# Oral argument — AKAMAI TECH v. LIMELIGHT NETWORKS > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A7093 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** November 18, 2011 ## Text The United States Court of Appeals for the Federal Circuit is now open and in session. Glad to stay with the United States and its honorable court. Please be seated. Our first case this morning is Akamai v. Limelight. Mr. Gunner, this afternoon I've been corrected. This happens. You get corrected occasionally, even if you're a chief judge. It occasionally happens to me too, Your Honor. Good afternoon and may it please the court. Your Honor, the single entity rule of the Akamai panel and the related cases creates a huge and, I submit, unfair loophole in the law. But more significantly, I submit it's unsupported by the case law, it's contrary to the statute, and it's out of sync with the common law of torts. Let me be a little more specific. All of the cases, Akamai panel, McKesson, Muni option, they all cite back to BMC. And BMC cites a number of cases. Those cases that it cites do not support the single entity rule that it enunciates. They are either directed to the all elements rule or they have no meaningful reasoning supporting the rule. And at least three of the cases reach the opposite conclusion. And those three cases cite two others which reach the opposite conclusion. It's inconsistent with the statute, 271A. Everybody focuses on that statute. The words are very simple. Whoever uses any patented invention, I've left out a few words in between, whoever includes both singular and plural under 1 U.S.C. 1. If you look at the dictionary, it says person or persons. And if you look at section 101, it says whoever invents. And we all know that you can have plural inventors on a patent. Section 102, a person shall be entitled to a patent. We all know that you can have plural persons entitled to one patent. Weimark focuses on the word use in whoever uses. But that tells you what infringes. It doesn't tell you who infringes. The pre-1952 case law actually supports divided infringement, a joint infringement. The NTP case, the Warner-Jenkinson case, both say that 271A in the 1952 act was a codification of pre-1952 law. And pre-1952 law supports Akamai's position. Didn't Arrow say, however, didn't the Supreme Court expressly note in Arrow 2 that the law of secondary infringement became confused prior to 1952 and the recodification was aimed at settling things? And isn't it also true that the legislative history, including testimony of our former colleague Judge Rich, clearly demonstrated that they were trying to get at curing the prior case law, not embracing it? Your Honor, all of that testimony and that legislation was directed to contributory infringement, which was, in the language of the reports, the Senate reports, confused and difficult to follow. It was designed to clear up that area of the law. I'm not talking today about contributory infringement. I'm talking about direct infringement. And the Arrow case and the Warner-Jenkinson case both say that the act was not intended to change the law of direct infringement under 271A. So I agree. Wasn't it always understood that direct infringement required that a single party would make, use, or sell all of the recited elements in a claim to mention? I disagree with that, Your Honor. The Peerless case, which we cite, the Seventh Circuit 1937 case, clearly had multiple actors. It didn't have a single actor. The Halliburton v. Honolulu Oil case and Ninth Circuit 1938 case also said the same thing. The McKesson brief cites still another case, Rumford v. Hecker, which says the same thing. Is there any indication that Congress knew about any of those cases in 1952 when it codified the statute? Your Honor, the cases pre-1952 I submit support. No, no, but the question is, was Congress aware? Is there any indication that Congress was aware of those cases? I cannot cite to anything that suggests they were aware of it, but I assume to the extent that congressmen were lawyers. Prosser says the same thing, and we all studied Prosser in law school. I'm not talking about infringement, but under the Supreme Court's decision about what constitutes ratification, you can't find ratification unless either Congress was aware of it, which you agree they weren't, or it was a well-settled and widely known rule, which this wasn't either, was it? Your Honor, what I'm relying on is the Supreme Court in Warner-Jenkinson and Arrow said that there was no intent to change the law. The pre-1952 law of direct infringement, whatever it was. And so whether they knew about what the law was is not what I'm relying on. So you don't have to satisfy the usual standards for congressional ratification of judicial decisions? I cannot satisfy the test that you are describing. Mr. Dunner, did the Congress in the course of this discussion about direct infringement cite indirectly to any of the cases implicating a single entity rule? Not that I know of, Your Honor. So your argument, I would assume, would be there's no evidence either way in terms of Congress's ratification that they were necessarily adopting either one. That's exactly right. Mr. Dunner, can I move you on to sort of the merits of your case? Both briefs, I think, state when one party directs another party's performance of the method step, the other party's performance of that step should be attributed to the directing party. First question I have is, would that test reach a situation where the defendant simply instructs another to perform all steps and performs none on his own? Or are you only talking about the circumstance in which the person is also performing at least one step with the method claim? For my case, I'm only talking about a situation where each party performs at least one step. Now, I know there's a different issue in the McKesson case, but my case only involves that issue. And is the position you take, therefore, one in which if one person performs 1% of the steps, one step out of 100, and another performs 99 steps, that person would still be liable? You mean the one that does the one step? Yes. Your Honor, we have a three-part test, and the third part is knowingly combining method steps. You don't need to know there's a patent. You don't need to know there's infringement. But you need to know that you are performing a step that is combined with other steps that you know about. So if you're performing one step out of 100, and that's in the prior art, the step you're performing, if you know that another person is performing 99 steps, then you're liable for infringement under that test? It doesn't make any difference whether you're performing a step in the prior art, because as we've heard many times from decisions of this Court, inventions are made up of old steps put together. So it's the combination that we're talking about. If you know of one step, and you know it's being combined with 99 other steps, and you know what those 99 other steps are, then you are potentially liable under our proposal. That's exactly right. But if someone is not performing any steps, they would have no liability, right, even if they are directing the performance of those steps, unless they meet the quite