Oral argument — California Dept. of Toxic Substances Control v. Northern Burlington

Please accord. Aaron Avila on behalf of the United States, with me at Council's table are Jamie McHale with the Department of Justice and Reid Sato for the State of California. I will be presenting argument for both the United States and the State, but Mr. Sato will use a few minutes of the government's time. I hope to reserve some time for rebuttal. The government has brought this action under the Comprehensive Environmental Response Compensation and Liability Act to recover their costs for remediating the highly contaminated Brown and Bryant facility, a former pesticide storage and distribution business. At the time of the district court's decision, the government had spent about $8.2 million in cleanup at the site. The government sought recovery from Shell Oil Company, as well as Burlington Northern and Santa Fe Railway Company and Union Pacific Railroad Company. The district court found Shell liable under CERCLA as an arranger for the disposal of a hazardous substance. The court found the railroads liable as the owners of the facility. The railroads have not appealed the liability finding by the district court. Although the court found both Shell and the railroads liable, it did not hold them jointly and severally liable for the single harm at the site. Instead, based on what the court called an equitable apportionment that the court sua sponte devised when it was crafting its opinion, the court concluded that Shell is liable for only 6% of the government's response costs and the railroads are liable for only 9%. This leaves 85% to be paid by the taxpayers. The district court erred when it reached that conclusion because it failed to hold the defendants to their burden of proof of establishing that the harm at the site is divisible. The court also erred because the divisibility inquiry is a question of causation alone, not one of equity. The court's analysis was guided by equitable considerations, not those causation principles. And finally, the district court erred because its analysis lacks evidentiary support. The parties agree that those liable under CERCLA are jointly and severally liable for the government's response costs unless the harm at the site is divisible. The parties also agree that the harm at the site is divisible unless the harm at the site is divisible. Kagan. What harm are we talking about? Is it the – it's not the leakage. Is it the contamination? Is it the cost of cleaning up the contamination? What is it? The harm is the contamination and the cost to clean up. We are – the government's Those aren't necessarily the same. That's why I asked the question. It is the – Whether it's divisible. I mean, you might find that the – that the cost, for example, is divisible because only certainly the – certainly the contamination is cleaned up, you know, and others isn't or – but it's – it matters. It's who's – it's who caused the contamination at the site. And showing that each defendant has the burden of establishing, since this is a single site – a single harm, pardon me, a reasonable basis for determining that defendant's contribution to the cause of that single harm. And the contamination is what's remained and has to be cleaned up. I'm sorry? The contamination is what has remained and has to be cleaned up. Correct. It's not what originally leaked, which – some of which may not be there anymore. Correct. It's – well, it's the current – it's the contamination that governments have to remedy, whether – I mean, one could show – so it's the current contamination as of, I guess, the time of the complaint, if nothing else. Just as a matter of curiosity, is there further remediation that needs to be done? Yes. The final remedy has not been selected yet at the site. So – Looks like a very difficult problem. It is, Your Honor. It's a highly contaminated site and it's a difficult – difficult one to remediate. And the governments are doing their best. The district court failed to hold defendants to their burden of establishing divisibility. With respect to the railroads, the district court specifically found there is no evidence to quantify the difference in the volume of releases from the railroad parcel or the B&B parcel. The court further noted that no party has specifically documented the relative contribution of contamination from either parcel. With respect to Shell, the district court found that Shell did not present evidence of how its product's contribution to the contamination at the Arvin facility can be apportioned. That failure of evidence – You're not making a waiver argument, such as was made by the State, as I understand it. No, it's – You didn't raise the issue. They – they put forth a theory that they were not liable. The district court – All right. That they were not liable for any – So ultimately, doesn't this all collapse into one problem? I mean, which is whether the way in which he chose to apportion it is adequate or isn't adequate? I mean, why does it matter – there is evidence in the record, and he used whatever he used, and it either works or it doesn't work. And why do we have to go through the question of whether they presented a theory or presented the evidence or whatever? Because it's their burden, first of all, to establish a reasonable basis. And they either did or didn't do it. And they didn't do it. And the government responded to the theory that they put forth. And their theory that they put forth at trial was that you can't – there's none of our contamination that we're responsible for is attributable to the contamination that you're remediating. That was their theory at trial, and that's the theory that the government responded to. But