Preview of Supreme Court Oral Argument in Google v. Oracle
On October 7, 2020, the Supreme Court will hold oral argument in Google v. Oracle, potentially one of the most consequential copyright cases in decades. This post will explain the complex facts and issues before the Court. For further background on the case, see here.
What is the case all about?
We’ll start with what the case is not about. It’s not about whether computer programs are subject to copyright protection. For forty years, it has been clear that computer programs are copyrightable. However, Oracle and its supporters misleadingly argue that Google is seeking to weaken copyright protection for all software.
Actually, the case concerns a much narrower issue: did Google infringe copyright when it copied the links—the interfaces—between different program elements? How the Court resolves this issue could affect the ability of software developers to create new programs that interact with existing programs. Because software drives such a wide range of products, the outcome of this case could have an enormous impact on competition and innovation throughout our economy.
What did Google copy and why?
In the 1990s, Sun Microsystems created a programming language it named Java. Additionally, it created a library of standard subroutines that a programmer could “call” rather than write the subroutine from scratch. In the Java language, the call creates a link to a line of code called a “declaration,” which in turn links to the “subroutine.” There usually are different ways to write a subroutine that implements a particular function. But a declaration can be written only in one way for a particular call to invoke a matching subroutine.
Java became one of the most widely adopted programming languages. There is no question that the Java language is not copyrightable. There also is no question that the pre-written subroutines, referred to as application programming interfaces or “APIs,” are copyrightable. But what about the declarations that link to the subroutines? That is the subject of the decade-long litigation between Oracle and Google.
When developing the Android smartphone platform, Google decided to use Java as the programming language. It quickly realized that the existing libraries of subroutines developed by Sun for desktops would not be appropriate for the smaller processors and limited memory and battery life of smartphones. Thus, Google wrote its own libraries of subroutines in a manner suited for smartphones. At the same time, because of the large number of Java programmers, and the large number of existing applications written in Java, Google sought to make Android as interoperable with the Java apps as possible. To do so, Google copied 11,000 declarations for approximately 20% of subroutines in the Java libraries. However, Google wrote its own code implementing these subroutines. In total, Google copied less than 0.5% of the code in the Java libraries.
By reusing some of the Java declarations, which matched the Java calls, Google made it easier for Java programmers to port their existing apps with the Java calls to Android without having to re-write the apps from scratch. Also, because the Java programmers did not need to learn an entirely new set of calls, they could more easily write new programs for the new smartphone environment.
When Google placed Android on the market in 2008, Sun issued a welcoming press release. But in 2010, Oracle purchased Sun’s assets, including the copyrights in the Java APIs. Oracle promptly sued Google for copyright infringement.
What happened in the lower courts?
In 2012, U.S. District Court Judge William Alsup, of the Northern District of California, ruled that the declarations fell outside the scope of copyright protection. His ruling was completely consistent with the precedents of the U.S. Court of Appeals for the Ninth Circuit, within which the Northern District of California is located. However, because Oracle originally sued Google for patent infringement as well as copyright infringement, Oracle’s appeal was heard by the U.S. Court of Appeals for the Federal Circuit, rather than the Ninth Circuit. The Federal Circuit has exclusive jurisdiction over all cases involving patent claims, even if the patent claims are no longer at issue, as in this case. Even though the Federal Circuit was required to follow the Ninth Circuit precedents, it instead followed long rejected precedents from the Third Circuit, and overturned Judge Alsup’s decision. Google petitioned the Supreme Court to review the Federal Circuit’s erroneous decision, but in 2015 the Court declined to do so.
The case was remanded to the district court for a trial on whether Google’s use of the declarations was a fair use. After a three-week trial, the jury in 2016 reached a verdict for Google. Oracle again appealed to the Federal Circuit, and the Federal Circuit again ruled in favor of Oracle in 2017. Once more ignoring precedent, the Federal Circuit accorded no deference to the jury’s decision. Instead, it treated the jury’s decision as merely “advisory.” Google petitioned the Supreme Court to review both of the Federal Circuit’s decisions, and the Court granted certiorari (cert.) in 2019. The oral argument was originally scheduled for March 2020, but was delayed because of the Covid-19 pandemic.
What are the issues before the Supreme Court?
The Court granted cert. on two issues. First, are the declarations copied by Google subject to copyright protection? Second, even if the declarations fall within the scope of copyright protection, was Google’s copying nonetheless permitted under the fair use doctrine?
With respect to the unprotectability of the declarations, Google makes two arguments. First, it argues that the declarations are “methods of operation” beyond the scope of copyright protection. (Copyright protects expression, not idea, systems, procedures, or methods of operation.) The declarations are the method for calling specific subroutines.
Second, Google asserts that any expression there may be in the declarations “merged” with their function of linking to the subroutines. If a program written in Java included a call for a particular subroutine, the program could be reused in the Android environment only if Google copied the Java declaration that matched that call. (The merger doctrine is a well-established copyright principle.)
In essence, Google argues that the declarations are different from regular program code. They have the specific function of linking programs to pre-written subroutines, and they can perform that function only if they are written in a certain way. Oracle, by contrast, argues that the declarations shouldn’t be treated differently from other code. And if they are treated differently, they should be accorded more protection because they reflect more creativity than other code.