stringent test of global tech, right? So the difference between the 0% and the 1% is pretty steep, correct? Your Honor, if there are no steps, then we are talking about indirect infringement, and that is not covered by my proposal. So the performance of one step out of 100 versus performance of no steps determines whether or not there's liability or not? That's exactly right. But we have gone out of our way to make clear that we are not trying to find liable or suggest that a person should be a liable who is an innocent infringer, who does not know the combination of steps. So by way of example, in the Akamai case, the Akamai facts, we have a limelight that is the content delivery provider, and we have the content provider. Each of them knows what the other is doing. If one of them did not know what the other is doing, that person would not be liable. For example, is knowledge all you require, or is it direction? I mean, it's direction. Your test, I thought, was direction plus knowledge, not just knowledge. Are you suggesting that just knowledge is sufficient? Your Honor, we have three tests. One test is direction or control, not and control, or control. The second test is concert of action. The third test is knowingly combining steps with another. So there are three separate tests, and in each case there has to be knowledge of what is happening. In the first case, direction or control, you have what somebody else might call a classic case. I don't know. But in any event, there are three different tests. We are not keying it to direction, and we think that the BMC case and the Munioption case and the Akamai panel case went wrong in limiting it to direction or control, and furthermore, further limiting the direction or control test to an agency or contract criteria. Mr. Dunner, I think I understand the gist of your first two tests. I'm having trouble getting a handle on the third, knowingly combining. Is that significantly different from the concerted action test? And if so, what does it cover that concerted action did not cover? Concerted action is a collaboration by agreement, direct or indirect agreement between parties. The third test that we're talking about is illustrated by the Peerless case, where you have a manufacturer of railroad equipment, including a gear, and the gear had a convex surface and had to be flattened, and so the manufacturer of that gear assumed that the users would actually, in practice, flatten the surface. So, in that case, you had no concert. There was no concert because there was no agreement between the parties. There was no collaboration between the parties. It was just a manufacturer selling a product, knowing that in use, the last step of the process would be performed. So, in that case, the manufacturer would be liable, but the user would not be liable because the user had no idea that it was doing anything connected with a process or multiple steps. So, if I work at a parts store, miscellaneous electronic parts, a radio set, let's say, and you come in and you say, I want to buy a part off the shelf. It's a standard part, and what I'm going to do is put it together with a lot of other devices. Now, let's translate this into a process and avoid the device, but if you explain the process you're going to follow, the first step of which involves the thing that I'm providing you, wouldn't that fall within your example of a combination of mine knowing, as the seller of the part, that you're going to combine it to do something which turns out to infringe a patent? Would you reach the Radio Shack employee under those steps? If the Radio Shack employee knows – let's say there are ten steps in your process. The first step is the obtaining of the part from Radio Shack, which the Radio Shack guy provides the part. And the Radio Shack employee knows that when it is taken, it will inevitably be used to complete the other nine steps? This is the project the fellow is going to use it for. He has, of course, no idea that that's an infringing use. But that's what he's told. This is the project that it's going to be used for. Does that infringe? He doesn't need to know it's an infringing use because this is a strict liability issue, 271A. If he knows that he's performing the first step and he knows that when he gives it to somebody else the other nine steps, if there's a ten-step process, will be performed, then he is potentially liable. Now, I'm confused. What happens to the substantial non-infringing use component of contributory infringement? You set that aside? If there is a possible infringing use or the salesman is told it's going to be used in infringing use, but it's an off-the-shelf product with other uses, which I assume – It was. That was my concern. It was just an off-the-shelf part. But it was going to be used, the salesman was told, for this impermissible, as it turns out, purpose. Your Honor, this is not a 271C situation. This is a 271A situation. We're talking about method steps. 271C deals with selling substantially all the parts where you've got an apparatus system that is sold, where you're performing – it doesn't focus on performing steps by the contributory infringer. It involves selling an apparatus that ultimately will be used for all the steps. What we're talking about is a process claim under 271A, and we're talking about two different people performing part of all the steps. Each one performs part of it. The other one performs part of it. There's got to be knowledge that what is being performed is part of a complete series of steps, the combination of which would in fact be the steps in the claim. But they do not need to know there's a patent. They do not need to know that there's infringement. Mr. Dunner, what are the implications, if any, of the rule you would like us to adopt for apparatus claims or system claims? As far as I'm concerned, there are no implications. Apparatus claims are covered by the NTP and related cases where you're talking about beneficial use, and you don't go through this analysis. What about somebody selling the components of an apparatus to another party that will take those components and put it into an overall device that's covered by an apparatus claim? Does the same rule apply? I am not suggesting that the same rule applies. Why would it not apply? Let me put it that way. You're advocating a rule based upon an interpretation of the statute stemming from the interpretation of the word whoever. Because the court, in its wisdom, has adopted a different rule for apparatus claims than it has for method claims. You have the NTP case, which is restricted to apparatus claims. You have the centillion case, which distinguishes between method claims and apparatus claims. The court has drawn a line between those two. No, but there were distinctions there based upon whether the act of infringement was a use or a making or a setting. That's exactly right. What I'm talking about is you're advocating, as I understand it, a rule of joint direct infringement liability. And the briefing, all of this has been in the context of the facts of the case, which is directed to a method claim. What are the implications if we adopt this rule for apparatus claims, where somebody is selling a component that