if this was a jury trial, you would have instructions stating what the law is, and you would have a number in the end, and then we would look and see whether or not the – there was substantial evidence to prove by preponderance to support what the jury found. And I don't understand why we're doing anything different here. I don't know whether theories matter. All that matters in the end is whether or not the evidence that the judge relied on for divisibility matches the legal standard for divisibility. I guess the point is, Your Honor, the reason the theory matters, because there's a threshold question of whether there's a reasonable basis. The government – under the way the district court went about it, the fact that the district court may or may not be able to cobble together some other divisibility theory really puts the government at an extreme disadvantage. How – the government had no idea to know that the district court was going to try and use the relative parcel size to determine divisibility, because the theory at trial was we didn't contribute to the contamination. If there's anything that's wrong with the relative parcel size, you're welcome to please tell us, because there certainly are problems with it. But I don't know why it's helpful to start three stages back. That's all I'm saying. I was just explaining what the prejudice to the government is during the trial phase, because we had no opportunity to respond to this theory. The problem with the parcel – using the parcel sizes is it has no bearing, no relationship to the relative contribution of contamination at the site that the government is mediating. There's no – the district court just assumed, without any basis in the record, that the level activities corresponded to the parcel size and the length of time that the railroad's parcel was part of the facility. And there's absolutely no basis for concluding that that approximates the amount of harms caused by the railroad parcel. With respect to Shell's portion of the supposedly divisible harm, the district court's analysis rests upon assumption upon assumption that the government had no opportunity to respond to, because that wasn't the theory at trial. The theory at trial was that Shell argued that they didn't contribute any contamination. Had there been – the district court provides no basis for concluding that, for instance, the level of contamination is directly proportional to the amount spilled. The district court is – Kagan. Aside from the arranger argument, how did Shell argue that it didn't have any contribution to the – to the contamination? I mean, the stuff did spill there. So what was their argument? That it never made it to the groundwater was their theory of the case. That – that spills that were arranged for by Shell had never made it to the groundwater. So with respect to that theory, with respect to the district court's theory that was – came up when it was crafting its opinion in this case, the underpinning of the entire analysis is the assumption that every gallon of DD spilled would result in the same uniform amount of contamination. But there's no evidence, expert or otherwise, to support the validity of that theory. In fact, the court itself specifically found that even a small spill of DD can cause substantial groundwater contamination. In crafting its own divisibility theory, the court failed to take into account the fact that spills could have been large or small, and that the location conditions may vary, and therefore the resulting contribution to groundwater contamination would differ substantially. In sum, neither the railroads nor Shell offered any theory of divisibility at trial. They only contended that they had no liability at all, and the district court rejected that contention. In such circumstances, courts have not hesitated to hold CERCLA defendants jointly and severally liable, and the district court erred when its sua sponte attempted to cure the defendants' failure of proof. Sotomayor, can we go separately through the railroads and Shell with regard to the divisibility? On the Shell, the judge did a series of calculations. There were approximations, but they were approximations based on some detailed records. Why weren't they good enough? Well, for instance, he based part of his calculation on activities that took place with respect to bobtails, with respect to one piece of equipment, and applied them to a separate piece of equipment, with no basis for saying why it was appropriate to use spills associated with one activity as being similar to spills associated with another activity. The district court also only used four years of sales records to start its calculation off of the amount of DD that was delivered. There's no explanation as to why those four years are representative. There's no expert testimony. I mean, in a lot of the divisibility cases, you have expert testimony saying why it's reasonable to make various assumptions or why, in one's expert opinion, it's appropriate to devise a harm like this. Here, there is no such evidence. So that is why that's not good enough in this case, Your Honor. And then further, you still have to make the next step from the spill of the DD to the groundwater contamination. And again, the amount of spillage that reached the groundwater will vary depending on the site conditions, the conditions of the spill, the site conditions, how the spill occurred, and things of that nature. So those things are all very relevant and aren't dealt with in the district court's opinion, and were not dealt with at trial by the defendants. With respect to the railroads, there's really just no basis at all for assuming that one-tenth of the site contributed one-tenth of the harm, which is