 You're watching FJTN, the Federal Judicial Television Network. The Federal Judicial Center presents Science in the Court Room, a series of programs for judges on science and scientific evidence. Program 3, Markman Issues in Biotech Patent Cases, a discussion moderated by the Honorable Fern M. Smith, Director of the Federal Judicial Center. Hello, and thank you for joining us for the third program in our series, Science in the Court Room. In this program, we'll address a number of thorny legal and practical issues resulting from the Supreme Court's holding in Markman versus Westview Instruments. We'll discuss these issues in the context of biotech patent cases and perhaps learn a bit more about microbiology along the way. But much of what we say will be applicable to the entire range of patent cases. Joining me for today's discussion are Judge Paul Michele of the Federal Circuit, Attorney Leora Ben-Ami, a partner with the firm Clifford Chance, Rodgers & Wells in New York, Judge Patty Saras of the District of Massachusetts, and Judge Roderick McKelvey of the District of Delaware. Let's begin with you, Judge Michele, if we may. And would you explain for us generally the Supreme Court's holding in Markman and what this does to the responsibility of the trial courts? Well, the court held that construing patent claims, much like construing statutory language, is a question of law as for the judge, not the jury, and is independently or freely reviewable on appeal. So it put a great burden on judges, particularly since patent claims have to be construed from the standpoint of people in the particular art or technological field. And of course, then the jury needs to be instructed by the court as to the meaning of the disputed claim terms. Okay, and Ms. Ben-Ami, you've litigated a number of these cases, and while the Supreme Court made clear the legal obligation for the trial court, it really didn't address the procedural issues at all. What have you found in going around the country that courts are doing in terms of scheduling and procedural matters? Honestly, they're all over the lot. Honestly, okay. And it can be helpful to understand what people are doing. Some judges do Markman hearings in the middle of discovery. Some do them at the end of discovery. There are some judges who do them as part of a summary judgment, and some do them as part of the pretrial. Still other judges will do them at the charging conference right before the jury comes in to deliberate. Some judges will do them purely on the papers. Others will have hearings, like a summary judgment hearing, and still others have an evidentiary hearing, like a mini-trial, so it really is very, very different depending on where you are. Okay, now where you are, Judge Saris, is in Massachusetts, and you've had a fair number of patent cases, what do you do in terms of scheduling, Markman hearings, and what kind of timing do you think is appropriate? Well, the first place to address the timing is at the Rule 16 Scheduling Conference, and I pull the attorneys in and I say, what about Markman? Do we have issues of claim construction? And often they say, we're not sure, or we're not sure how many claims, so they come in with lots of claims and then they can boil it down. So we sit down right there at the Rule 16 Conference and schedule in the Markman hearing. I find in general, the time it works the best is at the end of discovery, at the end of fact and expert discovery, is to schedule that Markman hearing in conjunction with the summary judgment motions. In that way, I have a sense of the big picture, what are the defenses, what are the issues here, the attorneys have really focused so they can boil down the number of claims that are actually in dispute, and at that point it's far enough before trial so that I can figure out the difficult technology and the parties can then prepare for trial in light of what I've ruled. The only downside is you sometimes need to have the summary judgment motions re-briefed because if they're like ships moving in the night, they essentially are not addressing the claim as you've construed it. So sometimes you need some re-briefing, but you need to give yourself enough time to be able to write the Markman opinion and explain how you came to the construction that you got, particularly if it's a hotly disputed claim. Okay, and do you find that the procedure you've just described works equally well for non-biotech cases as far as timing? Well, actually it's amazing to me. Every once in a while I get a patent that I can actually understand without any help and that's some basic straightforward technology. And sometimes even there, there's an issue of claim construction and the attorneys will say just settle this, just figure out what this claim means and we can maybe settle this dispute. So there I'll move it forward. Now try and do it earlier on. But the beauty of waiting on a very difficult case like a patent case or a computer case or something like that is that way I can learn the technology. I'm the one who only had biology my freshman year in college. I haven't done a lot of the science in a really long time. So I want to make sure I can get up to speed in the technology. Judge McKelvie, probably no court has a higher percentage of patent cases than the District of Delaware and you in particular have had a lot and I know your court has some established procedural rules. Do you agree generally with Judge Saras on timing or do you have a different point of view? I actually do mine a little bit later. I do my claim construction now. I set aside a day where I do pretrial conference in the morning, claim construction in the afternoon and then here's summary judgment arguments after that and that's all about four to six weeks before the jury trial. So it's probably a couple of weeks later in the process than Judge Saras does it. I tried it during the trial, found that very hard on both the judge and the lawyers hard because I want to take the time to think about claim construction and read the file history and I found that just hard to do. Tried to do claim construction earlier in the trials and found that also very hard to do earlier in the case meaning in the middle of discovery. I found it just too difficult to do. I didn't understand enough about the case. So I've settled in right now on cases including in biotech cases, trying to do it all in one day at the claim construction. And as I understand at the pretrial and... Okay and as I understand at Delaware tries very hard and it's pretty consistent about setting the trial date within 12 months after a case is filed. We try, there's a little slippage but we try to schedule a case so that it gets to trial within 12, 14, 15 months of the filing of the complaint. Okay. Ms. Ben-Ami do you find that the general