March 3, 2015


Thoughts on juries for intellectual property lawsuits

Here’s a thought that’s been stuck in my head for the past few days. It would never be practical, but it’s an interesting idea to ponder. David Robinson tells me I’m not the first one to have this idea, either, but anyway…

Consider what happens in intellectual property lawsuits, particularly concerning infringement of patents or misappropriation of trade secrets. Ultimately, a jury is being asked to rule on essential questions like whether a product meets all the limitations of a patent’s claims, or whether a given trade secret was already known to the public. How does the jury reach a verdict? They’re presented with evidence and with testimony from experts for the plaintiff and experts for the defendant. The jurors then have to sort out whose arguments they find most persuasive. (Of course, a juror who doesn’t follow the technical details could well favor an expert who they find more personable, or better able to handle the pressure of a hostile cross-examination.)

One key issue in many patent cases is the interpretation of particular words in the patent. If they’re interpreted narrowly, then the accused product doesn’t infringe, because it doesn’t have the specific required feature. Conversely, if the claims are interpreted broadly enough for the accused product to infringe the patent, then the prior art to the patent might also land within the broader scope of the claims, thus rendering the patent invalid as either anticipated by or rendered obvious by the prior art. Even though the court will construe the claims in its Markman ruling, there’s often still plenty of room for argument. How, then, does the jury sort out the breadth of the terms of a patent? Again, they watch dueling experts, dueling attorneys, and so forth, and then reach their own conclusions.

What’s missing from this game is a person having ordinary skill in the art at the time of the invention (PHOSITA). One of the jobs of an expert is to interpret the claims of a patent from the perspective of a PHOSITA. Our hypothetical PHOSITA’s perspective is also essential to understanding how obvious a patent’s invention is relative to the prior art. The problem I want to discuss today is that in most cases, nobody on the jury is a PHOSITA or anywhere close. What would happen if they were?

With a hypothetically jury of PHOSITAs, they would be better equipped to read the patent themselves and directly answer questions that are presently left for experts to argue. Does this patent actually enable a PHOSITA to build the gadget (i.e., to “practice the invention”)? Would the patent in question be obvious given a description of the prior art at the time? Or, say in a trade secret case, is the accused secret something that’s actually well-known? With a PHOSITA jury, they could reason about these questions from their own perspective. Imagine, in a software-related case, being able to put source code in front of a jury and have them be able to read it independently. This idea effectively rethinks the concept of a jury of one’s peers. What if juries on technical cases were “peers” with the technology that’s on trial? It would completely change the game.

This idea would never fly for a variety of reasons. First and foremost, good luck finding enough people with the right skill sets and lacking any conflict of interest. Even if our court system had enough data on the citizenry to be able to identify suitable jury candidates (oh, the privacy concerns!), some courts’ jurisdictions simply don’t have enough citizens with the necessary skills and lack of conflicts. What would you do? Move the lawsuit to a different jurisdiction? How many parts of the country have a critical mass of engineers/scientists with the necessary skills? Furthermore, a lot of the wrangling in a lawsuit boils down to controlling what information is and is not presented to the jury. If the jury shows up with their own knowledge, they may reach their own conclusions based on that knowledge, and that’s something that many lawyers and courts would find undesirable because they couldn’t control it.

Related discussion shows up in a recent blog post by Julian Sanchez and a followup by Eric Rescorla. Sanchez’s thesis is that it’s much easier to make a scientific argument that sounds plausible, while being completely bogus, than it is to refute such a argument, because the refutation could well require building up an explanation of the relevant scientific background. He’s talking about climate change scientists vs. deniers or about biologists refuting “intelligent design” advocates, but the core of the argument is perfectly applicable here. A PHOSITA jury would have a better chance of seeing through bogus arguments and consequently they would be more likely to reach a sound verdict.