the beginning of the amount. There's certainly some basis, which is essentially the – tell me if I have my facts straight. But there was no remediable contamination on the railroad parcel. Is that right? Correct. Okay. So anything that leaked on the railway parcel and didn't get carried off to the other parcel isn't a problem? Correct. So the only thing we're looking at is material that got carried off to the other parcel, whereas on the other parcel, we're talking about some stuff that just was spilled there. So at least on that basis, you need the transport to get from the rail – of water to get from the rail parcel to the other parcel, whereas on the other – on the original parcel, some stuff just went directly into the ground, and there's a problem for that reason. And the disturbing part of this, frankly, is that the reason a railroad doesn't know what's happening is because they weren't on the site. And I understand that they're responsible under the statute nonetheless, but the net result is that the least sort of directly culpable party is the least able to defend itself because they weren't there. Well, the district court did make a finding here that the railroads were aware of the activities. I understand they were aware, but they weren't there on a daily basis, and they didn't have their own percipient witnesses to be able to say what happened. If I do remember correctly, though, they did eventually add a term to the lease to require that the lessee use the lease in conformance with environmental laws. So it did know there was a problem going on there. It did seek indemnity and got indemnity. Of course, indemnity against someone who's bankrupt isn't much, but they did seek indemnity, didn't they? I believe that's right, Your Honor. Just briefly, with respect to Schell's cross-appeal on liability, the district court properly examined the totality of the circumstances in this case. Schell arranged for the delivery of DD to this facility. Schell owned and possessed the DD at the time it arranged for the delivery. Schell effectively controlled the shipment, delivery, and unloading process. Schell hired, paid for, and provided instructions to the common carrier that delivered Schell's DD to the facility. Schell knew that spills were an inevitable consequence of the arranged for delivery, and that Schell also knew that such spills occurred throughout the period over which Schell sold DD and arranged for its delivery to the Brown and Bryant facility. On appeal, Schell basically tries to take single factors and make them dispositive. And, in fact, the district court properly looked at all of the totality of the circumstances and found Schell to be an arranger. I'd like to reserve the balance and let counsel for the State of California. Thank you very much. Thank you. Good morning, Your Honors. Reed Sato, Deputy Attorney General for the State of California, representing the Appellant and Cross Appellee, California Department of Toxic Substances Control. As Mr. Avalos said, he is going to be making the principal arguments for the government on this issue. I just wanted to make a brief observation, and that is that nowhere in this extensive record that you have before you are you going to find an effort on the part of either Schell or the railroads to present any kind of percentage or volumetric basis for the divisibility of liability that has been assigned to them. I thought the railroads did try, but their way of — their method was rejected by the district court. Well, the railroads basically argued for a zero liability, as Schell did in a different kind of manner. Schell is saying as a matter of law we're not liable. The railroads are saying, look, based upon these factors, we didn't contribute anything to the site. You're correct. It's been rejected. But I don't think that that's the kind of divisibility analysis that the courts have required in other cases whereby a defendant has an affirmative duty to demonstrate a theory of liability and a basis for that division of liability. In addition, I think that absent that showing — Kagan. Was there a theory that none of the contamination from their site got onto the other site? Was that their basic theory? The theory is that none of their contamination caused there to be an incurrence or response caused by the government. Because it didn't get to the — to the other site. It didn't get — didn't get to any place that required remediation. Either it didn't hit the groundwater or it didn't cause soil contamination such to require remediation. Finally, I just wanted to mention briefly that, you know, the thing that we find most troubling about this case is the courts, the trial courts' efforts to try to use an equitable apportionment theory in order to make this kind of legal determination about causation. Causation, legal causation is the basis for making the divisibility determination when you have a single harm. That's what all the circle cases teach us. However, in this case, the trial court didn't look at any kind of — demand any kind of causation-based analysis. In fact, it observed that, in fact, that the railroads and shale had not been able to demonstrate a fundamental basis for causation, and yet went ahead and made an equitable apportionment, which is not proper in the reported cases. It is what he — he used that language sometimes, and there's no doubt that his — it seemed to me that his motive was that he thought that this was basically unfair. But ultimately, we have to judge the basis on which he did this as to whether it's adequate, rather than impugning his motives, don't we? Or can we just reverse, because he talked about equity at some points? Katyala, I think you can look at the record before you, and you