different worlds. We cannot communicate about the law (or most anything else) without decoding the various implicit and explicit narratives in play, and once we have done that, those narratives become part of our own experience and can become transformed.


Modern culture is breaking up the ancient binary oppositions between words and pictures, going back to “[i]n the beginning there was the word . . . .”28 Writing seems to have emerged out of the need to carry out public administration in the Bronze Age and law developed over time with it.29 For much of human history, reading and writing were the province of elites, and the teaching of literacy (or not) can be understood as a means of exercising power and control. Few learn to read without instruction and practice. Pictures, by contrast, seemed not to require teaching and were made in messy workshops where hands get dirty. Dirty hands are a marker for labor and not for elites. Pictures leap across linguistic barriers. (Those Internet statistics about uploaded visual material, cited above, reflect this idea.) The law, if it is not to be muddled, needs to understand pictures, if only because otherwise the meanings they import will be missed and good judgment diminished. Tushnet seeks to cure the problem by urging judges to consult experts when it comes to visual questions. While experts have much to contribute, they can, like all of us, assume that others interpret pictures in the same way they do.30 The cure for this problem is that judges, like the rest of us, need to begin by critically examining their own responses. Starting there, we can begin to find bias and elements that we ought to discount, and perhaps begin to imagine how varied other people’s responses might be. With our own responses, wider meaning rushes in. Practice in articulating responses and seeing questions may help concrete thinkers become more comfortable with complexity and complex thinkers more articulate about the issues.

If ancient prejudice about pictures clouds our vision, we will not be able to see the issues involving Internet policy. From the point of view of the end user, everything in a monitor is first a picture in a frame before it becomes all the other forms of content the user will experience. Legal content, too, is poured into this frame. For the end user, the Internet is not the fiber-optic cables carrying data under the sea or the vast server farms that store the data; rather, it is entirely representation.31 Most do not think of the software running routers with the protocols that knit it all together. This electronic environment is a creature that is simultaneously physical and virtual, spatial and temporal, binding what used to be binary into a new whole that is much closer to the human mind itself. Fair use, a fundamental requirement for culture to continue, meets the First Amendment all over the Internet. Copyright law was intended to confer limited monopoly for the benefit of authors and inventors, not to stop speech in its tracks. If Tushnet’s assessment that “[c]opyright law is literal”32 is correct, it is conceptually not capable of performing the governance role that it is being asked to play when assertions of copyright begin to affect the actual technological arrangements that constitute the Internet.33 What does it mean that we experience law now as a picture? How can we protect copyright if it closes down speech? Tushnet does not raise these questions, but her perceptive analysis provokes them.

1. Rebecca Tushnet, Worth a Thousand Words: The Images of Copyright, 125 Harv. L. Rev. 683 (2012).

2. I prefer to use the word “picture” instead of “image,” as this use makes it clear that we are discussing things that exist outside our heads in contrast to purely mental imagery.

3. Tushnet, supra note 1, at 684.

4. Id.

5. Id.

6. Id. at 687-88.

7. For an extended discussion of the differences between pictures and words, see Christina Spiesel, Reflections on Reading: Words and Pictures and Law, in Law, Mind and Brain 391 (Michael Freeman & Oliver R. Goodenough eds., 2009).

8. Ralph Bohnsack, The Interpretation of Pictures and the Documentary Method, 9 F.: Qualitative Soc. Res. (September 2008), http://www.qualitative-research.net/index.php/fqs/article/view/1171/2591.

9. Diego Velí¡zquez, Las Meninas, 1656, Covers & Citations, http://search.it.online.fr/covers/?p=63 (last visited Feb. 14, 2012).

10. Salvador Dalí­, The Persistence of Memory, available at http://www.artchive.com/artchive/D/dali/persistence.jpg.html.

11. Avatar (20th Century Fox 2009).

12. Maggie Koerth Baker, Nine Weeks of Weather in Three Minutes, boingboing (Jan. 3, 2012, 2:20 PM), http://boingboing.net/2012/01/03/9-weeks-of-weather-in-3-minute.html.

13. Statistics, YouTube, http://www.youtube.com/t/press_statistics.

14. Kay Kremerskothen, 6,000,000,000, Flickr (Aug. 4, 2011), http://blog.flickr.net/en/2011/08/04/6000000000/.

15. For the development of the pepper spray meme start here: Xeni Jardin, Video Remix: UC Davis Pepper Spray Incident Viewed from 4 Different Perspectives, boingboing (Nov. 21, 2011, 1:55 PM), http://boingboing.net/2011/11/21/video-remix-uc-davis-pepper-s.html. For a selection of examples see Xeni Jardin, Occupy Lulz, boingboing (Nov. 20, 2011, 1:49 PM), http://boingboing.net/2011/11/20/occupy-lulz.html. For the huge flood, see Search Results for “Pepper spray meme,” Google Images, http://images.google.com/ (search “pepper spray meme”). The Hitler reaction remix: Sarahharbin, Hitler Reacts to Pepper Spray Meme, YouTube (Nov. 22, 2011), http://www.youtube.com/watch?v=iU7y81FBxu4.