  1. avatar BlakeReid says:

    …but it breaks down in practice for the following reasons:

    a) Defining the PHOSITA to match real people is problematic. It’s one thing for the court to define a hypothetical PHOSITA, but another to define a real-world class of jurors.
    b) Depending on the subject matter, there may not be 12 PHOSITAs available in a particular venue (or anywhere, for that matter).
    c) Even if there are enough PHOSITAs in a particular venue, they may be impossible to find (as you mention).
    d) Even if they can be found, it might pose an unreasonable burden to repeatedly call them in for jury trials.
    e) Even if not, there are likely to be serious conflicts of interest in a tightly knit industry (i.e., the jurors may know the plaintiffs and/or defendants, be their coworkers, work for their competitors, etc.)

    Finally, even if such a system could be worked out, I’m not sure what problem it would solve. A very small percentage of patent cases actually reach juries, mostly because patent litigation is so expensive. PHOSITA juries wouldn’t do anything to change that.

    • Finally, even if such a system could be worked out, I’m not sure what problem it would solve. A very small percentage of patent cases actually reach juries, mostly because patent litigation is so expensive. PHOSITA juries wouldn’t do anything to change that.

      The fact that so few patent cases reach juries actually works in favor of trying to get some sort of PHOSITA jury. You wouldn’t have to do it as often as you assemble juries for, to pick an example, criminal cases.

      The problem they’d solve is that such juries would be better equipped to consider the evidence and reach sound conclusions. Such a jury could be presented with technical arguments having more depth, when appropriate. In the end, I would imagine that such a jury wouldn’t necessarily be more inclined to favor plaintiffs or defendants, but rather would be more inclined to favor the side with the stronger technical arguments in its case.

      Just consider obviousness. A jury of PHOSITAs could look into their own experience to say whether an idea is or isn’t obvious. (“Yeah, that’s how I’d do it.”) Present-day juries can’t do that at all.

      • avatar BlakeReid says:

        It’d still have to happen often enough that it’d be an administrability nightmare. Also, I don’t think PHOSITA juries would necessarily be more even-handed than regular jurors (who empirically seem to favor whoever files suit first, i.e., the patentee in an infringement suit or the potential infringer in a declaratory judgment suit). I imagine a pharmaceutical PHOSITA jury would be biased toward patentees, while a software jury would be slanted against.

        Finally, your last point is a strong argument against PHOSITA juries. Obviousness and anticipation are issues of law for good reason – witness any Slashdot discussion about a software patent. Obviousness must be deliberately determined from the time of invention, not the present, to avoid hindsight bias. Whether judges are effective in doing that is another matter, but even your off-the-cuff hypo (“Yeah, that’s how I’d do it”) suggests that juries would be unlikely to properly consider the issue.

  2. There may be a middle ground there somewhere. Juries are currently permitted to ask ‘legal’ questions of the court, so why not technical ones?

    Suppose the Court obtained/appointed one or more licensed PHOSITA ‘technical representatives’, certified as having no vested financial interest in the outcome, and allow the Jury equal access to each expert to conduct their own question/answer sessions? During deliberation the Jury members could ask any specific set of technical questions of any appointed experts, independently, with the experts being only permitted to answer those specific technical questions, and thereby refrained from a soap box exhibition on behalf of either side of the case. The Jury would then be enabled to understand the technical issues at hand and make a much better and informed decision.

    The trick here is to make sure that ‘the experts’ are non-biased at the start of the case and do not benefit by the outcome of the trial. Having them certified and employed by the Court System rather than the Lawyers/clients would keep them under control of the courts and not there just for personal gain. The cost of the experts could be from a pool of money from both clients, but the selection and payments must be controlled and monitored by the courts to keep things on the level.

    I have no idea how this would fly in the real world, but its a thought.

  3. My understanding from what patent attorneys say is that “ordinary skill in the art” is itself something of a misnomer. One the one hand, you often need someone seriously specialized to be able to put the patent into practice, but on the other, you need someone not much past their apprenticeship to make the invention non-obvious. So trying to get a jury of same might be difficult because the term itself is difficult, not because of a lack of people who might fit a reasonable bill.

    As for venues, don’t patent lawyers pretty much troll a handful of venues as it is?