someone else is going to use in a combination that's covered by a patent? Wouldn't the same rule apply and result in some liability? Your Honor, I think not. And the reason I think not is the court has, for whatever reason, put in a separate category apparatus claims. If somebody sold an apparatus, if there were no method claims, you have 271C, the question that Judge Newman raised. If you sell a material, a part of a whole, and it's a material part of the whole, and it has no substantial non-infringing use, it's not a staple, then you would be liable under 271C. So, in the fact situation you posited, it would be covered by 271C as long as it satisfied the language of 271C. Let's take an example like Deep South. Suppose it's the exact same facts as Deep South, except that the assembly takes place in the United States. Would Deep South be guilty of direct infringement under those circumstances, under your theory? You're talking about strictly apparatus, we're not talking about method claims? Yeah, apparatus claims. They put together, they offer a kit of the parts with the expectation that somebody else will put it together. Are they guilty, and the putting together happens in the United States, not abroad. So, is Deep South guilty then of direct infringement by doing that? My discussion is related solely to method claims. I understand, but what's the answer? Would Deep South be a direct infringer or not, under your theory? I think the answer is covered by 271C, pardon? Does the answer depend upon whether there's knowledge by the seller of the component of what the others are doing? Well, at the very least it would. If it's a 271, I think that's a 271C question. So, there's no direct infringement under those circumstances? I would think there's no direct infringement under those circumstances. Why should that be different than a process claim? Your Honor, it may logically be that it shouldn't be different, but I'm dealing with the case law as I'm standing here today, as it exists. And the court has enunciated separate rules for apparatus systems, because it focuses on use, and it has four method systems. But if the logic is that in the Deep South situation where the thing is done in the United States, Deep South is not a direct infringer, and that's the rule that's established, and I agree with you that Deep South is not a direct infringer, how could Deep South be a direct infringer in a process claim where exactly the same thing happens? Your Honor, the reason is that historically apparatus claims have just been treated differently. And method claims, I'm not trying to rewrite the statute. I'm trying to ask the court to interpret the statute. And historically, method claims have been found to be infringed when two different actors combine to complete the method covered by a claim. Historically, in the apparatus situation, that has not been the case. In the apparatus situation, the 271C, 271B have been held to apply. And I'm dealing with the law as I find it today. And if I were rewriting the law, I might not make a difference between those situations. Certainly the statute doesn't distinguish between the two. Pardon me? I mean, the case form is that the statute doesn't, in this context, distinguish between the two. Well, I think the statute covers single or plural actors. And I think that the case law, we're talking about vicarious liability here. And the DMC and the option case have already bridged the gap and said that if you've got one person whose acts are attributable to another person, it's the same as if that other person is performing the act. But the statute doesn't make any distinction in terms of 271A between apparatus claims and method claims. The cases are different only in the sense that they treat different types of infringement based on use, based on manufacture, based on sale. Those are different fact patterns that result in different types of analyses. But in terms of distinguishing between what is required for infringement under 271A, there's no distinction between apparatus claims and method claims at all, is there? Your Honor, if you are writing, the logic as I responded to Judge Dyke, the logic, in my view, is the same in the method situation or in the apparatus situation. So, and I agree, there's nothing in 271A that distinguishes between methods and apparatus. If this Court, I think the case law and the commentators and the history of the Patent Act are such that when you have a method claim, you should be able to combine the acts of two different actors under the circumstances that I've described. A directional control, concerted action, or knowledge of the acts being performed. If this Court wants to extend that to apparatus, I would be delighted. I don't think there's anything preventing you from doing so. I'm just telling you historically, the Court seems to have adopted a different rule for apparatus and products and what have you than it has for methods. I see no difference in the logic. It's not our role as a Court to be expanding the scope of coverage. If that's what is required, then that's for Congress. Your Honor, I don't think that it is expanding the statute. The statute says, whoever, and whoever uses any patent invention, whoever, I submit, covers single actors or plural actors. Historically, the Courts have extended it between single actors and plural actors. The 1952 Act, 271A, was designed to codify the law. It was not designed to change the law of direct infringement. There's case support and commentator support, and the rule of torts support, the restatement of torts, support the concept of direction, 876. It supports the concept of concerted action. But wasn't the codification of the circumstances before 1952 relating to multiple parties that which appeared in 271B and 271C? Wasn't that what 271B and 271C were intended to address? Yes, but the Supreme Court has held that the Patent Act of 1952, 271A, codified the law dealing with direct infringement, 271A. That is what the Supreme Court has held. The law of 271B and 271C was the subject of much discussion, including statements from Judge Rich, designed to clarify the law of contributory infringement to eliminate confusion that had been created in the Courts on that subject. Mr. Donner, you mentioned Section 876, and I'd like to ask you a question about acting in concert, your second theory of joint infringement. In particular, I'd like to know how you respond to comment C of the restatement, which says, and I quote, It is essential that the conduct of each actor be in itself tortious. Well, a person who performs only, say, one step in a method in the abstract is not committing itself a tortious act. So how does your theory of acting in concert squarely fit with the restatement and its requirements, which it says is essential? Your Honor, comment C is pretty much explained without reference specifically to that in Prosser. And what Prosser makes clear is that you're distinguishing here between somebody who's acting and somebody who's acting innocently. So in my hypotheticals, if you have two people and one knows that something that is being done, combined with something else, will result in five steps or six steps, whatever that claim actually is. So you have