can decide that, had the court proceeded based upon the findings that it made, that, in fact, there had been no demonstration by the railroads of certain facts, there had been no demonstration by shale of certain facts, that you can then conclude that they didn't meet the burden of proof, and therefore, there's no basis for making a divisibility-of-liability determination. And I'd like to reserve the rest of my time. Thank you, Your Honor. Roberts. Good morning, Your Honors. May it please the Court, my name is John Barg. I represent the railroads, Burlington Northern and Santa Fe Railway and Union Pacific Railway. With me at council table is one of my partners, Mark Gazzett-Patello and Mark — Mike Johnson, representing Shell Oil Company. I will have 14 minutes of our side, and Mr. Johnson will have 16. I'm curious about one thing. Yes, Your Honor. Your opponents in their briefs say that nowhere in the pretrial order was this divisibility issue mentioned. And I understand that you take the position that it is. Well, I've looked through some of this record, and I'd just like for you to tell me exactly where divisibility was raised as an issue that had to be litigated in the case. Yes, Your Honor. In the pretrial order, it is at the excerpts of record, page 85, paragraph 13. Page 85, page 13. There's a statement, railroads cannot establish any basis for divisibility. Right. Because there's no way to quantify where the hazardous substance is coming from. And they — Therefore, there's no basis for apportioning harm. And they make a similar statement acknowledging divisibility as a trial issue in the same pretrial order at page 96, paragraph 10. That's the government. Yes, the government. We put it in, Your Honor, page 1 — pages 112 to 113, paragraph 5. Read that, would you please? Would I, Your Honor? Would you read it? Sorry, Your Honor, I didn't bring volume 1, but one of my colleagues. No, we didn't. If you can't read it, can you tell me what it says? You aren't. Here it is. In pertinent part, Your Honor. Yes. It is well settled that where there are distinct harms or a single harm capable of being reasonably apportioned among several causes, liability is several, not joint. It then cites the Restatement, Second of Torts, Section 433A, and cites In re Bell, the Fifth Circuit case. And where in the record do we find that? Sorry, Your Honor? Where in the record do we find that? It's excerpt of record, page 112, paragraph 5. And then you go out and you basically lay out the theory, which was the one that you did try. I'm sorry, Your Honor. And you go out and you lay out the theory which was basically the one that you did try and which was rejected by the district court. Well, Your Honor, that's right. The railroads argued for a zero share. The government argued for 100 percent liability under joint and several liability. Both extremes were rejected by the court. And as Your Honor pointed out, this isn't sort of a novel approach that the factfinder would make a determination of percentage share, for instance, in a comparative fault case. Using all of the evidence that he had. And the notion of the law. I did leave – it does mean that the – there was something of an evidentiary vacuum, because you didn't try to – since you didn't at any point try to demonstrate on the thesis that some of the contaminant came from your parcel, there was really no detail. And that's why he went to this very gross sort of, you know, percentages of land and time. And then the part I really don't understand that I would really like you to explain to me, which is where he got the two-thirds from. All right, Your Honor. I don't – I don't agree with Your Honor's characterization that he made sort of a gross calculation based on – on limited evidence. This was a judge who worked hard through a 27-day trial, 36 witnesses, many of them presenting highly technical – He was totally frustrated at the end because he didn't have the evidence. So then – and it was quite apparent that he thought this was all very unfair, and he was going to try and find some way to find a basis for divisibility. But the evidence – I'm sorry, Your Honor. I haven't read a complaint by a district judge like the complaint in this case in 23 years or 24 years of doing this sort of thing. I mean, it was – and then he comes up with this whole construct on it. And that's – and that goes back to whether the issue was raised or not, I guess, as far as I'm concerned. Well, it most certainly was raised, and the government today talk about the railroads and Shell not offering this theory. They didn't fulfill their obligation to present a theory. The obligation is to present proof. We have 191 pages of findings of fact and conclusions of law from this judge who did work really hard through this case. And the evidence completely supports in an undisputed way the time of operations. Mr. Avala says it's irrelevant. It's irrelevant that Brown and Bryant, whom the – whom the district court found, engaged in release-causing activities every day of its operations, leased the parcel from the railroads in 1975. So there were 15 years from 1960 through 1975 – I guess that's 16 years – of operations by Brown and Bryant without the railroad parcel. And in that period of time, he used sumps and ponds on the Brown and Bryant parcel to discharge contaminants. All of the experts and the district court found that more than 99 percent of the contamination at the Brown and Bryant site is related to those sumps and ponds. Those sumps and ponds remained unlined until 1979. That's what the court found. So during the 15 – The bottom line of what you're saying, and it's very hard to quarrel with, is that definitely the railroads were not responsible for all of