procedures discussed by judges McKelvie and Saras are fairly universal or what do you see around the country generally in terms especially in differentiating between biotech cases and non biotech cases? I think in biotech cases the judges need more help. You need to get more background because as Judge Saras said most judges haven't had a lot of biology and it was quite a while ago. So people don't have the background to feel comfortable. And what I like to say is there are various ways to do Markman hearings but there needs to be what I call a period of intensity where you can get that tutorial. Judge McKelvie has tutorial tapes made by the parties where they explain the issues or a tutorial live tutorial or some means so that the judge feels comfortable that they're just not acting in a vacuum. I think that's very important for the courts. Do you have a lot of judges now requesting tutorials either live or by tape? Almost everyone now. Good. Well, I can be currently we're all learning something as we go along. Judge Michelle unfortunately a number of these cases wind up at your court sooner or later. All of them. And as you're reviewing the record do you notice any difference that timing makes in the dispute or do you have a sense of what might be optimal? I agree with what the judges here have said that after a fair amount of discovery but well before trial is the optimal zone. I think it varies some from one case to the next. That's probably why we didn't try to stipulate any procedures including on timing. And I think it's very interesting the stress put on getting comfort with a period of intensity and then also doing the analysis carefully in writing because when it does come to us on appeal which most cases do claim construction is always one of the issues. And even though the nominal standard of review is de novo we really don't start from scratch and construe the claims ourselves. We review the claim construction reasoning as explained by the district judge in the opinion. So having a clear line of reasoning there is enormously helpful. Do you get any sense in looking at these records as to what proportion of judges are doing Markman hearings at any given time early late? I think that it seems that almost everybody somehow connects it with summary judgment. We now see a great many cases where there's a grant of summary judgment and hence a pretrial appeal. So there is an opportunity requirement really for us to review the correctness of the claim construction before the case goes to trial. And so it seems like everybody has it pegged pretty close to summary judgment. There are lawyers who pressed to have claim construction done early in a case. Frequently a defendant will ask to have it done early. And in-house counsel will say that it can save a lot of money if we'll do claim construction early. I'll ask lawyers if they want to do claim construction early. If they both agree, we'll try to do it early. Sometimes it helps in discovery disputes understanding what the claims will mean. What do you do if one party gives you a pretty convincing argument that an early claim construction would be useful and the other party adamantly protests? Do you go with your feeling that it makes sense? I'll stay with my pretrial conference time. It happens that defendants will come in and say, you don't need discovery on the infringing product or process or tell us what the claim constructions mean first. And I'll allow a plaintiff to take discovery on the infringing product or process before we do claim construction. So my default mode is to stick with the pretrial conference. Okay. Judge Michelle, do you see many cases where the judge has adopted neither party's proffered claim construction but instead comes up with his or own version of what the claims mean? We see some of that. Usually it's more partial. There may be certain points on which one side's proposed construction was adopted by the trial judge, maybe on a few points the other side and once in a while a sort of a third approach by the trial judge themselves. But usually it's fairly clear cut on the critical claim terms that really drive the infringement summary judgment decision. As a lawyer, Ms. Binami, how would you feel if you weren't going to get the claim construction until during the trial and then the judge appeared on the bench and announced that the claim construction in fact was not what you'd asked for nor what the other side had asked for but was a whole new creature? You can imagine how it feels. The truth is it just leads to confusion. One of the goals of both the judges and the lawyers there is a consistency here. We want to try to keep things simple for the jury to the extent we can and when we have to argue the technology and the alternative, how the claims meet the products in the alternative, it just leads to a great deal of uncertainty for the jury. Experts will have to talk about things in the alternative and then if at the end the judge picks which claim construction he prefers, the judge may have been deciding the credibility of the experts and the jury finds out. So it just creates a lot of extra problems I don't think are particularly useful for anybody in this situation. Well, I know you've had cases before Judge McKelvie, before Judge Saris, before me when I was in San Francisco and before a lot of other judges given the fact that judges are all over the lot from the standpoint of the litigants and the lawyers, what do you think is the maximum time or the best time to hold a hearing? I think that for me, I would like it a little bit earlier than the pretrial because I'd like to be able to refocus my expert reports and I have a case in front of Judge McKelvie, so I'll be happy to do it at the pretrial also. But I think that if it's a little bit before the pretrial, it allows the parties to figure out if they need to do anything else in light of the claim construction. But it has to be well into the case unless it's an unusual case. Well, one of the advantages of the pretrial is you get this refocusing and it ends up being what I call a whack-a-mole approach which is that once the judge does claim construction, the parties will refocus and you'll do new claim construction and new claim construction. So one of the advantages of doing it late but before the trial is then the lawyers can't change their position and it forces the lawyers and the experts to come in with expert reports to deal with all of the potential claim construction so there's not a lot of shifting at ground afterwards. So this is a little bit of a benefit to the judge and a detriment to the lawyers in terms of timing. Do you allow the theory of shifting sands and claim construction? Of course I do because I want to get it right, Judge Michelle. As a trial judge, you say you want to