to know of the fact that it would be an infringement of a patent? No, you don't need to know that. But you'll notice... So then they don't need to know that the act, even in combination, is a tortious one. No, but what Prosser says is, and I'll just read this to you, A number of courts have held that acts which individually would be innocent may be tortious if they thus combine to cause damage. The single act becomes wrongful because it is done in the context of what others are doing. But does that sit within the concerted action second prong or within your third theory? Your three theories, and my concern is over the concerted action one, and what you just read to me seems to describe your third theory of infringement, about the individual actors and the innocent. Your Honor, I think it really relates to all three theories. And the reason it relates to all three theories is that if you have an act which alone, the example I give is polluting a stream. If you have somebody adding something to a stream, which alone doesn't do anything, but in combination with something that somebody else adds to the stream, you can have a tortious act in that situation. And all they're trying to do in the comment C that you read is distinguish between an act which becomes tortious, either alone or because it's combined with somebody else, and the words that immediately follow it, the act is innocently, rightfully, and carefully performed. And in fact, Prosser has that same language in another section in the exact combination that we see it there, where you in fact are, he talks about conducting tortious unless as distinguished from something that is done innocently. And I'm looking for it, and I will find it in a second. He says, it's furthermore essential that each particular defendant who is to be charged with responsibility shall be proceeding tortiously, which is to say with the intent requisite to committing a tort or with negligence. One who innocently and carefully does an act which happens to further the tortious purpose of another is not acting in concert with the other. So we're talking there about acting in concert, and that's almost the same language that is in comment C. If one of the actors is innocent, then you don't have acting in concert, correct? That's what you just seem to have read to me. Maybe I misunderstood you or misheard you, but there's no acting in concert if one of the actors is acting innocently. That is true. They would not be acting in concert in that situation. Mr. Gunner, I have a question on the direction of control test, or rather the three-part test that you're articulating. I don't see clearly why the two separate prongs of concerted action and knowingly combining action, why they're not subsumed under direction of control. Wouldn't any instance where you have concerted action also involve elements of direction of control, and those being questions of fact? It could be, but I've always considered concerted action to be cooperative efforts between multiple parties, whereas direction and control may not involve any cooperation. There may not be any agreement. You may control somebody to do something without there being an agreement, so there's overlap. So I may give instructions to somebody on how to use a particular article and yet not be having concerted action with that person, but wouldn't my giving instructions, detailed instructions, here's what you do, wouldn't that be direction, or wouldn't that fall under control? It could be both. It could be actually both. You could give direction and control to a candidate in the Akamai situation. You could argue that there's concerted action there. I think, in fact, you could reasonably argue that. You could also argue that there's direction and control there, because there's total direction. There is what we feel is a contract, I know the Akamai panel said there wasn't, a contract in which they get a performance guarantee in return for doing certain things. We give them instructions. We tell them how to do it. We give that we send engineers or put engineers in contact with them, tell them how to do it, and they can't get the performance guarantee unless they do it. So there we have direction. There we have, I think, concerted action, because they have an agreement. So, in many cases, there will be total overlap between the two. Mr. Dunman, maybe I could exert a little control. I think we need to hear from Mr. Panner now. Okay. Thank you. Thank you. Chief Judge Rader, may it please the Court. I think it's probably useful if I begin, perhaps, with where Mr. Dunner left off in response to Judge Rainer's question with respect to direction and control. And the question was, well, if instructions are provided, doesn't that constitute direction and control, and it does not. And this Court, in BMC and then again in Muni Option and the panel opinion in this case, understood that there's a very sharp distinction between the attribution of conduct under principles of vicarious liability and joint tort fees or liability, where there is wrongful conduct by multiple actors that may lead to liability for both of them and shared responsibilities that are resulting harm. And so, if you look at the way that Congress codified the infringement tort in Section 271, there's a very sharp distinction between direct liability under 271A, which requires the performance of all of the elements of the patent and in which liability is strict. It doesn't depend on knowledge. It doesn't even depend on knowledge that the conduct is being performed. What it depends on is simply the use of the invention, in this case, the use of the method by performing all of the steps. And Section 271B and C, which have actually an extraordinary standard for intent, they require not merely knowledge of the patent, not merely knowledge of the conduct, but knowledge that the conduct that's contemplated is infringing and an intentionality to defeat the patent holder's rights. And so, to import into Section 271A the possibility of liability at a lower standard, where mere knowledge and mere inducement of conduct by another would be a basis for liability, would undermine what is a very fundamental distinction under Section 271. And the Supreme Court has made clear in cases like Central Bank of Denver, Mayer v. Hawley, and the Janus case last term, that where Congress has established a cause of action that's based on the conduct of the defendant itself, or himself or herself, that where Congress has done that, it is not appropriate to use principles of joint tort liability to expand the liability that Congress has recognized. Again, Congress has provided for joint tort fees or liability in some circumstances under Section 271B and Section 271C, where it is codified what was pre-1952 contributory infringement. And if you look at the cases from pre-1952, where the courts talk about joint tort fees or liability, what they are talking about are situations of contributory infringement, not direct infringement. And direct infringement has been understood for a very, very long time to mean the use of the invention, in the case of a method claim, the performance of every step of the method. And there are cases that are cited in our brief and in some