Fred Wilson links to his partner Brad Burnham’s post, "We need an independent invention defense to minimize the damage of aggressive patent trolls:"

I know of no case where the engineers in one of our companies were aware of the patents that are now being used to attack them. The moral rightness of this screams at me. If, as an engineer focused on solving a problem, I happened to come up with an idea that is in some way similar to yours, then that in itself should suggest that it was obvious and not patentable. Unfortunately, that does not really help. There, the burden of proof is still on the startup and it is still smarter to settle than to burn precious capital on a defense.

If, on the other hand, the troll was required to show the startup had some prior knowledge of their technology, the burden would be shifted to the attacker, and this blatant abuse would come to a grinding halt. If you believe as I do that innovation is key to social progress, please support patent reform. It is a complicated issue, but an independent invention defense is an obvious place to start.

(Emphasis mine; keep reading to see why.)

Though I sympathize with Brad’s concerns — patent infringement litigation is both high stakes and notoriously expensive, and thus risky and burdensome even for defendants likely to prevail at trial — I have a couple issues with an independent invention defense. 

First, the defense already exists to some extent in the form of differing damages for "infringement" compared to "willful infringement." If the plaintiff cannot prove at least "objective recklessness" — which is quite hard to do in the wake of In Re Seagate Technology, since the defendant has no affirmative duty to avoid infringement — then the plaintiff cannot recover treble damages or attorneys’ fees. The stakes are thus lower in genuine "independent invention" cases.

Second, an independent invention defense would discourage individuals and businesses from doing an adequate patent search before investing resources into novel solutions. One of the primary reasons we have a public patenting process (rather than merely protection of private trade secrets) is to make inventions easily available to the public for use. How many times would the wheel have been reinvented if it had been kept secret? Though frustrating to businesses, from a societal standpoint patent licensing is generally preferable to the redundant investment of time, effort and money into solving problems with known solutions.

Third, an independent invention defense would be ripe for abuse. Independent invention is already a defense to to a willful patent infringement claim; making independent innovation a complete defense would give defendants an even greater incentive to manufacture "evidence" showing their "independent" invention. Worse, genuine invention is often quite messy; the independent invention defense could thus perversely protect only those defendants who from the start knew to create a trail of "evidence."

That’s not to say we don’t have problems with our patent system. We do. I just don’t think an independent invention defense is the way to go.

So let’s talk about the part of Brad’s post I emphasized: obviousness and the burden of proof.

A recurring theme in the comments to Brad’s post and Fred’s post is the complaint that many patented "inventions," particularly in the Silicon Valley industries, are not particularly inventive. Patent law is supposed to guard against this problem. Indeed, the most common defense in patent suits is a counterclaim by the defendant that the patent is invalid because it is too obvious.

35 U.S.C. § 103 forbids patents where "the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains." The Supreme Court has explained the analysis as such:

Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented.
Graham v. John Deere Co. of Kansas City, 383 U. S. 1, 17–18 (1966); see also KSR International v. Teleflex (2007)(quoting Graham). Though the law is in flux, apparently the Supreme Court now believes — despite prior precedent (i.e., Graham and KSR) holding otherwise — that "obviousness" is a factual question for the jury to decideWhether "obviousness" is a question of law for the court or a question of fact for the jury is important, but neither answer would fundamentally alter the dynamics of patent litigation.


More pertinent here is how a finding of "obviousness" is the only tool courts have to mitigate the high stakes and high expense of patent infringement cases with questionable, but not outright dubious, merit.

Finding "obviousness" as a matter of law is a nuclear option; it requires the court dismiss the case, simultaneously creating precedent barring the defendant from any future litigation on the patent in the future and creating a basis for any current licensees to stop payment to the defendant. Courts’ hesitation to use that nuclear option — particularly given the standards governing summary judgment and courts’ unfamiliarity with the state of innovation in highly technical industries — is understandable.

So what to do?

Change the burdens.

It’s not a new idea; employment-discrimination cases routinely apply the McDonnell-Douglas burden-shifting framework at summary judgment.

Right now, in terms of "obviousness," a court has two options:

  1. Find the patent "obvious" as a matter of law, thereby blowing up the case and invalidating the patent;
  2. Leave "obviousness" up to the jury, where the defendant has the burden of proving the patent is "obvious."

As Brad complains, #2 may be a hollow remedy.

So why not add another option? Why not allow the court to put the burden of nonobviousness upon the plaintiff if the defendant shows, pre-trial, a "cogent and compelling" argument that the patent was obvious? ("Congent and compelling" isn’t foreign to courts either; such a showing is required in securities fraud cases under Tellabs.)

A court could further mitigate the pre-trial risk to the defendant by ordering bifurcation of the trial; i.e., first there’s a trial on "obviousness," and then, if the patent is "nonobvious," there’s a trial on infringement. Adding some element of fee-shifting — e.g., a plaintiff who loses the "obviousness" trial has to pay the trial fees of the defendant — would create a market for contingent fee defense of patent infringement suits, and thereby mitigate the "burn[ing] of precious capital …"

Just a thought.