get it right and if a party comes back and shifts, it's difficult because you can't just put it aside but I think the best thing to do is to force the lawyers to deal with what the different options are going to be and then do a claim construction decision. Now, one of the complications is you can pick one from A, one from B, one from C and then the lawyers will be a little unclear about it. Okay, briefly Judge Saras, what do you want to conclude with? Well, I think from a judge's point of view, the problem with Markman is it really has turned potentially a summary judgment hearing into two separate hearings. You have the Markman hearing, you construe the claim and then there's a summary judgment hearing and there's the horror that we all face of seeing these huge briefs and having to write two separate opinions which takes a long time. So I think our goal should be to try and condense as much as possible the Markman into at least the summary judgment if not the pretrial so that you're not delaying by three and six months the time it takes to write each one of the opinions and that's a hard thing to do. Well, that's helpful Judge Saras. We've come to the end of this segment. We'll be back in a moment. Judge McKelvie, Judge Saras has told us she doesn't have a science background. Do you? No, actually in economics in my opinion. Okay, so how do you prepare for one of these hearings once you've decided when to hold them? What do you do to get ready? Well, first I don't set a particular structure for the lawyers. I suggest that they reach an agreement on briefing and filing papers before the claim construction hearing. Then I ask them for a couple of papers. As I mentioned, I have the pretrial order already. I ask the lawyers to give me a copy of the file history so I can read it. I ask the lawyers to give me a videotape early on in the case so that I have a sense of what the technology is during the course of discovery. And then I also have the briefing on the summary judgment motions and so those documents all together, I'll sit down a couple of days before the pretrial conference go through them. And they pretty much get me to the point where I think I understand what's going on with the technology and prepared for the hearing that it will take place using the afternoon of the pretrial conference. Do you give the parties any parameters about the videotape, for example? Do you tell them you want animation? You want a lecture? Or do you leave it up to them? A couple of things to it. One, I include in my initial scheduling order the requirement that the properties prepare and file. We'll talk during the initial conference about it. Some parties are a little reluctant because it's expensive to prepare. I basically tell them I use this to educate myself, educate law clerks. I use it for a transition for law clerks each year that come in in August or September to find out what's going on in the cases. So that the parameter basically is if you think it's going to be helpful for me, that's fine. Usually they're filed under seal and I treat them as if they're a brief. It can be a lawyer talking. It can be animation. Generally, the things that are most helpful for me are the pictures. I find that the pictures help educate me is lawyers draw pictures about the technology. It brings me up to speed much more quickly than trying to pour through the briefs. Or at least it helps me pour through the briefs. Okay. And Judge Saras, I know you and Ms. Banami had a case together. What kinds of things did you ask her and opposing counsel to do to help you get ready in that case? Well, I knew I was in trouble when the first thing that the parties gave me were two huge textbooks being used at Harvard Medical School. And I said to them, I'm the one who went to law school, not med school. And I took a read through the summary judgment motions at some point and the motions preliminary injunction. I said, I don't really understand this. I just fundamentally don't understand the technology. So what I asked the parties to do and they were terrific at doing it, was to teach me. And I found it most useful to be taught as like having a private tutorial. They each brought in professors from some of our area schools and some of them were undergraduates. Not necessarily postdoc undergraduate professors like there was a professor from Wellesley College. There was one from MIT and they came in and they taught me. They started at the beginning, DNA, and they worked all the way through. And I found that that gave me the tools to even understand the legal arguments and to understand the technology because after all it was cutting-edge, Nobel Prize-winning technology. So the other thing is I can't emphasize enough this use of video tapes because I forget things. You're onto a criminal case. You're onto a securities broadcast. You're onto other things and you forget when you're writing. And I would sit there just replaying and rewinding the tapes. You know, what's the technology? What is the issue here? And so that a month down the road while you're writing it up, if you say, gee, I don't really understand that, you can go back. So I think it's important of the video tapes. I love being taught in court and being able to ask questions. And finally, I found it was useful. I actually went back to those Harvard Medical School treatises just to understand basic term definitions. And I found if you have that, we all know how to learn and they can teach you how to learn. And that was an important piece for me in that case. We'll never get those books back, will we? I'm still hoping one of my kids goes to med school. Ms. Ben-Ami, is that the procedure you tend to follow in educating most judges? I think there are a few ways that it's done right now. Oftentimes, judges will say, come together and find a mutual expert just to give us a tutorial. But then it's a pure tutorial. Other times, they'll say each one of you bring in an expert and each one can give a tutorial. That'll be more focused on what the issues are in a case because a pure tutorial will not necessarily say, and this is where the dispute is. A focused tutorial might be helpful. Other times, the lawyers will do a tutorial. It's our job to explain the technology to the jury ultimately and then lawyers should be able to explain the technology to the judges. But one way or the other, I say most judges want a tutorial. And are they typically on the record, off the record? Both ways? Both ways. Both ways. People are doing it all different ways. Okay. Judge Michelle, when you see these cases coming up to you from your perspective, I mean, you've heard what Judge Saras and Judge McKelvie do to get ready. And obviously, they look at a variety of extrinsic material and intrinsic material. What do you think the trial judge ought