September 16, 2020 Feature

A Seventh Amendment Right to Fair Use Determinations?

Justin Hughes

©2020. Published in Landslide, Vol. 13, No. 1, September/October 2020, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.

Practically everyone in the intellectual property (IP) community knows the Oracle v. Google litigation1 has centered on the copyrightability of the Java language’s application programming interfaces (APIs) and, if they are protected by copyright, whether Google’s use of 37 Java APIs was fair use. The second of these issues has generated interest in a topic that previously had received little attention from courts and commentators: the respective roles of judges and juries in fair use determinations. In Oracle v. Google, an initial jury concluded that the APIs were protected by copyright but deadlocked on fair use; a trip to the Federal Circuit (because the case included patent claims) confirmed the jury’s copyrightability determination and remanded for fair use to be considered anew. A second jury concluded that Google’s repurposing of the 37 APIs was fair use, but the Federal Circuit disagreed, reversing the fair use determination and leaving Google liable for infringement.

Because this is widely believed to be the first time that an appellate panel has reversed a jury’s finding of fair use—and did so on a de novo review standard—the ruling has triggered a lively discussion about the relative roles of judges and juries in fair use determinations. Fair use is a mixed question of law and fact, so, of course, there is the policy issue of how these mixed questions should be resolved and who should make those resolutions. But before the policy debate, there is the constitutional issue: whether defendants in copyright infringement cases have a Seventh Amendment right to jury determination of fair use. Presumably, if there is such a Seventh Amendment right, judges—trial or appellate—may only review jury determinations on a “clearly erroneous” standard.

As in many such discussions, some people are jumping to conclusions colored by the substantive outcome they seek in the litigation; others are simplifying things in a way that undercuts a careful analysis of the precedent, doctrine, and history. In Oracle v. Google, the Federal Circuit presented such simplification in a footnote: “Justice Joseph Story described fair use as a ‘question of fact to come to a jury’ in 1845.”2 An amicus brief before the U.S. Supreme Court repeats this.3 Both the Federal Circuit footnote and the amicus brief refer back to Justice Story’s 1845 decision while riding circuit in Massachusetts, Emerson v. Davies.4

The problem is that Justice Story said no such thing in Emerson v. Davies. Instead of appealing to historical authority that really does not exist, this article proposes that there is only a moderate case for a Seventh Amendment right to jury determination of 17 U.S.C. § 107 fair use—and there are many junctures at which one could conclude that no such right exists.

To see this, we must put the history of copyright cases pre-1791 against the Seventh Amendment analytic framework the U.S. Supreme Court uses. There is no question that an action for copyright infringement could be brought before a jury in 18th century England, although there is also no question that many or most cases occurred in courts of equity.5 But in modern Seventh Amendment jurisprudence, the question is not whether the cause of action would have been heard before a jury; the question is whether the particular trial decision or its “analogue” came before juries. So the answer to the Seventh Amendment question turns partly on what, if anything, is understood as the “analogue” of modern fair use in 18th century English case law.

A Brief Pre-1791 Copyright History

Professor Ned Snow—also cited on this point in the amicus brief—believes that “[t]he early history of copyright law demonstrates that English common-law courts heard the issue of fair use, expressly reserving the issue for juries to determine,” and, more boldly, “[f]or over two centuries, courts have treated the issue of fair use in copyright law as a question for the jury.”6 For anyone who learned that the fair use doctrine traces back to Justice Story’s 1841 decision in Folsom v. Marsh,7 Snow’s historical claims are jarring. And that is because they are a real stretch.

There is no question that when a plaintiff sued at law, early English cases sent the general question of infringement to juries, sometimes framed in words and concepts that now sound like fair use considerations. Professor Snow’s argument is that “fair use” was built into these primary findings of infringement, i.e., “[e]arly English copyright cases contemplated the sort of use that a defendant made of a work, deciding whether the use was permissible or not.”8 To prove this, Snow relies principally on language in two cases: Sayer v. Moore (1785)9 and Cary v. Kearsley (1802).10

Before we delve into what is actually said in those cases, remember that in the 18th century, English courts were divided into courts of law (king’s bench, common pleas, or exchequer of pleas) and courts of equity (chancery).11 A plaintiff could bring a copyright infringement suit in a court of equity, which could award injunctive relief, and many of them did. Legal historians have yet to find any ruling or statement that a copyright infringement defendant in a court of equity had the right to refer the case to a court of law or to have any elements of the infringement action decided by a jury sitting in one of the courts of law. Instead, as we will see below, just the opposite transpired.