of the amicus briefs that go back to the 19th century that make clear that if one does not perform every step of the method claim, one has non-infringed. And that's the case, the General Foods case has a very nice discussion of that point, of course. In this case, the court in Canton Biomedical, as well, made very clear that for there to be direct infringement of a method claim, every step of that claim must be performed. Mr. Penner, do you understand the test, or at least part of the test that is advocated by Akamai, which is that it shouldn't be necessarily direction and control, it should be something called instruction. Do you understand that to be a clear line, or instruction being a highly factual component that could vary from case to case, depending on the relationship of the parties, the industry itself, and so forth? How do you see it? I think that's exactly right, Your Honor. And I think that there is, and building on that point, there's this basic distinction between the attribution of conduct, the imputation of conduct under principles of agency law, or where someone has procured a particular performance, where the principle can be held liable as if he had, or she had, performed that conduct, him or herself. And situations in which there is wrongful conduct, wrongful procurement of some conduct, or wrongful participation in a tortious scheme, which again is completely outside of what Section 271A has been understood to be directed at. And for that reason, the line that this Court has clearly drawn, and I would submit that it was clear in BMC, it certainly is clear, it was also clear in Muni-Austin, and I think the panel decision in this case got it quite right, that there is, built deeply into the structure of the law, a distinction between the imputation of conduct under principles of agency, effectively, and broader joint tort-feasor concepts where the basis for liability is fault, is the wrongful joint conduct. This case, and it's important to emphasize, this case is solely about direct infringement. In this case, Akamai claimed that there was literal direct infringement of their method claim. But it is undisputed that Limelight does not carry out one of the steps of that method. There's no, again, there's no dispute that Limelight itself does not do that. But Limelight knew that someone would set into motion the taking of that action. Well, Your Honor, really two points. It is true that Limelight presumably hoped that its customers would engage in the action that Akamai says constitutes tagging, and that that would result in the use of the service. But it didn't have any ability to control that. The customer was under no legal obligation, if the panel correctly found, was under no legal obligation to Limelight to perform that step. And, again, the issue is whether we did literally what is described in the patent, and what is literally described in the patent. But would Limelight's activity be worthless without a customer doing the tagging? Well, Your Honor, in order for the customer- Wasn't the customer suspected and hoped- I think all of those things are true, Your Honor, that Limelight was hoping that the customer would use the service, and that involved doing what was found to be a tagging step. Wouldn't that be direction, then? No, Your Honor. The direction and control test comes from agency law. That is to say, and again, because the critical distinction is between imputation of the customer's conduct and a standard that would require some sort of mens rea function of wrongful intent. And so, in the case of- the direction or control test comes very expressly out of agency law. But an agent acts for the benefit of the principal, isn't that right? That is right, Your Honor. But here, the tagger is acting for a joint benefit, for his benefit plus Limelight's benefit. Well, they're acting for their own discretion, for reasons of their own. They're under no obligation, and it's precisely for that reason that the method is different. And the question here is a narrow one, which is, is there literal direct infringement under 271A when it's conceded that Limelight does not carry out one of the steps? And so the question becomes, is there a basis for imputation of that conduct? And there are very well-settled standards in the law for when such imputation can be found. And that- those are the very standards that this Court has alluded to in its prior decisions in the line of cases that, of course, are now before the- Mr. Perry, would you find a way to impute responsibility on these facts? This is perhaps an extreme case, but it's one that, to me at least, addresses the problem of subterfuge that sort of looms here. Suppose that I invent a very much improved method for making Ford hybrid engines. Ford finds out about it, and they read my patent, and they say, wow, that will save us 50% of the cost of making hybrid engines, but we don't want to pay the licensing fee. So here's what we're going to do. We're going to call in ACDelco, and we're going to say, we want you to build hybrid engines, and we slip a copy of the patent and say, you know, use any method you want, but you might want to think about this one. We want you to build hybrid engines except don't put the case on it and don't put the terminals on the engine's battery. Okay? And Ford then gets the almost completed engines and batteries, and they do the last step themselves. No liability or is there a way of getting at that problem and saying, no, no, they can't get away with an obvious subterfuge like that? Okay. Well, Your Honor, the first point I want to make is that there is no direct infringement. That is to say, in the – well, what you've described – Remember now, in the transaction, they're very well counseled. Ford did not say, we direct you to produce this in this way. They simply said, just sell us an engine, an incomplete engine. Let me say that as you pose the hypo, let me give first perhaps a somewhat trivial answer, and then I think the answer that may be more responsive to your question. The somewhat trivial answer is it seems to me a trier of fact on that fact pattern could find that although in form there was no direction, there was pretty clear direction given. And if you handed a copy of the patent and gave the wink and the nod and paid for that performance, that – Would that then put it in inducement? Is that where you go with that? I mean, because it's pretty clear that Delco did not do all the steps because one of the steps was to put a case and terminals on the back. I guess what I'm saying is there's two possibilities. So there's a possibility that Delco really may have been acting as Ford's agent in your hypothetical and that if one were to say, you know, can you avoid having this person imputing the conduct where it has been so clearly directed notwithstanding some of the formal language around it, that might be a question of fact. But let me get to your – assume that perhaps the facts are a little less extreme. And the question – if the problem is self-refuge, if the problem is intentionality, it does not live under 271N. If there is not literal direct infringement, that is not a problem that needs – that should be dealt with under 271A. But what under 