16. There are a number of versions of this painting, including a very large one in the United States Capitol Rotunda (1817-19) in Washington, D.C. Yale University’s Trumbull Collection has a smaller version from 1832. What is generally regarded as a good reproduction can be viewed at Declaration of Independence, Architect of the Capitol, http://www.aoc.gov/cc/art/rotunda/declaration_independence.cfm (last visited Jan. 1, 2012).

17. Powerful interest groups often attempt to stifle or circumvent public debate that would challenge their positions. Consider, for instance, the desire to protect or resist state secrets, whistleblowers, and Freedom of Information Act (FOIA) requests. Two very recent news stories provide examples. The first is the story of the Komen Foundation trying to slip a change of policy into its grants – delving into an issue sure to provoke the outcry that came. See Pam Belluck et al., Ban Is Reversed by Cancer Group, N.Y. Times, Feb. 4, 2012, at A1, available at http://www.nytimes.com/2012/02/04/health/policy/komen-breast-cancer-group-reverses-decision-that-cut-off-planned-parenthood.html. The second example is the Bulgarian government, which had signed the Anti-Counterfeiting Trade Agreement (ACTA) without any open debate; a number of members of the Bulgarian parliament ultimately protested against that act. See Bulgarian MPs Wear Guy Fawkes Mask to Protest ACTA, Sofia News Agency (Feb. 3, 2012), http://www.novinite.com/view_news.php?id=136340.

18. See Timothy B. Lee, Study: Patent Trolls Have Cost Innovators Half a Trillion Dollars, Ars Technica (Sept. 20, 2011, 10:21 AM), http://arstechnica.com/tech-policy/news/2011/09/study-patent-trolls-have-cost-innovators-half-a-trillion-bucks.ars; Chris Foresman, Apple May Be Using Patent Troll to Do Its Legal Dirty Work, Ars Technica (Dec. 12, 2011, 3:25 PM), http://arstechnica.com/apple/news/2011/12/apple-may-be-using-patent-troll-to-do-its-legal-dirty-work.ars.

19. Tushnet, supra note 1, at 694.

20. Id. at 692.

21. Id. at 695.

22. Id. at 698.

23. Interestingly, this worry echoes Socrates’s concerns about the written rather than spoken word:

[Y]ou who are the father of letters, from a paternal love of your own children have been led to attribute to them a quality which they cannot have; for this discovery of yours will create forgetfulness in the learners’ souls, because they will not use their memories; they will trust to the external written characters and not remember of themselves. The specific which you have discovered is an aid not to memory, but to reminiscence, and you give your disciples not truth, but only the semblance of truth; they will be hearers of many things and will have learned nothing; they will appear to be omniscient and will generally know nothing; they will be tiresome company,having the show of wisdom without the reality.

The Phaedrus, in 1 The Dialogues of Plato 278 (B. Jowett ed., 1920).

24. See H.P. Grice, Logic and Conversation, in The Motion Aftereffect (George Mather et al. eds., 1998).

25. Tushnet, supra note 1, at 697.

26. Elizabeth Mertz, Recontextualization as Socialization: Text and Pragmatics in the Law School Classroom (American Bar Foundation, Working Paper No. 9418, 1994).

27. Anton Tomazic,”E-Justice Based on e-Law,” (June 1-3, 2008), available at http://www.tomazic.info/Tomazic-Conference2008.doc.

28. John 1:1 (King James).

29. Compared to cuneiform, our alphabetic writing is very young indeed. See Lesley Adkins, Empires of the Plain (2003). For the social context of early writing, see Hans Jörg Nissen et al., Archaic Bookkeeping (1993).

30. The work of the Cultural Cognition Project has demonstrated that values govern what people see, how they make judgments about facts, etc. See Dan M. Kahan et al., Whose Eyes Are You Going to Believe? Scott v. Harris and the Perils of Cognitive Illiberalism, 122 Harv. L. Rev. 837 (2009).

31. See Andrew Blum, Netscapes: Tracing the Journey of a Single Bit, Wired (November 4, 2010), http://www.wired.com/magazine/ff_internetplaces/. Contrast this account with The Atlas of Cyberspace, http://personalpages.manchester.ac.uk/staff/m.dodge/cybergeography/atlas/atlas.html (last visited Jan. 14, 2012).

32. Tushnet, supra note 1, at 684.

33. This is the world of the Stop Online Piracy Act and the Protect Intellectual Property Act recently under debate in the United States.