As to Sayer v. Moore and Cary v. Kearsley, both cases were before the king’s bench (law), and both involved what we would now consider principally factual works: Sayer concerned sea charts and Cary concerned road books, an early form of road atlases.12

In Sayer, the plaintiffs had expended substantial monies in preparing their sea charts, which used a “plain” projection, while the defendant’s charts used the more modern Mercator projection. From this difference alone, we know that the “expression” of the facts presented in the two sets of sea charts was quite different. Professor Snow quotes Lord Mansfield’s comments in Sayer: “In all these cases the question of fact to come before a jury is, whether the alteration be colourable or not? . . . [T]he jury will decide whether it be a servile imitation or not.”13 From those words, Snow concludes that “Lord Mansfield considered the issue of whether a defendant’s use was permissibly fair or impermissibly infringing to be one for the jury.”14

But when one reads the entirety of what was testified and what Lord Mansfield said, it becomes evident that Mansfield’s prospective jury charge addressed what we now call the fact/expression dichotomy and the Arnstein-esque question of whether the defendant copied at all.15 Lord Mansfield says:

The Act that secures copy-right to authors guards against the piracy of the words and sentiments; but it does not prohibit writing on the same subject. As in the case of histories and dictionaries: in the first, a man may give a relation of the same facts, and in the same order of time; in the latter an interpretation is given of the identical same words. In all these cases the question of fact to come before the jury is, whether the alteration be colourable or not? There must be such a similitude as to make it probable and reasonable to suppose that one is a transcript of the other, and nothing more than a transcript. So in the case of prints, no doubt different men may take engravings from the same picture. The same principle holds with regard to charts; whoever has it in his intention to publish a chart may take advantage of all prior publications. There is no monopoly of the subject here, any more than in the other instances; but upon any question of this nature the jury will decide whether it be a servile imitation or not.16

Mansfield says that when dealing with two factual works (“histories and dictionaries”), the jury must consider the possibility that independent creation produced similar works: “[t]here must be such a similitude as to make it probable and reasonable to suppose that one is a transcript of the other, and nothing more than a transcript.”17 In modern American copyright parlance, this is “probative similarity” as evidence of copying.18 When Mansfield says “with regard to [sea] charts; whoever has it in his intention to publish a chart may take advantage of all prior publications,”19 he might have been addressing the copying of expression, but he definitely was addressing the copying of facts.

Sure, there may have been some fair use–like considerations in Mansfield’s thinking, but that should not overshadow how much the dispute was about the fact/expression dichotomy and whether the defendant copied little more than facts or only facts from the plaintiffs. (And this passage above is what Justice Story was quoting in the 1845 Emerson v. Davies case; it wasn’t Justice Story’s own words. Emerson v. Davies was again about whether the defendant copied at all, and if so, how much—not fair use.20)

Similarly, the works at issue in the Cary v. Kearsley case were akin to road atlases, and the defendant’s work had the same town names, the same distances between towns, and even some of the same errors that had appeared in the plaintiff’s work. Lord Ellenborough noted that “the defendant was authorised to use a work published as this of the plaintiff’s, to make extracts from it into any original work of his own”21—a statement that makes sense given that the plaintiff’s work was largely a compilation of facts. Supporting that the dispute was about extraction of facts from a fact-based work, Lord Ellenborough compared what the defendant did to “only using an erroneous dictionary.”22

Nonetheless, Lord Ellenborough and the plaintiff’s counsel had a colloquy that seems to go much further. When the plaintiff’s counsel asked whether copying “a whole essay” would be a “piracy,” Lord Ellenborough replied that “a man may fairly adopt part of the work of another.”23 He added, “the question will be, Was the matter so taken used fairly[?]”24 Those comments do sound like a discussion of fair use considerations. Lord Ellenborough then made some comments about how two works compiling surveys and distances of various roads would be similar, saying he would convey all this thinking “to the jury, leaving them to say, whether what [was] taken or supposed to be transmitted from the plaintiff’s book, was fairly done . . . or taken colourably, merely with a view to steal the copy-right of the plaintiff.”25

Is Cary v. Kearsley “the origins of fair use,” as Bill Patry believes?26 Sure, but saying one detects the origins of fair use is not saying one detects fair use. It is fine to say that amino acids are the origin of life, but their detection in a petri dish does not mean one has detected life.