271A prevents viewing these actors, whether you call it collaboration, if this isn't a one-step-in-a-hundred situation that's been presented, it's quite straightforward. So are you saying that 271 actually prohibits viewing these actors as together, carrying out the patented measure? Judge Newman, I think I started my answer to Judge Bryson by saying that in an extreme case like that, one might actually be able to impute the conduct. It might be an issue of fact. But what is that? Are you drawing the line? You say this is extreme. What makes it extreme? What made it extreme was the provision, was the fact that Delco was hired by Ford to conduct certain staff. And under Judge Bryson's hypothetical, there was a – But then you would say that's direction and control, not an extreme collaboration. I may not have understood Your Honor's question exactly. I'm trying really to understand why the statute prohibits viewing this extreme situation, which may or may not arise spontaneously among the actors or may be suggested by one or the other. Why it prohibits looking at what each one does and saying, ah-ha, every step of the method is being performed. Well, because the fact that every step of the method is being performed in the passive voice does not indicate that there's infringement. Let me put it to you a little more directly. Suppose Ford says there's seven steps in this method. We're contracting with you to perform the first six steps, and we, Ford, will perform the last step. Does that involve direct infringement? Yes, it does. Because in that case, if you contract out the steps and you have the legal ability to require your independent contractor as acting probably in that circumstance, at least for purposes of what you contracted with them to do as your agent, yes, that would be something where you could impute the conduct. And so the very distinction I'm trying to draw – No, go ahead. The distinction that I'm trying to draw is between a situation in which, as in the hypothetical, there's a contracting out where the actor whose conduct is being imputed is under a legal obligation. That direction or control standard, again, comes out of agency law, comes out of a circumstance in which the principal has a legal right to direct or control the conduct of the agent, and it is that legal control that allows the imputation of the conduct. Well, why in Judge Breis' hypothetical couldn't you infer an agreement from the facts, the wink and the nod? Well, I think that's – Yes, you could, right? You could. What I'm kicking myself about here is that I'm trying to answer two questions, and maybe I should be clear about one. There are situations where you could have a factual – you know, you would have a factual issue as to whether, in fact, there was direction or control of a – you know, whether the principal hired the agent to do something in a particular way, even though there was in form a contract that didn't necessarily dictate that. I think that what I was also trying to get at, though, was a less extreme situation and to suggest that in a circumstance where there is – where there isn't that sort of clear direction to do something in a certain way, that it might be a harder case. And what is the statutory support for this decision you've just drawn? That's what I'm trying to focus on. So under Section 271A, the basis for liability is the use of the invention. And so the person, the defendant, must use the invention. And so if the defendant has not carried out every step of the method, the defendant has not used the invention. And so the distinction that is – that I think is fundamental and is between a circumstance in which the conduct of the – if you were trying to impute conduct of the third party to the defendant, there has to be a basis for finding vicarious liability. But the statute says whoever. And it doesn't say whoever alone performs an invention. It says whoever. And doesn't that include the plural? And there are multiple steps here. There are only two parties. All the steps have been performed, and the parties have a relationship with respect to each other, a buyer and a seller. Certainly the statute says whoever, and that could be multiple parties if all of them have, in fact, carried out – have used the invention. But that's not the circumstance here. You don't think between two of them they haven't performed all the steps of the invention? They may, between the two of them, have performed all the steps, but neither of them has used the invention. And that's the critical distinction, is that the liability doesn't exist in the air, as it were. That infringement is conduct by a person or persons. Pardon me, Your Honor. It does occur in the air. Isn't this the Internet? In the cloud. The point is that there has to be an infringer. There has to be, under Section 271A, it's a basis for liability. If there is a defendant who is carrying out all of the steps of the method, then that would be use of the method. It's one of the answers to Judge Greisen, perhaps, and I may be wrong about this, that this is a matter that – that's why we pay patent prosecutors so much, and that this is something that could be avoided in the drafting. I mean, that's why when people sit there and draft their claims, they can anticipate these instances where even assuming – and I know you agree that there might be liability in this case, and Judge Dyke made the point that there likely is. But even if there were not in some sort of circumstance, isn't the answer to that that people have to draft – they are able, at least, to draft claims that avoids that situation? Certainly, Your Honor, I think that careful claim drafting in the case of method claims can address the problems that are anticipated. And, you know, I think it was the amicus brief that was filed by Cisco and others that points out that this is something that has been understood among the patent prosecution community for many years, that it is critical to make sure that one drafts the claims from the point of view of ensuring that there's a single direct infringer. You're not challenging Judge Post's premise that patent prosecutors are paid so much. Your Honor, I did hear a laugh or two behind me. I'm not a patent prosecutor myself, so I won't – Mr. Fair, let me ask you a question. The Facebook amicus brief argues that – it argues in support of your position and says vicarious liability is the correct standard. But it defines vicarious liability a little bit differently than you do, or even than it appears this Court might have so far. It actually says that vicarious liability under traditional tort principles would encompass a joint enterprise. Do you agree that your standard would encompass a joint enterprise? Yes, Your Honor. I think in a circumstance where there is, in fact, a joint enterprise, that is, where there's a mutual right of control in, for example, in a partnership or a joint venture, then the law has understood joint venturers, for example, to be mutual agents. And so that could be a circumstance where there would be liability. Again, I think it's important to recognize functionally what is being recognized when one is talking about imputation of conduct and vicarious liability. What matters is that