to be looking at and how should the trial judges be using those materials? Well, first, I think to get the background about the invention and the technology itself, the trial judge should make use of any sort of source that seems useful. Expert witnesses, textbooks, anything. With respect, however, to construing the actual disputed words and phrases in the claim that are likely to drive the infringement finding, they have to be a little more careful. There's a cardinal rule that you cannot contradict clear meaning in a patent claim on the basis, for example, of the testimony of an expert witness. And there is also a sort of a hierarchy of preferred sources where the claim at issue is perhaps first on the list and then next the surrounding claims and next the written description and then the drawings and then the prosecution history and maybe cited prior art and so on down until you finally get to the expert witness who came into the case long after the patent issued. And that, of course, has the problem that competitors can only look to the patent to know what's forbidden and what's permitted and therefore testimony that arose only in the course of litigation wasn't available to warn anybody and so that's the least preferred source. But if none of the other sources will adequately illuminate the understanding in the relevant art at the relevant time of a particular term then as a last resort the testimony is perfectly permissible provided it doesn't contradict clear meaning discernible from the patent documents itself. Okay, all right, thank you. And Judge McKelvie, within that framework, what do you do to get inside into the specific disputed terms and how do you winnow the large number or what appears to be the large number of disputes down to something that's manageable? Actually, the lawyers end up doing that. They'll arrive with the case early on. It may not have a sense of how many claims are going to be an issue. I do a couple of things in terms of trial management. One, just in terms of getting the case to the jury, I'll force the lawyers if I can to eliminate certain patents or certain claims so that we have a certain representative group if there seems to be too many. Second, I found that in the Markman briefing, that is the briefing before the hearing that the lawyers will find out that they don't disagree on certain terms. So I've even had, during the course of Markman hearings, lawyers say, oh, I didn't understand it. We agreed on this term or those terms and then they'll essentially get it down to two, three, four terms that the judge will have to resolve the dispute between the parties. Okay, and do you hold your tutorials, I guess? I do videos early. But do you have a separate tutorial segment and if so, do you have it on or off the record? No, I do. I received the videos early on in the case. And then for the Markman hearing, I just tell the lawyers, you've got two hours. You can put on what you like. I have found that since Vytronics and Pitney Bowes, the two federal circuit cases, lawyers have been very reluctant to put on extrinsic evidence. But I invite them to put on evidence about the infringing product or process, bring in expert witnesses if they'd like to. Lately, I've found that what's happening is the lawyers are coming in and presenting argument. They do, in essence, the tutorial on the technology, the tutorial to get me up to speed on what the terms are that are in dispute and how they believe I should decide on one way or the other. Okay. It's rare now that I get witnesses during Markman hearings. You know, it fascinates me that neither of you have a particularly scientific background nor do I nor do the majority of judges on the federal circuit. So we have to do a similar learning experience for purposes of claim construction to what you've done in the pretrial phase. And we depend primarily on the lawyers and on the district judge's opinion as opposed to more technical or scientific treatises or things of that sort. That's very interesting because I think a lot of people assume that virtually everybody on the federal circuit has a PhD in some very esoteric science. So it's comforting to know that you have to struggle the same way we all do. Judge Serres, do you find, as Judge McKelvie does, that there are fewer and fewer cases in which testimony is being offered or are your cases different in some way? Well, the area in which testimony is being offered is where there's a dispute as there often is as to what people of ordinary skill and the art at the time of the invention would have construed a claim to mean. And that's the toughest area, I find, in general in the Markman hearings because sometimes you have wonderful experts on each end. We sometimes call them bookends and they just happen to disagree. They're not hired guns. They disagree as to what 1979 eight-term meant. And to give another example from this one, biotech case, there was a dispute about what the term gene fragment meant. And it was not a defined term in the patent. The literature cited in the patent and the pictures gave an ambiguous picture as to what exactly it meant in my opinion. So I had to go to extrinsic evidence which is a less preferred kind of evidence. And it was actually a dispute between two leading people as to whether it meant a single-stranded DNA molecule or a double-stranded DNA molecule which was important to one of the main issues in the case. I had hired a court-appointed neutral expert, a woman whom both sides agreed on and we did it on the record to help me through it. And essentially what I said to each of them was I didn't just want their opinion as to what it meant but it was a show me. Show me where... Give me an article where they used it as a single-stranded or a double-stranded. Help me through this as to not just what you understood it to mean but what did the community understand it to mean? And that's the toughest area. I think you do often need to go to extrinsic but it's got an unclear status in these Markman hearings. It does and maybe Judge Michelle you can help clarify some of it. You wrote the Vitronic's opinion which is widely cited and then we're also involved in some of these other cases. Would you tell us what the current status is if you can on intrinsic versus extrinsic? I wrote not only Vitronics but also a follow-on case called Pitney Bose. So I guess I'm the problem. You're the target. I think that the basic dichotomy between extrinsic and intrinsic evidence is okay as a starting point but it can end up being confusing. There's nothing inherently always wrong about extrinsic evidence and there is no flat rule in our case law that the construction has to be based entirely on intrinsic evidence but there is a preference for intrinsic evidence because that's what the competitors could