In her careful “micro” level legal history of the Cary case and its antecedents, Isabella Alexander notes:

[I]t becomes clear that words like “originality” and “fair use” are simply not being used in the same way that we use them today. While this does not mean that the cases are of no value, it does mean that judicial uses of such words can be less easily deployed as precedent supporting contemporary argument.27

Instead of positing that Cary proves “fair use” was sent to a jury, Alexandra Sims is another academic who reaches the more measured conclusion that Cary “represents the beginning of a judicial recognition of fairness in relation to the use of factual materials in the creation of new works.”28

Turning from these cases, another way one could argue for a Seventh Amendment right to jury determination of fair use is through “fair abridgment” case law. Scholars also generally agree that the defense of fair abridgment that sometimes arose in 18th century cases was the doctrinal ancestor of modern American fair use; indeed, Folsom v. Marsh was a fair abridgment case.29 One can believe that fair abridgment should be treated as an “analogue” to modern fair use for Seventh Amendment analysis and use this to reason that there is a right to jury determination of fair use.

How Seventh Amendment Jurisprudence Might Apply

How all this plays out also depends partly on the choice between a couple of prior occasions when the Seventh Amendment met intellectual property.

In the 1998 Feltner v. Columbia Pictures decision,30 the U.S. Supreme Court considered whether there is a Seventh Amendment right to jury determination of statutory damages under 18 U.S.C. § 504(c). The Court noted that English “copyright suits for monetary damages were tried in courts of law, and thus before juries,” both under the 17th-century common law and the 1710 Statute of Anne, which itself established statutory damages.31 The historical record recited by the Court also showed that some of the pre-1791 American copyright statutes specified that infringement actions would be tried at “law” or in an “action of debt,” both of which would go to a jury—and that three of those state statutes had statutory damage provisions. This evidence, coupled with one 1789 Connecticut decision in which a jury awarded the copyright owner £100 (under a statute with statutory damages),32 led the Court to conclude that the “right to a jury trial includes the right to have a jury determine the amount of statutory damages, if any, awarded to the copyright owner.”33

Two years earlier, the Court had engaged in a more elaborate Seventh Amendment analysis in Markman v. Westview Instruments.34 The Markman Court concluded that judges should handle the interpretation of patent claims and that such patent claim construction was not a “guaranteed jury issue” under the Seventh Amendment.35 The 1996 Markman decision highlights one critical piece of the Seventh Amendment machinery: while “[t]he right of trial by jury thus preserved is the right which existed under the English common law when the Amendment was adopted,”36 that does not mean every issue in such a trial must be decided by the jury. As the unanimous Markman Court noted:

[W]e ask, first, whether we are dealing with a cause of action that either was tried at law at the time of the founding or is at least analogous to one that was. If the action in question belongs in the law category, we then ask whether the particular trial decision must fall to the jury in order to preserve the substance of the common-law right as it existed in 1791.37

This was in keeping with the Court’s reasoning in earlier cases that “the question whether a jury must determine [some issue] in a trial in which it must determine liability” turns on “whether the jury must shoulder [that] responsibility as necessary to preserve the ‘substance of the common-law right of trial by jury,’”38 and that “the Amendment was designed to preserve the basic institution of jury trial in only its most fundamental elements.”39 Again, in its 1999 City of Monterey v. Del Monte Dunes decision, the Court said the issue is “whether the particular issues, or analogous ones, were decided by judge or by jury in suits at common law at the time the Seventh Amendment was adopted.”40

So, can we conclude that 18th century “fair abridgment” is an appropriate analogue to modern fair use and that fair abridgment was an issue for jury determination before 1791? First, it is not clear that prior to 1791 there was any regular practice of juries deciding fair abridgment.41 The vast majority of cases were brought in equity (before the chancery). Chancery court could not impanel a jury, but, generally speaking, a defendant sued in chancery could request that factual issues involving witnesses be sent to a court of law to be tried before a jury.

Did they have a right to such a jury determination? That sort of request is what happened in the 1740 chancery case Gyles v. Wilcox, Barrow, and Nutt42 concerning a work that had verbatim copied a prior law book with “only some old statutes hav[ing] been left out” and some translations added. The Lord Chancellor noted that “[w]here books are colourably shortened only,” that would be an infringement of the copyright, whereas “a real and fair abridgement” that showed new “invention, learning, and judgment” would not be.43 (If you want to liken that to a determination of transformative use, go ahead.)

