« The Ikea Effect and Locke's Theory of Property | Main | The Sixth Circuit's ruling on Affirmative Action & Holmes' "Prophecy" Theory of the Law »

Saturday, July 02, 2011

The Constitutionality of First-to-File

There has been a big debate brewing in the patent law community about the constitutionality of “first-to-file,” with rival camps of law professors sending letters to Congress arguing for and against this issue.

As a bit of background, the U.S. has traditionally used a so-called “first-to-invent” patent system.  That is, when two people independently invent the same thing (which happens quite frequently, e.g. Edison and Swan with the light bulb), the person who gets the patent is supposed to be the one who first conceives the idea.  The rest of the world uses a “first-to-file” system, where the patent goes to the first person to file in the patent office.  Congress is considering a pending bill that would change the U.S. system to a first-to-file rule.

The constitutional argument is that the Constitution only authorizes patents for “inventors,” and those who argue that changing to first-to-file is unconstitutional basically argue that "inventor" means the first inventor as a matter of historical meaning and practice.  I don’t have very strong views on the history.  But I do have rather severe doubts about the argument.

The reason is that the opponents of first-to-file have never argued the current system is unconstitutional.  And while the current system in theory gives the patent to the first inventor, in practice it has an extremely heavy presumption that the first person to file is the first inventor and will obtain the patent.  It seems rather odd to say that it is constitutional for Congress to create a presumption so heavy that it is de facto impossible to rebut; but cannot go ahead and make the presumption irrebuttable.

Posted by Tun-Jen Chiang on July 2, 2011 at 02:35 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference The Constitutionality of First-to-File:



I partially agree with your assessment. For me, if one is first to invent (inventor A), but decides to hide the invention by trade secret, then they should loose the ability to patent the invention. If later on, another person invents the same thing (inventor B), but decides to patent it, then he should be able to get that patent even if he was not the first inventor.

The constitutional argument does not make sense to me because it seems to argue that inventor A (who deprives the public of an invention) can get a patent later on to reap the benefits of a patent only when someone else tries to get a patent on the same subject matter. It seems to me there could be gamesmanship involved to get the most out of your monopoly by first using trade secret and then the patent system only when someone else tries to patent that invention.

I realize that 102(g) states that you cannot get a patent if you abandoned, suppressed or concealed the invention. But from what I understand 102(g) will be removed or dramatically altered in the new reform bill.

Posted by: Shine Tu | Jul 7, 2011 3:48:41 PM


They both sound horrible.

Posted by: Praetor | Jul 3, 2011 3:11:24 PM

Praetor, if you mean that the first independent inventor can lose because of the timing of getting to the patent office, that is accurate. And in one sense it is arbitrary.

But we have lots of "arbitrary" rules on when you have to get to a government office to file your paperwork, on pain of losing something. A race-notice recording statute makes the first buyer of property lose their property by virtue of getting to the recording office too late.

More directly, my point in the post is that this "arbitrary" rule applies on the ground in both a formal first-to-file system and our system. In our current system, the first independent inventor is saddled with an almost-irrebuttable presumption merely because he filed too late. In a formal first-to-file system, he is saddled with a completely-irrebuttable presumption. If you think that the latter makes no sense, I fail to see how you can find the former significantly better.

Posted by: TJ | Jul 3, 2011 2:02:24 PM

Is the only difference that in first-to-file the first independent inventor can lose because of timing? That sounds arbitrary.

Posted by: Praetor | Jul 3, 2011 1:56:14 PM

Praetor, that is orthogonal to the debate, in the sense that the theft issue is treated the same under either system. In the current system, Claimant A who files first is still presumed to have invented independently. Claimant B can argue that Claimant A stole the invention from him, and Claimant B would get the patent if he proved the allegation. But Claimant B would bear the burden of proof.

Posted by: TJ | Jul 3, 2011 1:04:58 PM

I think my problem with first-to-file is that you never know whether both invented independently. It would nearly impossible to prove your idea was stolen if the first filer, who is a thief, has a presumption that he did not steal.

Posted by: Praetor | Jul 3, 2011 12:54:47 PM

Praetor, in case my post was not clear, the difference between the two system only occurs when each person who files independently invents the same thing. If Inventor A invents the light bulb, and Inventor B steals the idea in the sense of breaking into A's lab in the dark of night, then no system would give the patent to B.

Posted by: TJ | Jul 2, 2011 10:59:50 PM

As someone who rarely makes arguments in advance of fairness, it would seem that if you conceive of the idea, others should not be able to steal your idea and reap rewards that otherwise would have redounded to you.

Posted by: Praetor | Jul 2, 2011 4:47:48 PM

Post a comment