it, and they maybe weren't responsible for most of it. All right. Is that good enough? And then what do you do with that next? It is good enough, Your Honor. I mean, what this case comes down to, I believe, are two issues, the burden of proof and the standard of review. The governments seem to be looking for some degree of certainty in the proof that's greater than a preponderance of the evidence. We have this judge making findings that Brown and Bryant, for these 15 years, actually 20 years prior to lining those sumps, and that's 16 years of operations without the railroad parcel, engaged in contamination in those sumps and ponds. So the question becomes, well, what portion of the – of the contamination was caused by something that happened on the railroad parcels? The district court found that there were these continuous columns of contamination from the surface of the soil all the way to groundwater on the Brown and Bryant site. There was not evidence of one on the railroad. Yes, but he also found that there was runoff from the railroad parcel that would have found its way into the sumps and the ponds. He did indeed, Your Honor, and yet – and yet there was no trace of that, or certainly not a trace. But now you're quarreling with his – with his findings. That's different. What I'm – what I'm saying, Your Honor, is he did find the railroad contribution there, and he – and he based that on evidence that the – that the surface soil ran in that direction, and yet the surface soil on the railroad parcel didn't require any remediation. The contamination levels were within – within guidelines. But you're not in a position, and I don't think you want to quarrel with his findings of fact. I don't. Okay. I don't, Your Honor. You're doing it. I'm – I'm – this – this, too, was – was within Judge Wainsworth's findings of fact. The government at trial pointed to the swale at the south end, kind of a low point that would have created a collection point for any surface runoff before it went onto the Brown and Bryant site. And the district court found that the soil on the railroad parcel at the swale was uncontaminated. You know, counsel, this case is made very difficult because of the bankruptcy of B&B. In an ordinary – orderly presentation of this type of case, you would not struggle so hard with divisibility, which is a very difficult thing for any of the parties to establish. And then, when you come to the contribution phase, all of these arguments would be very good, and you'd probably get B&B responsible for most of it. But we've got the problem here simply of we've got this contamination. We know it's been contributed to by the railroad and probably by shale. And no real way to find the division of responsibility for parts and parts. And so that's what I see myself faced with. All right, Your Honor. You know, in a sense, Your Honor, the court used the parcel size and the duration of the railroad involvement, in a sense, as surrogates for sort of a volumetric share, thereby overstating the contribution that could come from the railroad parcel. And the court, of course, made factual findings that the railroad parcel could not have been responsible for as much as 10 percent of the site contamination. The fact that there is this B&B share and they are insolvent, Your Honor, well, no provision of CERCLA 107 allows some sort of reallocation of that. The court did the right thing in allocating among all defendants, including B&B. Oh, wait. The statute doesn't cover allocation at all, and the restatement does seem to provide the possibility of some special consideration where there is a bankruptcy. That's right. And the district court's discussion of that is a little peculiar. So there is a possibility, which wasn't really addressed at any particular length by the district court. Right. Well, he did. This was — this is what bothers me about the government's argument that the district court improperly engaged in sort of an equitable analysis and didn't consider causation. That's just not true. If you read the 191 pages, you find the separate section on equitable allocation And he does at some point say he's doing an equitable analysis. I don't know why that matters. I'm not sure I'd be, you know, disposed to reverse because he used the wrong language at some points. But he did use the wrong language at some points. Well, he did use — he did discuss equitable allocation after discussing the appropriate standard, the causation-based standard, and putting in all this evidence, this mountain of evidence that he found based upon his evaluation of all the witnesses and their credibility and so on. But it was the government that — sorry. None of that evidence went — seemed to have anything to do with the numbers he came up with, because then he went all the way back to these very basic facts about the size of the parcel and the amount of time, and then I really would like you to explain to me this two-thirds, where that came from. All right. And so he did do mountains and mountains of evidence, but it didn't seem to inform this decision. But I believe it did, Your Honor. The two-thirds — let me cover that real fast — is there were three chemicals of concern on the site, DD, Nemagon, and Dynaseb. And the Court found that one of the three chemicals, DD, was not an issue on the railroad parcel. It was. It was there just as much as the others. There were — there were apparently tanks there that were — where did he get them? No, no. There were no installed tanks. There was — sorry. There were — there were vehicles parked. All right. But it was on the site. And so on. But there was no evidence that this volatile DD caused any response costs and so on on the railroad parcel, or that it could have