First-to-File as a Rule of Evidence

Tun-Jen Chiang

Assistant Professor, George Mason University School of Law

In 2011, Congress passed the Leahy-Smith America Invents Act (AIA) and reformed the U.S. patent system. Most significantly, the Act replaced the “first-to-invent” patent system with a “first-to-file” system.1 The law is now the subject of a constitutional challenge.2 In this short Essay, I will explore two obstacles to this challenge: one that I view as insurmountable and one merely very difficult.

At the outset, I should make clear that I have no strongly held views on the wisdom of first-to-file as a policy matter. In my view, the empirical questions of who will win and lose from the change, and by how much, are difficult to measure and have not been measured to any degree of confidence. Nothing that I say should be understood as endorsing first-to-file as a good idea. But I have much confidence that it will and should be upheld as constitutional.

So what is the whole kerfuffle about? A common situation in every patent system is that two people will independently conceive the same invention at roughly the same time. This has happened repeatedly throughout history: Bell and Gray with the telephone, Edison and Swan with the light bulb, etc.3 The question is who should receive the patent in such a situation. Roughly speaking, there are two types of rules for how such a fact situation is to be resolved. In a first-to-invent system, the first person to conceive of an invention in her mind receives the patent. In a first-to-file system, the first person to file a patent application describing the invention receives the patent.4 The debate concerns whether the change to a first-to-file system is consistent with Congress’s constitutional powers.

I. The Challengers’ Argument: The Constitutional Limit of Granting Patents to “Inventors”

First, let me summarize what I understand to be the challengers’ argument, which at this point is underdeveloped because the case is still in its early stages. Article I, section 8 of the Constitution empowers Congress to “promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” The argument is that “Inventors” in this clause refers to the first inventor, i.e., the first person to conceive the invention in her mind.5 Accordingly, the argument goes, Congress lacks the power to grant patents to any other person, including a second inventor who independently conceives the same invention later. Since a first-to-file system will permit such a result to occur, it is unconstitutional.

The challengers seem to recognize that no patent system can ever achieve perfect factual adjudication. Even if Congress mandates a legal standard that the patent must only be granted to the first inventor, some errors in adjudication will occur when a person who is actually the second inventor hoodwinks the United States Patent and Trademark Office and the courts into granting the patent to him. In apparent response to this practical objection, the challengers argue that Congress cannot enact a statutory standard or objective other than granting patents to the first inventor,6 implicitly conceding some allowance for adjudicative errors. This problem of adjudicative error is important, and I will return to it in Part III.

II. The Slippery Slope Problem: The 1952 Act and Second Inventors

My first objection to the challengers’ theory is that it raises a severe slippery slope problem. Even taking the standard as whether Congress can explicitly provide for the grant of patents to second inventors, it turns out that the patent system has done so for a long time. Thus, the challengers cannot challenge the new Patent Act for failing to have an explicit goal of granting patents to only the first inventor, without making the exact same challenge to the old Patent Act for the exact same failure. This objection is not insurmountable—the challengers can simply argue that every Patent Act since 1870 has been unconstitutional—but I have serious doubts that the challengers are willing to make that argument or that a court will accept it.7

Although it is commonly described as a first-to-invent system, the Patent Act of 1952 in fact made two important exceptions to the general principle of granting patents to the first inventor.8 By the Act’s own terms, the second inventor of a particular invention would receive the patent over the first inventor when: (1) the first inventor conceived the invention outside the United States,9 or (2) the first inventor had abandoned, concealed, or suppressed the invention, most commonly by practicing the invention as a trade secret.10 There are reasonable policy arguments for granting a second inventor priority in such cases,11 but the point here is that these exceptions are on-their-face inconsistent with an argument that the Constitution allows patents only to be granted to the first inventor and to no other. If there is a constitutional requirement to grant patents only to the first inventor, we have been openly violating it for sixty years. In fact, we have been openly violating it for over 140 years, since the Patent Act of 1870 had the same two exceptions.12

Anticipating this argument, a group of law professors who support the constitutional challenge to the first-to-file provision have attempted a response. According to these professors:

The Constitution establishes the presumption that first inventors are secured a patent, but it does not mandate that first inventors must receive patents regardless of their own actions. Thus, Congresses and courts have identified instances in which post-inventive actions by a first inventor can result in a default on the right to receive a patent.13

There are two problems with this response. The first is that it is factually incomplete. One exception to the first-to-invent principle in the 1952 Patent Act is post-inventive misconduct by a first inventor that results in the default of his rights, namely where the inventor engages in abandonment, suppression, or concealment of his invention. But the other exception to the first-to-invent principle in the 1952 Act had nothing to do with post-inventive actions—it simply disfavored foreign first inventors and awarded priority to a second inventor when the first inventor’s conception occurred overseas.