mutual legal right of control between the parties where one is trying to impute one's conduct to the other. And so in a circumstance where there was, in fact, a joint venture or partnership or some other formal joint enterprise – and I think this is actually addressed well in the Facebook brief – that one would, if there were, in fact, that legal relationship that one could impute the conduct of one to the other. To the extent that Mr. Dunner argues that there's a gap in the law that this Court should fill because we're not covering all the bad conduct that's out there, if you encompass joint enterprise liability as well as the other grounds for vicarious liability, the gap in the law isn't really as big as it might otherwise appear. Well, I obviously take issue with the idea that there's a gap in the law. I mean, the point is that – and I think that Judge Prost made reference to it – Akamai drafted its claims. It described a process in which several steps had to be carried out, including tagging. My understanding is that Akamai itself provided service using that method at times. And so there's nothing about the way the patent was drafted that doesn't encompass certain conduct, but it doesn't describe what Limelight does. Limelight does it a different way, and the customer does the tagging. And so, certainly, if there is a circumstance in which parties get together, or in the case of the hypothetical that Judge Bryson used where someone contracts out certain steps, it's been clear in the law for a long time that one cannot avoid liability by contracting out or by hiring an agent to conduct something. But that, again, reflects the principle of vicarious liability and imputation of conduct. And so the line that's very important to maintain, consistent with the strict liability nature of Section 271A and the really very powerful remedy that 271A provides – direct infringement is an extraordinarily powerful remedy for the patent holder because it doesn't depend on knowledge of the patent. It doesn't depend on anything but performance of the steps. But at the same time, that's why it's necessary to insist upon the performance of all of the steps. And Judge Rich in his testimony said it much better than I can in terms of saying, once you're not talking about doing exactly what the patent says, we're outside the realm of direct infringement. You just said you're not arguing that there's a gap in the law, but isn't that your whole point? There is a gap in the law that it's for Congress to fill, not the court? Well, it depends on – I guess it's easy for me to say, to the extent you think there's a gap in the law, it's for Congress to fill. But I guess what I am saying, and I think this case illustrates it, is that Akamai drafted a patent that covers certain steps, and we don't do that. And in fact, there's no dispute that Limelight itself does not literally do all of the steps of the patent. And so – But Delco, which is a patent hub, which is great, makes the entire process, performs all of it except putting the last plug, making that final last connection, which is required by the method step. It then sells these out to open market, and then Delco goes forward, buys a whole bunch of them, and says, oops, plug attached, patent not infringed, because I did the last step and they did all the others. There's no other purpose or use for this invention, nothing else to be done. I know you say it doesn't matter since it's strict liability, but Delco knew about the patent and they did this intentionally to avoid infringement. And you're saying, too bad, no harm, no foul. I mean, clearly there's harm, the patent deed is harm, but you're saying Congress just didn't mean to cover this kind of conduct under 271A? Well, Your Honor, I think that that hypothetical likely sort of reinforces potentially the point that Judge Prost made about claim drafting. In that example, if what's material about the manufacturing process that's different could be embodied in the patent, then perhaps it would be covered. I understand that the hypothetical assumes that there's some truth to that. But it doesn't have a useful invention until the last connection is made. Before that point in time, it doesn't work. So I'm not sure. I mean, maybe there are ways for prosecutors who are far more clever than I to draft around this. But it seems like that is pushing the problem off somewhere else as opposed to telling me why it doesn't fall under 271A. Well, the reason it doesn't fall under 271A, and you're right that it doesn't in our view, and the reason is that 271A is limited to the literal performance of all of the steps. But all the steps are literally performed in my example. But not by the defendant. And so the question then becomes... So I have to buy into the single entity rule for you to prevail. Well, I don't love the term single entity rule because I think it's not quite accurate, Your Honor. That is to say that it's not this court's position, it has not been the position of the panels that have adopted the rule that's at issue here, that a single entity has to perform, has to be liable. That is, there's liability for the conduct of an agent, there's liability for contracting out. But yes, it is true. I would agree with the single entity rule characterization to this extent that under Section 271A, the question is whether the defendant has used the invention, whether the defendant has carried out all the steps of the method claim. But then the problem is that you can't have infringement under 271B or 271C unless you have direct infringement under 271A. So it isn't as if in this situation which on its face just reeks of, let's say, inequity, that there are other remedies under other portions of the statute unless, according to at least some of our cases, they say unless there's direct infringement, there can't be any form of indirect infringement. So what we're doing, and perhaps this is where the flaw in precedent arises, is at that very, very point, because in the hypothetical we postulated, all of the steps are indeed being performed without dispute. We just seem to somehow have created a situation where there is no remedy, even under the simple straightforward cases. That is not one step out of a hundred, but one step out of two. And it seems to me that if we get to that stage, then we really are approaching the abuses and the horribles that some of the amicus briefs have pointed out in terms of loss of innovation and standing and a lot of the other consequences of an ineffective patent statute. Well, Your Honor, the concern that Your Honor seems to be articulating is with respect to deliberate avoidance of a patent and a circumstance in which there is, and in this case, there was a strict liability instruction. And it's under Section 271A, and it was the instruction that Akamai relied on at trial was they don't even need to know about the patent, they don't need to be aware of it, there's absolutely no requirement of intentionality. But that's different, isn't it? That could be to the damages. If you're innocent, even if all of the steps are performed on an innocent contributor, there was one step out of a hundred or one step out of two. Maybe a joint tort fusion, but not full liability. The