  4. I agree with your assessments of the situation: both that experts testifying about hypothetical PHOSITAs to juries of laymen is a poor way to run a patent case, and that it’s very impractical to change that.
    Which is why I think that reform of our patent system should focus on reducing the amount of litigation that’s necessary by narrowing and clarifying the scope of what’s patentable. This doesn’t necessarily have to change the legal standards directly — programs like Peer to Patent have the potential to improve the legitimacy of the patent system at-large. It seems like one of the biggest problems right now is that a stamp of approval from the USPTO doesn’t mean nearly as much as it should. No one really knows if their patent is valid until it’s been litigated, which shifts the responsibility of considering the PHOSITA standard from patent officers to laymen juries.

  5. avatar Mitch Golden says:

    I guess my question regarding this idea is this: what is so special about patents? Any situation could call for people with particular expertise.

    For example, someone I know was involved in a malpractice case against a dentist. What the dentist had done was clearly outside of the bounds of normal practice. But the trial devolved into precisely what you described: dueling experts – two for the insurance company and two for the plaintiff. Had the jury consisted of just people with dental expertise they would have found for the plaintiff, but because they were bored and confused it went the other way.

    And it could even be less technical than that: I don’t know the first thing about gangs, but if I were asked to sit on a jury I’d be asked to assess whether a claim about one was reasonable.

    It’s just the way the jury system works…

    • You make precisely the point I’m trying to make. If the facts are that the dentist screwed up, then other dentists would clearly see that. Maybe you could say that other dentists from the local area might be buddies with the defendant and thus shouldn’t be on the jury, but it’s certainly the case that a jury of dentists would have an easier time sorting out the dueling experts’ opinions.

      The gang example is also interesting. On the one hand, jurors who live in gang-controlled areas will have more familiarity with how gangs behave and could more rationally evaluate claims as to whether a gang member would or would not behave in a particular fashion. On the other hand, jurors from such areas might have conflicts because they fear retribution from the gang or they may even have an anti-police bias.

      It’s just the way the jury system works…
      Yes, but should the jury system work that way, or could it work differently?

    • I actually really like this idea. While I may be way too optimistic I think a version of it could actually work.

      The trick is to be just a little more selective about choosing who sits on a jury. Today the jury is pretty much made up of 12 random people picked off the street. Which is why need “experts” so frequently. They need to tell the jury what is “common sense” in this case. So instead of getting random people, you get people in related fields. They have most of the “common sense” necessary. In the dentist case you don’t need twelve dentists, you just need twelve people involved in medic where they be brain surgeons or veterinary assistants. They all have some of the same “common sense” for the medical field. Stuff like giving shots, sterilizing things, dealing with medical charts and the like.

      So for patent cases you wouldn’t need people that could actually build the device that was patented maybe just people that could identify it. Many people couldn’t tell the difference between an car engine and a bunch of dirty hoses, wire and something heavy taped together. Yet at least %10 of the population can. If you have a patent case about engines you want to draw from that %10. To get this group you would have the potential jurists answer some simple questions related to the subject matter. For example:

      Do you know how to give someone a shot?

      Construction or architecture:
      Which is heavier steal or aluminum?

      Name as many fluids as you can that are necessary for a car to run.

      What would this program print to the screen?

      print "hello"
      x = 1
      while x < 10 {
      print x
      x = x + 1

  6. One challenge, even if you could get twelve PHOSITAs on a jury, is that what matters is what a PHOSITA would have known at the time of the invention.

    Imagine a patent lawsuit in 2009, over an invention made (say) back in 1999. The PHOSITA jurors would have to pretend that they don’t know anything that they have learned since 1999 — not an easy thing to do. Worse yet, they have to pretend that they don’t know anything about the patent that is the subject of the trial — also not easy to do, if you are hearing days of testimony about that very patent. In short, they have to ask what would be obvious to hypothetical past versions of themselves.

    What you really want is to get hold of a time machine, go back in time to 1999, kidnap twelve PHOSITAs and time-travel them back to the present to act as jurors. (Don’t forget to sequester them!)

    • Maybe this is the real reason that Skynet invented time travel.

      • avatar BlakeReid says:

        1. Invent time travel.
        2. Go back in time and patent time travel.
        3. (There is no step 3.)
        4. Profit!

        • Here’s an improvement on Blake’s method:

          1. wait for somebody else to invent time travel
          2. go back in time and patent time travel
          3. cackle villainously
          4. profit!