have gone to look for at the time of patent issuance. And as I said I think the best way to think of it is as a series, a sequence of preferred sources starting with the patent document and then working downward from there as far as you need to go. Much as Judge Sarah said if you can settle it, the dispute can be settled from the face of the patent document. That's the way to do it. If it can't then you have to go to whatever sources you can go to. I do think there's a great preference for sources that existed at the time and that were widely available to the public like technical dictionary or articles or patents of the time as opposed to sheer opinion by experts and I think the biggest thing to watch out for is who is the expert. For a long time we saw innumerable cases where the record was the testimony of a patent lawyer or a former patent commissioner construing technical terms in a patent when the law says it's supposed to be from the standpoint of the practitioner in that field not from a lawyer standpoint. So I think a lot has to be thought about who are the legitimate witnesses not only what is their ultimate source. Okay, Ms. Benami as you go around the country do you think this concept is clear at the trial level or do you see some confusion or at least difference among trial judges and their attitude towards intrinsic versus extrinsic? I think most judges right now feel that intrinsic evidence is safe and extrinsic evidence can lead to reversal. It's just fair. I think we need to separate out what experts can do because there is a way that I think experts are very useful but it's still guiding you in the evidence that existed at the time. It's not useful necessarily for an expert to say this is just my opinion. I was there and I think this what an expert can do that I think is very helpful for the court is to guide you through the intrinsic evidence when it says this in the specification that the plasmid was made this and this in this way well yes that's there it's on the face of the document but no one knows what it means you still need someone technical to help you understand that. Go through the prosecution. What did the prior art mean? Go through the literature of the time. What did that mean at the time? And not necessarily say well it's just my opinion. And so there's a difference there. Okay, one of the most neglected sources I think of claim construction are patents that are cited in the patent that's being enforced as being the relevant prior art because in a certain sense they're almost incorporated by reference and it's very convenient to be guided through them by the attorneys or by somebody giving a tutorial because then you don't have to spend a lot of courtroom time hearing live testimony. You can sort of walk through the key phrases in those cited patents. One thing that happens is lawyers will tell you at Mark Winner before there's certain things you can't hear certain evidence that can't come in. For example evidence as to the infringing product or process or testimony by experts. And I think that's a misreading about vitronics and Pitney Bowes because I think it's very helpful for a judge to not just understand the technology but to understand the case or controversy. That is, why is it that these terms are in dispute and is it a genuine dispute or not? So I don't think judges should limit what information lawyers can give them. They may want to limit what they rely on in terms of reaching a decision as they write out their opinion on claim construction. I think that's the key. We've never reversed a judge for what they admitted. We have reversed on a few occasions where they relied on extrinsic evidence to contradict rather clear meaning discernible in the patent but we've never reversed for what they admitted in order to get whatever benefit it had. Well that's comforting, helpful and I'm sure that reassuring to a lot of our audience we'll be back in a moment with our next segment. Let's focus on the conduct of the hearing itself now. Judge McKelvey in recent years the courts, the trial courts have been given not only the responsibility for construing patents but also gatekeeping responsibility and assessing the reliability of expert testimony. Do you see these two cases crossing paths and patent issues that come before you? Very rarely do I see Dalbert issues come up in patent infringement cases. Actually the time I do see them come up is in the context of infringement issues where a party is presenting testing expert testimony but not in claim construction. My sense is patent lawyers are happier to do cross-examination at trial rather than try to knock an expert witness out early. Okay Judge Michelle you mentioned something about the judges gatekeeping responsibility in Pitney Bowles. Do you see this issue coming up at all in patent cases that you review? Certainly not the come-ho kind of issue because usually both experts have very very good credentials. There's no equivalent of junk science. You don't see junk patent theories. So I just don't think that we have that kind of a problem. We really have a different problem which focuses more on what can you rely on and in what context rather than what can you listen to if it would be helpful to understand the technology or the invention or the background of the lawsuit. And there it's not really a gatekeeping function. It's the final step in the construction analysis. Given that let me let me turn to you then, Miss Ben-Ami and ask this. We don't have reliability problems. On the other hand all of us have seen cases where you have two blue ribbon experts wonderful credentials and yet the bottom line is they disagree over the meaning of an important disputed term. How do you wind up with these dueling experts and what happens is the best way to resolve them? I think there are a few important points. Number one, it's rare in a patent case that you'll get an expert witness for either side that's completely off the reservation. They tend to be people with very high credentials. Number two, they don't tend to have black versus white differences. They tend to be gray. They're overlapping and then the fringes are a little bit different. And that's the problem for the judge I think is to tell who's being the strict constructionist, who's being the broader constructionist and we have to remember they're often going back 25 years in time and reading articles and trying to say how would that affect the community because you're not asking them what they thought. What would the community have thought? And that's really the difference. All right, Judge Saras, you've got a case before you. You've heard the testimony from these two experts and this is the situation that you have. There's a critical term and basically their answers are different and it's up to you to