The second problem is even more fundamental. As the law professors supporting the challenge say, a rule that first inventors can lose their patent rights due to their own misconduct is not contrary to the principle that Congress can only grant patents to first inventors—the two principles are reconcilable because Congress can legitimately grant no patent at all. But the logic of their argument does not allow Congress to then grant the patent to someone else, which is precisely what the 1952 Patent Act does. The challengers have a plausible textual argument that “Inventors” means only “first inventors”; they do not have a plausible textual argument that “Inventors” mostly means “first inventors,” but then means “second inventors” when the first inventor commits misconduct or is a foreigner. If the AIA is unconstitutional because it allows patents to be granted to a second inventor, then so is the 1952 Patent Act.

III. An Overlooked Problem: How to Determine the First Inventor

As above, I have serious doubts that the challengers can persuade a court that there is a constitutional principle that patents must be granted to the first inventor and no other. As a predictive matter, one can add that the Supreme Court has been notably unwilling to find any kind of judicially enforceable constraint on congressional power in the area of intellectual property and has been notably enthusiastic in blessing foreign IP harmonization as a compelling government interest (every other major country operates on a first-to-file system).14 Putting all that aside, and assuming arguendo that the challengers are correct that there is a constitutionally mandated principle that only first inventors should receive patents, their argument that the first-to-file system of the AIA is unconstitutional would still have an insurmountable problem.

The problem is that there is no magical methodology by which we can be certain that someone is the first inventor of a particular invention, given that conception is a mental act that is not easily discerned. There must be some way to determine first inventorship. My core argument here is that, even assuming that awarding patents to the first inventor is a constitutionally mandated end, a first-to-file system is a permissible means for Congress to achieve that goal.15

What the challengers seem to assume is that establishing a constitutional principle that Congress should aim to grant patents only to first inventors then necessarily implies that the determination must be done through case-by-case adjudication. Moreover, they assume that the Constitution dictates that the decision-maker in such an adjudication must be permitted to look at a broad range of evidence beyond the filing date of the patent application. But there is no plausible argument, I think, that a short textual mandate that Congress should grant patents to “Inventors” comes with an elaborate set of Constitutional Rules of Inventorship Procedure and Evidence that dictate the means by which the first inventor must be determined.

If we accept that the Constitution is silent on this matter, then it almost necessarily follows that Congress has substantial discretion to craft the mechanism by which the true first inventor is determined.16 And I see no reason that Congress cannot make a legislative judgment that, as a general matter, the first person to file a patent application is the most likely to be the first inventor in fact. A case-by-case totality-of-circumstances evaluation of a subjective mental fact—when a person actually conceived an invention in her mind—is notoriously difficult and prone to fraud.17 There is no reason to assume that case-by-case totality-of-circumstances adjudications will more systematically reach the correct constitutional result—to award the patent to the actual first inventor—than an across-the-board rule that infers the first inventor from the objective fact of first filing.

A rule that gives absolute weight to a reliable objective record to prove a subjective mental fact, and thereby excludes all other evidence and eschews totality-of-circumstances evaluation, is exceedingly common in our legal system.18 For example, the theoretical goal of contract law is to enforce the parties’ actual meeting-of-the-minds bargain, but since that is not discernible, courts instead enforce the provisions of an integrated written contract as the best evidence of that mental agreement, and they exclude extrinsic evidence because it is less reliable. Many of the Federal Rules of Evidence similarly involve elevating what are considered reliable sources of evidence (e.g., the Best Evidence Rule that requires original documents) and correspondingly excluding less reliable evidence.19 The first-to-file rule in the AIA is properly conceptualized as a rule of evidence reflecting a judgment that a filed patent application is the best evidence of actual conception and all other evidence should accordingly be excluded. This legislative judgment is, to my mind, quite reasonable.20

Unless the challengers to the AIA want to argue that Congress cannot set rules of evidence for determining first inventorship, or that it can do so but cannot find the filing of a patent application to be the most reliable evidence for this inquiry, then it seems to me that there is no possible argument that a first-to-file system is unconstitutional, even if we accept for the sake of argument a constitutional principle that Congress must aim to give patents to the first inventor. There is no inconsistency between a first-to-invent theoretical principle and a first-to-file real-life system, once we understand that determining the first person to invent requires a practical mechanism involving reliable evidence.