Confronted with a request to send the “factual” issues to a jury, the Lord Chancellor judging the case said:

Mr. Attorney-General has said that I may send it to law to be determined by a jury; but how can this possibly be done? . . . The court is not under an indispensable obligation to send all facts to a jury, but may refer them to a master, to state them, where it is a question of nicety and difficulty, and more fit for men of learning to inquire into, than a common jury.44

Again, this was a case of a court sitting in equity and does not directly show what a court of law would have done. At the same time, Gyles is not the only occasion where an 18th century English court of equity sent the fair abridgment question to a special master, not a jury.45

Even if we found meaningful evidence that the question of fair abridgment before a law court was typically decided by juries, one could reason that fair abridgment is an ancestor of fair use without being what the U.S. Supreme Court calls an “analogue” in Seventh Amendment jurisprudence. In their careful reconstruction of the history of intellectual property in Great Britain, Brad Sherman and Lionel Bently conclude that the 1710 Statute of Anne was understood at the time as a “subject-specific statute” and that “there was no (modern) copyright law (as it is understood today) until the middle of the nineteenth century.”46 If Sherman and Bently are correct, that is another reason why pre-1791 English jury determinations of fair abridgment may not be a reasonable Seventh Amendment analogue to fair use, especially in challenging technology-related areas—think intermediate copying for reverse engineering, Google Books, or copying APIs to build a new operating system.

All this makes for an equivocal case that there is a Seventh Amendment right to jury determination of the overall fair use inquiry.


In 2010, copyright scholar Ronan Deazley wrote, “definitive readings as to the meaning, impact, and legacy of the Statute of Anne, and the case law that it prompted, remain as elusive as ever.”47 Perhaps that is an overstatement. There are things we definitely know about 18th-century copyright, but Deazley does speak to the modesty with which one should approach the historical record. Considering what English jurists said and did in 1740, 1774, 1785, and 1802, there is probably no clear answer on whether there should be a Seventh Amendment right to jury determination of modern § 107 fair use. If there is not such a right, then we fall back to the policy question of how we think the fair use mixed question of law and fact is best handled. For some, the answer will be juries know best. For others, fair use’s role mediating copyright and free expression—and everyone’s needs for reasonably predictable outcomes—means the bench is a better place to lodge final fair use determinations.


1. Oracle Am., Inc. v. Google LLC, 886 F.3d 1179 (Fed. Cir. 2018), cert. granted, 140 S. Ct. 520 (2019) (No. 18-956) (set for argument Oct. 7, 2020).

2. Id. at 1194 n.3.

3. Brief of Amicus Curiae American Intellectual Property Law Association in Support of Neither Party at 26, Google LLC v. Oracle Am., Inc., No. 18-956 (U.S. Jan. 13, 2020) (citing Oracle v. Google, 886 F.3d at 1194 n.3).

4. 8 F. Cas. 615, 623–24 (C.C.D. Mass. 1845) (No. 4,436).

5. H. Tomás Gómez-Arostegui, What History Teaches Us about Copyright Injunctions and the Inadequate-Remedy-at-Law Requirement, 81 S. Cal. L. Rev. 1197, 1222–23, 1273 (2008); H. Tomás Gómez-Arostegui, Equitable Infringement Remedies Before 1800, in Research Handbook on the History of Copyright Law 195, 196 (Isabella Alexander & H. Tomás Gómez-Arostegui eds., 2016); Isabella Alexander, Sayer v. Moore (1785), in Landmark Cases in Intellectual Property Law 59, 66 (Jose Bellido ed., 2017) (noting that in the 1780s, “the Court of Chancery tended to be the forum of choice for copyright plaintiffs of the time”).

6. Ned Snow, Who Decides Fair Use—Judge or Jury?, 94 Wash. L. Rev. 275, 276, 279 (2019).

7. 9 F. Cas. 342 (C.C.D. Mass. 1841) (No. 4,901).

8. Snow, supra note 6, at 292.

9. Sayer v. Moore (1785) 102 Eng. Rep. 138; 1 East 358. The spelling of Sayer/Sayre is contested; many think “Sayer” is the (more) correct form.

10. 170 Eng. Rep. 679; 4 Esp. 168 (1804). Of course, by the lights of Seventh Amendment jurisprudence, the 1802 decision can only be evidence of what English courts might have been doing prior to 1791.

11. The chancery court also had some limited jurisdiction as a court of law not relevant here.

12. Isabella Alexander, “Manacles upon Science”: Re-evaluating Copyright in Informational Works in Light of 18th Century Case Law, 38 Melbourne U. L. Rev. 317, 326 (2014) (exploring a line of cases ending in Cary v. Kearsley and noting that the litigations concerned “road books [which] were an early form of road atlas”).