make the magic critical decision. How do you do it? Well, this is the toughest area in the Markman hearing it. The law is uncertain on this subject and it's hard to figure out what exactly to say. It is not a question of burden of proof. This is at least theoretically a question of law so you can't resolve it based on burdens of proof. You also are sort of treading in difficult territory when you start talking about credibility. I find this one more credible or this one less credible. At least on the few times, I don't have as many as Judge McKelvie that I've had to come into this area. I think the safest is my statement before, which is show me, which is finding an undisputed basis in the record for what the community was thinking at the time. That isn't always easy because it's often a new area. That's why it's a patent and it claimed an invention. But if at least you can find not just what the opinions of the experts are but where it exists in the generally available literature or patents, I think you aren't pretty safe grounds. I think you're on the least safe grounds when you start turning it on credibility. I think that's true. Judge Michelle, from the standpoint of the appellate court, however, when you look at it and say, this is a question of law, is it correct that despite that, there are certain factual underpinnings that the trial judge is simply going to have to deal with? I think the Supreme Court and our court has acknowledged that there's a factual aspect of the underpinnings and really just insisted that the final step is pure law like statutory construction. But I don't think that credibility is very often the determinant. One good sort of a guide or default rule or tiebreaker is if a particular proposed construction would leave the preferred embodiment described in the written description outside the scope of the claim, that can't be right. So that's one good beacon to navigate by. And I think that context is critically important. And the correct construction usually makes a lot of sense if you put it in the context of the whole patent, of the other claims, of consistent use of terms. And I think the best single set of show me documents are the particular patents and articles that are cited in the patent being asserted because they are close in context whereas other articles that could be mentioned by experts may be in a very different context even though in the same field of technology. Judge McKelvie, you'd mentioned the fact that you get fewer and fewer experts testifying now in Markman hearings. But for those judges who do, and based on your past experience, what do you do when the inventor is one of the experts and the inventor says, this is what I believe the term means or this is what my opinion. Is he a lay witness? Is he an expert? Does he get more or less value because he's the inventor? Well, I look on the inventor and I allow the inventor to come in and testify as one skilled in the art. I know there's some case law on it that says I have to be careful about the extent to which I rely on it but I don't preclude a party from offering testimony from the inventor in part because I don't know exactly what the inventor is going to say until I have a chance to hear the inventor and hear them on direct and on cross. But I'm pretty careful in light of the Federal Circuit case law, I'm pretty careful about the extent to which I'll cite and rely on the inventor in claim construction matters. Okay. Something that judges are using in different ways is the neutral expert or the third party expert. I mean, not a third party but a neutral or outside expert. You've had some experience with that, Judge Serres. How did you use this process? Well, I found the technology so difficult I wanted someone who was going to be neutral and the parties, luckily for me, agreed to pay for the expert because to my knowledge there's no separate pot of money sitting out there. So the parties here agreed to pay for the expert and they agreed upon a person. It was a longer process than I thought because different experts have ties to different organizations but they agreed upon a neutral person and she wrote up, she was a postdoc and she wrote up a report on her construction of the claims and she helped me ask questions to some extent of the other experts as to the basis for their opinion. That was the good news. The bad news was that they wouldn't agree to let me do it off the record. Everything was on the record and I was worried that it would be unfair for me to discuss with her in an ex parte kind of way what anything meant or even just some basic technology terms. So it was more stilted that way and I think that the attorneys usually play an important role in teaching experts how to teach me and no one taught this woman how to teach me. So she often talked at levels that were higher than my understanding so that it was not a perfect use of an expert. I think I'd work on it a little differently. I think I'd twist arms a little more heavily to try at least to have some ways for me to help her at least coach me. Tell her how to get it down to a level of a non-scientist but on the whole I think that's a safer approach unless you can get the parties to agree for ex parte communication because otherwise I think there's some issues on the ex parte. Judge McKelvie, have you had any experience in using either a court-appointed expert a technical advisor? If so, what's your view and how do you feel in particular about the ex parte issues? I haven't done it and in no case have the lawyers asked me to do it and I'd actually be very concerned about ex parte communications with an expert. It makes me nervous and I know it makes the lawyers very nervous about the information I'm getting about the technology. So for example, we don't go head to the library very often to get technology books or other information. I'm assuming the lawyers know what they want me to know and because of having lawyers on each side I assume they'll be able to educate me as to the matters and issues. So one, I don't do ex parte communications with experts and two, on court-appointed experts I've had lawyers tell me that it just adds a layer of expense and uncertainty to a case that they are reluctant to ask me to do. So actually I haven't done either. I'm open to it. I may not sound like it, but I am. If you were for whatever the reason to go check some books out of the library, is that something you think needs to be disclosed to the other side that you've looked at a particular reference book and even what pages you've scanned? This is critical. I think it's fair. If it's critical, if I'm going to, certainly if I'm going to cite anything in an opinion and rely on it and tell the federal circuit here's what I look to, I think it's very important to tell the lawyers