concept of separating what it costs you as a participant in some form of joint infringement and whether, in fact, the patent is being practiced is a question that we'll need to get to, but perhaps that's where the solution lies. Your Honor, may I answer? Please. Thank you. Your Honor, as between the parties who are conducting themselves in a way that does not fall within the literal terms of the patent as written and as granted by the patent office, and the drafter who has the ability to draft the patent to reach the... No, I was postulating that it did fall within the literal terms. I understand, Your Honor, that you're saying that all of the steps were performed, but if the method claim is written to encompass several steps and none of the parties are carrying out those steps, either directly or vicariously through the conduct of one that can be imputed to them, then no one has used that patent. One can easily imagine just how threatening a patent system would be in which the performance of multiple steps by parties with some association could lead to infringement. Well, of course, some wise court would have to devise a rule of reason or something else in terms of the degree of liability, but perhaps we really do need to focus on separating whether, in fact, the patent is practiced by one or more entities and the extent of culpability of the various practitioners. Well, Your Honor, I think that what's clear is that when you're talking about Section 271A and literal direct infringement under Section 271A, the idea of importing into that notions of intentionality and state of mind would really be quite inconsistent with what has always been a very fundamental distinction between direct infringement and other types of infringement that are based on very elevated standards of knowledge, and one wonders why there would be any need for someone to allege contributory or inducement, especially inducement of infringement of a method claim if one could simply say, well, you're associated with that person and contributed to their having performed those steps and that's enough to hold you liable as a direct infringer. And so I think that it's critical to maintaining what Congress intended in terms of the division between direct and indirect infringement to maintain the integrity of the direct infringement toward under Section 271A. Thank you, Mr. Klamer. Thank you, Your Honor. I want to respond to a few of the points that were made, particularly the claim drafting point raised by Judge Crouse, but first I'd just like to mention preliminarily that we have a jury verdict here and there was a proper instruction. Nobody objected to it. Direction and control. The jury found there was infringement and there is substantial evidence, I submit, that there was at the very least direction. And the direction comes for all the reasons I've mentioned. There was direction in the extreme in this case. They were told what to do. They were given a tag. They were told how to apply the tag. They were given engineers to help them apply the tag. And wholly aside from what I feel is a contract, which should apply for those who actually use it, not for those who don't use it. Those who used it were bound by this contract to perform certain things. As a matter of process, before you move on, if we should agree with everything that you just said, what is it that you would suggest that this Court would hold? And the particular question I have is, are there other open issues that were presented on appeal that either weren't addressed by the panel and need to be or what? Because we didn't look at the whole case. And I believe I might have argued that they appealed certain other issues. Your Honor, if you agree with me on that, I think you should reverse the Jane Law decision below and remand for reinstatement of the jury verdict. I don't think there are any open issues as to the claims we're talking about on this appeal. Now, as to, I'd like to address the claim construction point that Judge Crouse talked about. There's a problem with claim construction. The Golden Hour case is a perfect example. There are two collecting staffs and one integrating staff. And what they had was a strategic agreement to divide up the thefts, and the end result was no infringement. In the appealist case, the— Mr. Panner seems to think maybe that wouldn't come out the same way under his test. Pardon me, Your Honor? I understood Mr. Panner to suggest that if there's a joint activity like that, it could be direct infringement. Well, but the court held there was no direct infringement. The court held there was no infringement in that case. Maybe we were wrong. Well, I think you were wrong twice. But the fact is, it's not only the Golden Hour case, the peerless case, the Shields case. If you look at the claims in those cases, they were designed to be acted on by one party, and clever people will find ways to do it. In fact, we refer to an article by Tanya Gray in which she tells people to have a contract in which nobody's the mastermind. Yesterday I read an article by a fellow who was suggesting outsourcing one or two steps. No matter how you write those claims, there could be a problem. Moreover, this invention, the invention of the Akamai case, and many other inventions, are by their very nature joint inventions. The whole concept is to have two people operate so that you give one party expanded capacity by permitting him to shift demand for the objects from his or her servers to somebody else's servers. Under Section 1, Shelf Paragraph 2, if you're going to particularly point out and distinctly claim that invention, you need to be able to claim a joint invention. And it is totally an arbitrary thing, I suggest, to cut that off and to prevent people from doing that. Mr. Panner acknowledges that I think he's wrong on the single actor rule. He construes whoever to mean that each party must individually infringe. The commentators in the cases don't support that. But he's already bridged across the line. He's already said that it's okay to have agency or contract. If you look at the restatement of agency, Section 212, they tell you that they talk about directions and they say it's not dependent on the law of agency. So the restatements are indeed to agency, and I think that once the line is crossed, there's no logical reason to stop the line at the point at which he says it should be stopped. He mentioned that the early pre-'52 cases are contributory infringement cases. I assume he has in mind peerless. If you look at peerless, they may have used the word contributory infringement, but that was not a contributory infringement case. There were two actors. They divided. There was no direct infringement under his definition. I submit, Your Honors, that given the law as it existed before 1952, given the commentators, given the restatements, given the logic of the situation, given the huge gaping hole that is created by this arbitrary rule, I suggest that there's no reason to draw the line as you do. But even if you do, we can satisfy the direction of control test. Thank you, Mr. McDonough. The next case is McKesson v. Brewer. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A7093. Public record. Not legal advice.