  1. Leahy-Smith America Invents Act (AIA), Pub. L. No. 112-29, § 3(b), 125 Stat. 284, 285-87 (2011) (to be codified at 35 U.S.C. § 102).
  2. See Complaint, Madstad Eng’g, Inc. v. USPTO, No. 8:12-cv-01589 (M.D. Fla. July 18, 2012), ECF No. 1 [hereinafter Complaint].
  3. Mark A. Lemley, The Myth of the Sole Inventor, 110 Mich. L. Rev. 710, 720-22 (2012).
  4. There is a qualification to this, which is that the first person to file must still have conceived the invention using his own efforts and not through theft. See AIA § 3(i), 125 Stat. at 289 (to be codified at 35 U.S.C. § 135) (providing for derivation proceedings to adjudicate theft).
  5. Complaint, ¶ 20. See also Karen E. Simon, Comment, The Patent Reform Act’s Proposed First-to-File Standard: Needed Reform or Constitutional Blunder?, 6 J. Marshall Rev. Intell. Prop. L. 129 (2006).
  6. Complaint, ¶ 17 (arguing the first-to-file standard is unconstitutional).
  7. Cf. Margo A. Bagley, Patently Unconstitutional: The Geographical Limitation on Prior Art in a Small World, 87 Minn. L. Rev. 679 (2003) (arguing that the 1952 Act’s disfavoring of foreign prior art is unconstitutional).
  8. See Letter from Timothy Holbrook, Professor of Law, Emory Univ., et al., to Lamar Smith, Chairman, House Judiciary Comm. (June 13, 2011)(arguing these exceptions prove that “Congress has discretion in defining the term ‘inventor’”),
  9. Patent Act of 1952, ch. 950, § 1, 66 Stat. 792, 797 (codified at 35 U.S.C. § 102(g)) (denying priority to the second inventor only if “the invention was [previously] made in this country by another . . . .”) (emphasis added). This exception was partially repealed in 1999. See 35 U.S.C. § 102(g) (2006) (allowing foreign conception to suffice in an interference proceeding). Even with the repeal, the fact remains that the 1952 Patent Act allowed later domestic inventors priority over foreign first inventors for nearly 50 years, and nobody thought this was unconstitutional. Moreover, 35 U.S.C. section 104 still bars using activity in non-WTO countries to prove first invention.
  10. Patent Act of 1952, ch. 950, § 1, 66 Stat. 792, 797 (codified at 35 U.S.C. § 102(g)) (denying priority to the second inventor only if “before the applicant’s invention thereof the invention was made . . . by another who had not abandoned, suppressed, or concealed it”).
  11. For example, because we want to incentivize independent reinvention when the first inventor keeps the invention as a trade secret. W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1550 (Fed. Cir. 1983).
  12. See Patent Act of 1870, ch. 230, § 62, 16 Stat. 198, 208 (excluding evidence of foreign activity unless patented or published); Thomson v. Weston, 19 App. D.C. 373, 381 (1902) (“[B]y deliberate concealment or suppression of the knowledge of his invention he subordinates his claim . . . to that of another and bona fide inventor . . . .”).
  13. Letter from Daniel B. Rachiver, Lecturer in Law, Benjamin N. Cardozo Sch. of Law, et al., to John Boehner, Speaker, House of Representatives, and Nancy Pelosi, Minority Leader, House of Representatives (June 17, 2011),
  14. See Golan v. Holder, 132 S. Ct. 873 (2012). See also Timothy R. Holbrook, The Treaty Power and the Patent Clause: Are There Limits on the United States’ Ability to Harmonize?, 22 Cardozo Arts & Ent. L.J. 1, 6 (2004) (noting that the rest of the world uses first-to-file).
  15. To see the difference between my argument and the conventional analysis, compare Holbrook, supra note 14, at 6 (conceding that first-to-file is unconstitutional if “Inventors” means first inventors, and arguing that Congress should “redefine” the meaning of the term), and Jane C. Ginsburg, No “Sweat?” Copyright and Other Protection of Works of Information After Feist v. Rural Telephone, 92 Colum. L. Rev. 338, 376 n.180 (1992) (same), with City of Boerne v. Flores, 521 U.S. 507, 519-20 (1997) (holding that Congress cannot redefine the substantive meaning of the Fourteenth Amendment but has discretion over the means of enforcement).
  16. McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421 (1819).
  17. See, e.g., Precision Instrument Mfg. Co. v. Auto. Maintenance Mach. Co., 324 U.S. 806 (1945) (applicant fabricating evidence of inventorship).
  18. At a more fundamental level, this is a rules-versus-standards tradeoff. See generally MindGames, Inc. v. W. Publ’g Co., 218 F.3d 652, 656-57 (7th Cir. 2000) (Posner, J.) (“A rule singles out one or a few facts and makes it or them conclusive of legal liability; a standard permits consideration of all or at least most facts that are relevant to the standard’s rationale.”).
  19. Fed. R. Evid. 1002.
  20. I am not saying that Congress must use an absolute rule. Congress is free to choose a more standard-like procedure such as having a rebuttable presumption that the first filer is the first inventor, subject to case-by-case adjudication of whether the presumption is rebutted, which is the procedure under the 1952 Act. 37 C.F.R. § 41.207 (2012). Or it can even have totality-of-circumstances adjudication with no weight given to the fact of first filing at all. All I am saying is that the Inventors Clause of the Constitution does not dictate how this rules/standards choice is made. If Congress can enact the existing rebuttable presumption, it can also make the presumption absolute.