13. Snow, supra note 6, at 294.

14. Id.

15. See Arnstein v. Porter, 154 F.2d 464 (2d Cir. 1946).

16. Sayer v. Moore (1785) 102 Eng. Rep. 138, 140.

17. Id.

18. See, e.g., Johnson v. Gordon, 409 F.3d 12, 18 (1st Cir. 2005) (“[T]hat a sufficient degree of similarity exists between the copyrighted work and the allegedly infringing work [gives] rise to an inference of actual copying. We have referred to that degree of similarity as ‘probative similarity.’”); Peel & Co. v. Rug Market, 238 F.3d 391, 394 (5th Cir. 2001) (explaining that “factual copying” may be inferred from proof of access to the copyrighted work and “probative similarity”); Alan Latman, “Probative Similarity” as Proof of Copying: Toward Dispelling Some Myths in Copyright Infringement, 90 Colum. L. Rev. 1187 (1990).

19. Sayer, 102 Eng. Rep. at 140.

20. 8 F. Cas. 615, 623 (C.C.D. Mass. 1845) (No. 4,436) (“And the question then comes to this, whether [the defendant] has, in substance, copied these pages, in plan, method, arrangement, illustrations and tables, from the plaintiff’s work, with merely colorable alterations and devices to disguise the copy, or whether the resemblances are merely accidental, and naturally or necessarily grew out of the objects and scheme of the defendant, Davies’s work, without any use of the plaintiff’s.”).

21. Cary v. Kearsley (1802) 170 Eng. Rep. 679, 679 (emphasis added).

22. Id.

23. Id. at 680 (emphasis added).

24. Id.

25. Id. at 681.

26. William F. Patry, Patry on Fair Use § 1:11 (2019).

27. Alexander, “Manacles upon Science, supra note 12, at 324.

28. Alexandra Sims, Appellations of Piracy: Fair Dealing’s Prehistory, 2011 Intell. Prop. Q. 3, 21.

29. Matthew Sag argues that the fair use doctrine should be “understood as the continuation of a long line of English fair abridgment cases, dating back to the beginning of statutory copyright law in 1710.” Matthew Sag, The Pre-History of Fair Use, 76 Brook. L. Rev. 1371, 1372–73 (2011). However, Sag notes that the transition from fair abridgment in the mid-1700s to the modern fair use doctrine was both “gradual” and “haphazard.” Id. at 1411. And he warns that “[d]espite the continuity and congruity between the premodern and contemporary understandings . . . the parallels should not be overstated.” Id. at 1404.

30. Feltner v. Columbia Pictures Television, Inc., 523 U.S. 340 (1998).

31. Id. at 348–49.

32. Id. at 350–51 (citing Hudson & Goodwin v. Patten, 1 Root 133, 134 (Conn. Super. Ct. 1789)).

33. Id. at 353.

34. Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996).

35. Id. at 380.

36. Id. at 376 (alteration in original) (quoting Baltimore & Carolina Line, Inc. v. Redman, 295 U.S. 654, 657 (1935)).

37. Id. (citation omitted).

38. Tull v. United States, 481 U.S. 412, 425–26 (1987) (quoting Colgrove v. Battin, 413 U.S. 149, 157 (1973)). In contrast to Feltner, in Tull the Court concluded that “the trial court and not the jury should determine the amount of penalty” under penalty provisions of the Clean Water Act, even assuming that there was a right to trial by jury on the basic question of liability. Id. at 427.

39. Galloway v. United States, 319 U.S. 372, 392 (1943).

40. City of Monterey v. Del Monte Dunes at Monterey, Ltd., 526 U.S. 687, 718 (1999) (emphasis added).

41. It is not even clear what kind of abridgments were thought prohibited, if any at all. See Ronan Deazley, The Statute of Anne and the Great Abridgement Swindle, 47 Hous. L. Rev. 793 (2010).

42. 26 Eng. Rep. 489.

43. Id. at 490.

44. Id. at 490–91.

45. According to Gómez-Arostegui, in Strahan v. Newbery (1774), at the interlocutory injunction phase, the court asked a master to decide whether a “Book published by the defendants is a fair abridgement.” Gómez-Arostegui, Equitable Infringement Remedies, supra note 5, at 209.

46. Brad Sherman & Lionel Bently, The Making of Modern Intellectual Property Law 208 (1999).

47. Deazley, supra note 41, at 816. For a deeper dive into these and other issues on whether judges or juries should decide fair use, see Justin Hughes, The Respective Role of Judges and Juries in Copyright Fair Use, 58 Hous. L. Rev. (forthcoming 2020).


Justin Hughes is the Hon. William Matthew Byrne Distinguished Professor of Law at Loyola Law School, Los Angeles, where he teaches intellectual property and international trade courses.