because they could say right up, Judge, you're looking at the wrong year for the textbook. And I think it's very important, but I haven't done that. And if it did become critical, I certainly would tell the lawyers I had a time before I relied on it. How about from the standpoint of the advocates, Ms. Ben-Ami, what has been your experience in having neutrals or other parties brought into the litigation? Well, I've had all different ways of this being done as well. In Judge Saris' case, we had a Rule 706 expert. And I think that if we did that again, it would be useful for the parties to get together and give a tutorial to the expert. This was the same expert Judge Saris had referred to. Because we have to be, the scientists are very good when they're in their labs. Now you bring them into a courtroom and they're not on their turf. And it's not really fair to them to sit them down in a courtroom. They have no idea what lawyers are doing. They have no idea why the questions are relevant. Why are you focusing on 1984? Why are you focusing on what's in the patent documents as opposed to something else? And I think you need to give the scientists a tutorial so that they're just not a victim in this whole process. And that's not fair to the scientists. I have had technical experts who act as a technical clerk, for example, to the judge. That is very difficult for the parties because there is no record of what is transpiring. And if it's impacting the judge in any way, there's no record for the parties to understand what's going on. That troubles me a great deal. Other judges have used special masters to help them. They've been rejected at times. It's very expensive. I'd hope that the judges could give the lawyers a chance first to try it their way. And if the lawyers aren't doing a good job, you always have another chance. Okay. Judge Michelle, is this something you see raised on appeal frequently if at all issues or disputes over the use of a neutral or an outside expert? Almost never. There's a lot of talk about it at conferences and in articles, but I very rarely see it done. And as far as I can tell, it doesn't need to be done except maybe in some very unusual circumstance. From the standpoint of reading appellate records, my favorite expert witness is a university professor who's testified in some previous patent trials, knows the patent, law background, and knows how to explain to non-scientists what's going on in the field. Okay. Ms. Ben-Ami, have you ever run into a situation where the judge has a point at a neutral over the objection of the parties? No, because the judges recognize the cost and how uncomfortable it makes everyone feel. Okay. And let me ask you one closing question for this segment. If you were a judge, what do you think would be the most important thing to hear from the experts? Where do you agree? Thank you, Ms. Ben-Ami. That concludes this part of the segment. We'll be back in a moment. Let's talk now a little about how Markman fits into the case as a whole. Judge Michel, often the way the court construes the patent is going to determine whether or not summary judgment will follow and be granted or denied. What does or should happen next at the Court of Appeals? What are the chances, for example, of an interlocutory appeal at that stage? Well, as for interlocutory appeals, we've had several dozen requests and to date we've never granted one. So I guess we have to assume the prospects will be slim to none. I can't imagine a case, but I think it's a rare case where it would be justified. Part of our problem is just that we already have a big backlog and this would only add to it if we took those cases. Of course, if summary judgment is granted with respect to infringement overall, then the losing party has a right to review and then the claim construction is taken up immediately without the expense of trial. But there may be a lot of cases where the trial can't be avoided. Do you think or do you see instances where you feel a summary judgment as motion has been granted, perhaps on a somewhat shaky basis simply to get it up to your court? It's hard to know exactly what all the circumstances were, but we do see some summary judgment cases where the analysis is very shallow and that's not helpful. Most of the analysis is very careful and we take it from there and even though it's a de novo review, we rely very heavily on the analysis of the trial judge. Another issue that I know concerns a lot of trial judges is the perceived reversal rate when a case finally does get to the federal circuit and there are comments that the reversal rate is 40%, 50%. Is that in fact accurate and if so, is that unusual? Well, there's a footnote that says it's over 40% and maybe it's a good illustration of the evil of footnotes and opinions. But we have made a careful count in recent years of the actual reversal rate on just unclaimed construction. The overall reversal rate is around 45% on at least some issue, reversal in part. But reversal rate on claim construction is only about 25%. So the widely stated reversal rate is actually not true. It's only half of what is said. Okay, well, that's good knowledge to know and I hope we'll be somewhat reassuring to trial judges watching. Ms. Benami, what do you think the litigants think of the process as it works now? What is their satisfaction or lack thereof? It's been an interesting development because when Markman came out, the patent bar all said they'd be out of jobs. There'd be Markman hearings and that would decide the entire case and no one would ever get to trial. The truth is that Markman is just a step in the process. It may lead to a finding of no literal infringement but not likely to a finding of no doctrine of equivalence infringement. It may raise more validity issues but it's now just a staggered part of the process and people look at that way. Well, another part of the process is preliminary injunction motions. Judge Serres, how do you mold that into the case as a whole as far as timing and how you view them and approach them? Does it mean another hearing or set of hearings? Well, that of course is your nightmare when your court and deputy comes in and says there's a group of attorneys downstairs that wants to have a preliminary injunction. Right away, you're forcing all these careful measured steps that you have to construe a claim into a very rapid period of time and I think that enhances and makes it likely it's going to make a mistake so that my first goal will be to look at it and see if there are serious technology disputes, serious claim construction disputes. Do I even understand the technology well enough to see whether it infringes? And if I can't make a genuine assessment that there's a likelihood of success, I go no further and I deny the motion for preliminary