[This post has been authored by Ajeeth Srinivas, a 4th year student at School of Law, Christ University, Bangalore.]
“If the majority is going to speculate about what Oracle might do, it should at least consider what Google has done.”
On 5th April 2021, the Supreme Court of the United State of America passed a judgement which could potentially change the fundamentals of copyright law forever. This judgement, in the case of Google LLC v Oracle LLC, deals with the copyrightability of Application Programming Interfaces (“API”). The judgement however explicitly sidesteps this question by stating that the relevance of its judgement would diminish due to fast paced change in technology today. The case, with its long and complex history, for the sake of brevity, can be explained as a simple case of copying. In short, the case revolved around Google’s development of the Android platform, during which time, it copied the codes of Sun Java, originally created by Oracle. Google copied around 11,500 lines of code verbatim from the library created by Oracle. The SCOTUS’s majority judgement overrules previous opinions of the Federal Court which held that such use of code would not amount to fair use under copyright laws. Here, the Court stated that Google’s use of such code was fair, since Google copied the ‘declaring code’, and not the ‘implementing code’.
This article seeks to analyse the dissent, presented by Thomas. J. In this process it shall highlight the rationality of the dissent, by noting the significant differences between the rulings of the majority and the dissent, on similar points of law. This article analyses two major implications resulting out of the judgement, which are its effects on the idea expression dichotomy and merger doctrine in copyright law, and the effects on the fair use doctrine formulated in the United States. In the first part, through an understanding of the format of JAVA, the article covers the impact of this ruling on the Idea-Expression Dichotomy, and analyses future implications on the merger doctrine. The second part seeks to analyse the perspectives on the fair use doctrine, as adopted by the majority opinion and refocuses on the inexorability of the results of these perspectives, on the future of IP disputes relating to fair use. However, to understand the ramifications of this case, the structure of Oracle’s Sun Java needs to be understood.
The Format of JAVA
The mechanism of JAVA has two specific kinds of code which ensure the functioning of the software, when synchronised; these are the ‘declaring’ and the ‘implementing’ codes. The declaring code is the one which gives life to the functions of the code itself by creating different classes and variables and tasks which may be performed using the code. The implementing code, on the other hand, provides specific instructions on how each of the functions and tasks, which were created using the declaring code, perform in a specific setting. The declaring code cannot perform any functions without the implementing code, since there would be no instruction on how the specific tasks are to be performed. Similarly, the implementing code cannot function without the declaring code since there are then no definitions – or “declarations” – of the tasks that the implementing code needs to execute. The two cannot exist in isolation from each other.
Why the Dissent Matters
Thomas. J., delivering the dissent on behalf of Alito. J. and himself, explicitly notes that the majority has failed to acknowledge and answer the pivotal question in this case – i.e., whether the API is copyrightable?
Consequently, one of the main shortcomings of the majority opinion is encountered at the very beginning, where it is assumed that API is copyrightable, without entering the Intellectual Property thickets that APIs have created. This assumption is laid down on the basis of rapidly changing technological, economic, and business circumstances. In addition to this, the majority’s judgement simply fails to understand the extent of the detrimental effects that the copying has had on Oracle. The problems with such an assumption also leads to problems in the way the fair-use argument has been proved by the Court – if APIs were indeed copyrightable, the Court’s fair use analysis would turn out to be inherently flawed, and this is noted in the dissent.
With this background on the case, and on the format of Java, its implications on the Idea-Expression Dichotomy and the Merger Doctrine may be studied.
Implications on the Idea-Expression Dichotomy
One of Google’s primary contentions was that since Oracle’s Sun Java was in the form of a declaring code, it was functional rather than expressive, and therefore an unprotected idea as per Baker v Selden. This would allow free use of the declaring code, resulting in only individual implementing codes being construed as the expression of that idea. However, in holding the declaring code to be an idea, the code is given a static setting, meaning that the code cannot be expressed differently. As later noted by the dissent, the expression of the declaring code can vary from person to person. Oracle’s manner of declaring functions would be inherently different from the manner in which other users would have expressed the code. The majority’s reading of the declaring code as an idea, simply means that there could be no other way of expressing this declaring code, which in the field of computer science, is false.
Section 102 of the US copyright legislation, in essence, codifies the idea-expression dichotomy. The dissent also uses the application of the noscitur a socii principle espoused in United States v Williams to state that the term ‘method of operation’ mentioned in Section 102, when in regard to the implementing code, needs to be understood from the language surrounding it, and not independently. Methods of operation are purely the functional aspects of codes. However, as noted previously, declaring codes by themselves do not provide any function – it is their combination with the implementing code, which gives them value. Without the declaring code, providing the implementing code copyright protection would be of no value. Therefore, in the author’s opinion, when declaring codes in combination with implementing codes are analysed, their expression of a particular idea, and the manner in which they emulate such an idea, is copyrightable. The lack of such an interpretation by the majority, when read with the following analysis on the Merger Doctrine, further complicates matters.
The Effect on the Merger Doctrine
The purpose of the merger doctrine is to prevent ideas, which have only one form of expression, from being copyrighted (See Herbert Rosenthal Jewelry Corp. v. Kalpakian). As previously mentioned, Oracle itself could have expressed this declaring code in numerous ways. This is the reason why the merger doctrine argument perpetrated by Google fails. Oracle’s declaring code, by its very nature, is not an idea but an expression of an idea. Therefore, the application of the merger doctrine to this case cannot be permitted, since the debate in this case does not surround an idea and the subsequent expression of that idea, but different forms of expression itself. To illustrate this further, an example would be a programmer creating a button which said “Accept T&C” to accept the terms and conditions on a website. The idea would be the button itself. The expression of that idea would create the function of the button using the declaring code, and the performance of the button would be dictated by the implementing code working in tandem with the former. In this case, the declaring code would be useless without the presence of the implementing code, and vice-versa. Therefore, the dichotomy in this case is not that the declaring code is an idea and the implementing code an expression, but that the idea subsists in the purpose of the code, and the functions it sought to achieve, and not the declaring code created by Oracle. In this case, the declaring code created by Oracle would, in fact, be a form of expression.
In rendering its judgement however, the majority sought to fit facts into its specific narrative, disregarding the possible issues that it may cause to the interpretation of the basic tenets of copyright law. The next part of this Article will further analyse the majority’s interpretations of the fair use doctrine, and the problematic results of such interpretation.