With respect to fair use, Google argues that the four fair use factors set forth in the Copyright Act tilt in its favor. Of special importance here are the first factor, the purpose and character of the use, and the fourth factor, the impact of use on the market for the copied work. When assessing the first factor, courts look at whether the use is “transformative:” whether it adds something new, with a further purpose or different character, and does not substitute for the original use. Google argues that incorporating the declarations in the Android platform was transformative; the highly constrained smartphone environment is completely different from the desktop environment for which the Java APIs were originally written. Oracle, on the other hand, asserts that Google’s use is not transformative because the declarations have exactly the same function in the Android platform as in the Java platform.
In connection to the fourth fair use factor, Google argues that the Java APIs were not in use in the smartphone market, and thus Android did not harm the market for the Java APIs. Oracle counters that some mobile products did, in fact, use the Java APIs.
What is ironic about the positions taken by Oracle?
During the 1990s, both Oracle and Sun argued in the United States and in the European Union that software interfaces fell outside the scope of copyright protection. Indeed, a Deputy General Counsel of Sun was the chairman of the American Committee for Interoperable Systems, an advocacy group that lobbied Congress and filed more than a dozen amicus briefs making this point. Now that Oracle sees the potential for a significant damages award and harming a Silicon Valley rival, it is taking the opposite position.
What curveball did the Supreme Court throw?
After the parties completed briefing the protectability and fair use issues discussed above, the Supreme Court ordered supplemental briefing on the appropriate standard of review of a jury determination of fair use. This suggests that the Court might be looking for a procedural way out from having to decide substantive legal questions turning on complex software technology. The Court could reverse the Federal Circuit on the grounds that it did not afford sufficient deference to the jury’s fair use determination. The Court could rule that the Federal Circuit should have overturned the jury’s verdict only if no reasonable jury would have found fair use, but there was ample basis here supporting the jury’s verdict. Indeed, given the broad industry support for Google discussed in the next section, suggesting that no reasonable jury could have found for Google is absurd.
Who supports each side?
Google enjoys broad support from the technology industry. Amicus briefs supporting Google were filed by CCIA, Internet Association, the Developers Alliance, IBM, Microsoft, Red Hat, Mozilla, Medium, Reddit, Python Software Foundation, Engine Advocacy, and 85 computer scientists. 97 law professors also supported Google’s position, as well public interest groups such as the Center for Democracy and Technology, the National Federation of the Blind, Electronic Frontier Foundation, the American Antitrust Institute, R Street, and Public Knowledge. The American Intellectual Property Law Association filed a brief in support of neither party, but the positions it advocated clearly favored Google.
The concern animating these briefs is the impact this case might have on software interoperability. Oracle and the Federal Circuit have summarily dismissed concerns about the impact of this case on interoperability, asserting that Android and Java are not interoperable. While Android and Java are not completely interoperable, they are as interoperable as possible given that they function in radically different environments. Moreover, the Java declarations at issue in this case are a form of software interface. If the declarations are subject to copyright protection, then other software interfaces in other products are as well. The declarations cannot be distinguished from other software interfaces in a manner that makes sense from a computer science perspective. Accordingly, affirmance of the decisions below would have a profound negative impact on interoperability.
Oracle has much weaker support from the technology industry. Supporting briefs were filed by four software firms (SAS Institute, Synopsis, Mathworks, and Dolby Laboratories); ten computer scientists; and a telecom trade association whose most important members refused to join the brief. Eleven law professors filed briefs in favor of Oracle. Oracle also received strong support from the associations representing the entertainment industry: the Songwriters Guild, the Motion Picture Association, the Association of American Publishers, the Authors Guild, the Copyright Alliance, and the Recording Industry Association of America. These briefs appear motivated more by anti-Google animus than the actual issues involved in this case. This litigation concerns the application of copyright to a highly specialized element of functional computer programs: software interfaces. It has nothing to do with how copyright treats highly expressive works such as novels, films, and songs. These briefs attack the strawman position that Google is seeking to use copyrighted material simply because it is “popular” with engineers. Briefs filed by entities such as the Digital Justice Foundation, the Internet Accountability Project, the Hudson Institute, and the American Legislative Exchange Council seem similarly motivated by anti-Google animus or extremist views on intellectual property protection.
The most significant brief supporting Oracle was filed by the Solicitor General. This brief failed to discern the obvious differences between software interfaces, such as the Java declarations, and other computer code. It also rejected Google’s merger argument, although it acknowledged that the calls and the declarations were “created contemporaneously and specifically designed to work in tandem.” The SG brief suggested that questions of interoperability were better addressed by fair use than by merger, but then advocated an application of fair use so stringent that no interoperable product could ever be created.
What might happen at the oral argument?
Oracle likely will use misleading analogies to Harry Potter and other traditional copyrightable works, as it has throughout the litigation. It will suggest that Google is attempting to free-ride on its fanbase, just like someone other than J.K. Rowling writing an eighth Harry Potter novel. Google, in contrast, probably will stress the highly functional nature of the declarations, perhaps employing the analogy to labels on files in a file cabinet, as it did in its briefs. The SG might attempt to minimize the impact affirmance would have on interoperability. The Justices’ questions could focus on the appropriate standard of review of a jury’s fair use determination. This conversation could go very far afield, including the right to jury trials in copyright cases in Eighteenth Century England.
This is the first fair use case in the Supreme Court since Campbell v. Acuff Rose, decided in 1994. And it is the first software copyright case in the Supreme Court since Lotus v. Borland in 1996, when the Court reached a 4-4 tie, because of Justice Stevens’ recusal. Thus, the Court’s decision could provide clarity on significant copyright issues implicating competition and innovation in the technology industry. At the same time, it is possible the Court will resolve the case on the basis of the standard of review of jury determinations of fair use.