injunction. I think that frustrates attorneys who think they're going to go in for a quick hit and I know that time is money in a lot of these areas where the technology is moving so quickly but the bottom line is we have to make sure that the likelihood of success is there and that's too hard to do quickly so I tend to not do that but then I'll try and move the trial much more quickly than I would otherwise do to get a faster resolution of it and in general the attorneys have been willing to work with me on that. Judge McKelvie, what's your experience on preliminary injunction motions? The experience in Delaware is pretty similar which is I think the trial judges are pretty reluctant to grant preliminary injunctions. I know when I get them and they'll come in certain types of cases, pharmaceutical cases and electronics cases. I try to talk the lawyers out of presenting a prompt preliminary injunction application under the theory that I'd like to give a defendant a little opportunity to take some discovery on validity defenses among other things. It gives me more time to focus on the technology and other issues so I'll see if early on we can talk the lawyers into having a trial in three or four months as opposed to 12 months. One of the little interplays here is between the trial judge making decisions on claim construction and on probability of success versus the parties presenting a case to a jury so that one of the issues is are the parties going to give up a right to a jury to get a prompt trial or would they hold off on it? Ms. Benami, what do you do strategically on this issue? When do you file preliminary injunction motions and why? Preliminary injunctions are successful where there's been a prior adjudication usually. If somebody who had strong bargaining power went ahead and took a case and tried to invalidate a patent and it wasn't invalidated, it gives comfort to the judge where it's a real copyist. Someone just took your patent and just copied it. There can be circumstances where it's really necessary but by and large it says the judges have said judges don't want to do it. On a biotech case, have you ever been involved in a case either winning or losing where a preliminary injunction was granted? I did have one where we won but the judge it had been previously adjudicated and the judge gave us a mini trial so that's a little different situation. What about at your level Judge Michelle? Would you see these and how do you, if so, what is your reaction? We see a fair number of preliminary injunctions both granted and denied and of course they're reviewed for abusive discretion so that they're rarely reversed on appeal whichever way they came out and my rough impression is about half of them were granted and half of them were denied so from that sense I don't think that it's rare for them to be granted but it may be rare in particular technologies. That was going to be my next question. Do you see that these occur more often in certain types of patents than in others? I think they occur where it's clear cut an injunction will be granted where it's clear cut whether you have outright copying or where you have fairly simple technology so that you don't need a huge sensitive claim construction exercise before you can make a pretty good prediction of who's going to win at the eventual trial. So it's not the industry that's determining it but really more the history of the litigation. I think it's more the facts of the case although I think the more complex the technology probably the less likely to get an injunction. Let me ask you another question when it comes to reviewing a case what's your reaction when the findings or the judge's conclusions have been done orally from the bench versus a written opinion? Do you see major distinctions in the way the federal circuit views these cases? Well, I don't know that I can show a difference in the outcomes but from the standpoint of our comfort level I'd always rather have a written opinion not necessarily terribly long but written rather than oral from the bench. There are some trial judges who are exceedingly good at being very very clear and very thorough and still concise from the bench but most people can give a much better explanation when they've had a chance to write it out. Okay. And now finally Judge McKelvie you have a number of these cases that do go to trial. When you get to trial what do you do with the claim construction? How do you use it in the trial itself? The place where I put claim construction is in the final jury instructions where we have the claims set out and I will for example after a particular words where there's been a dispute as to the meaning of the words I'll put brackets in typing out the claims put brackets in and say by that we mean or that is the following and then I'll quote it in the jury instructions. All right. And in these cases there are often a number of terms that are complicated. They're not common terms but the parties don't have any dispute between themselves about what those terms mean. How do you let the jury know what those kinds of terms mean? Generally we have preliminary jury instructions that we give the jury at the beginning and actually we pass them out type them up and pass them out and they'll have attached to them a glossary of terms glossary of technical terms where the parties agree on the meaning of the terms while they may not be the types of words that the jury might understand so we'll define them and agreed upon definitions then we'll actually also attach a glossary of legal terms such as for example the words file wrapper or other words that may come up during the course of the trial. All right. Thank you. We've come to the end of this program. While there are different views some of the major conclusions are the following. First, timing for Markman hearings depends on the case but generally speaking the most optimal time will be after most discovery has been done but well before the trial date. Second, judges need not be afraid of science. There are a variety of tools with which to learn especially the experts. Utilize them all. Third, you have latitude and discretion to hear both intrinsic and extrinsic evidence. Just make sure that your findings point out what intrinsic evidence you utilized in order to reach your conclusions. As we said in the beginning although we've used a biotech case as a model the concepts and principles discussed apply generally to all patent cases. The Federal Judicial Center is very grateful to all of our panelists for giving their time and sharing their experience with us. We hope this helps those of you who are watching and encourage you to check the FJTN Bulletin and the Center's DCN site for the schedule of other programs in this science series. On behalf of the FJC thanks for watching.