been involved in off-site transport, surface transport back to the pond. Your Honors are aware, I think, that — getting back to the Brown and Bryant share for a moment in response to your question, Judge Fletcher, the Eighth Circuit in the Hercules case made the observation that the purpose of divisibility is, in the words of the Eighth Circuit, a sensible way to avoid imposing on parties excessive liability for harm that is not fairly attributable to them. And, of course, that's what the — that's what Judge Wanger was really concerned with. And — and, obviously, he struggled with the best way to accomplish that. And — and he — he based his ultimate decision on all this evidence concluding that railroads can't be responsible for the first 15 years of Brown and Bryant operations when that wasn't — the leased parcel wasn't part of the Brown and Bryant site. So he took the rest of it, and he said, you can be responsible for that percentage. And he — and he — he took that portion of the entire facility and carved out the — the railroad parcel and said, you can only be responsible for that percentage of this overall site. And — and then he — he assumed in that process that there were releases every day throughout that period of time on all the parcels in order to reach this ultimately 6 percent railroad share that he then augmented to reach a conservative number. He was bending over backward to reach a conservative number that — that would not be unfair to the government. Now, you brought up Hercules. Yes, Your Honor. Which my brother Woolman authored. And this is indeed a tour de force on all of this subject. But I see — I see in it the statement, thus where causation is unclear, divisibility is not an opportunity for courts to split the difference in an attempt to achieve equity. Isn't that kind of what the district judge did here? Oh, no, Your Honor. I don't believe so. With all respect, I — I think it's quite clear from his findings that he knows and we know and Your Honors know that the cause of the contamination that resulted in response costs was the sloppy operations on B&B, was the unlined sumps and ponds that existed for 15, 16 years before the railroad parcel even became part of Brown and Bryant. So he wasn't splitting the difference. He wasn't basing his decision on an absence of evidence. He used causation-based evidence. And in — in order to reach a result that was consistent with the component of — of Your Honor's colleagues' Hercules decision talking about divisibility based on geography, based on time, based on volumetric share, or — It went down the whole drill, Your Honor. Or he says other evidence. And he went through all of them. And — and — and this is — this is the decision he reached. Much like Your Honor noted a jury would — would reach. He looked at all of it, and he made a decision that I don't believe this — this Court, sitting here now, could say, well, we'd have done it differently, or — or we think he reached the wrong result. This is a result that is supported by all of that evidence that — that — that he laid out. It might not be the way Your Honors would do it, but — but it's not clearly erroneous to have — to have used those factors and done it in — in the way he did. You are running down your time on the — Sorry, Your Honor? You're running down the time in terms of the allocation you suggested. That's up to you, but I just wanted to — All right, Your Honor. Well, I — I will turn it over to — to Mr. Johnson. We do have a — a — we — we recovered a — a — our share from Schell, and I do believe that Judge Wanger's decision with respect to Schell as an arranger, based on the totality of the circumstances, was a correct decision. Thank you, Your Honor. Good morning, Your Honors. Michael Johnson, representing Schell, in this case as appellee and cross-appellant. We would join in much of what the Railroad said with respect to the trial court's effort and the trial court's causation-based analysis of divisibility. It's clear this trial court did a lot of work. Governments have argued that this was some kind of equitable analysis that he did, that he was splitting the difference. The proof of the pudding is in the eating, okay? To determine what this trial — what the trial judge did, you need to look at what he did. He did not allocate Schell a 6 percent share — divisible share based on equitable considerations. He went through, looked at the evidence, made calculations based on the evidence, to arrive at a 6 percent volumetric contribution, okay? Case law is quite clear that that's appropriate. It's permissible. And that's what he said he did, and that's what he did. There are references that — I'm sorry about that statement. Wasn't there some statement in there that this — this screams out as inequitable, works to that general effect? He did say that. And, you know, I don't think that the fact that the judge thought that his divisible decision was the right decision makes it legal error. In fact, I believe the fact that the divisibility analysis coincides with his sense of fairness makes it likely that it's a good result. Well, but he also — I mean, he did, both in his original — in his findings of facts and conclusions and also in his opinion on reconsideration, he did keep talking about, like on page 487 of the — of the excerpts, he talked about equitable apportionment reached by applying all the facts to the law. And I went through and looked, and he did keep saying this, not once but many times. So the question is, what do we make of that? Well, I think we make of that that there's some loose language in a decision that the judge wrote over a three-year period by himself without the assistance of law clerks based on a voluminous record. So I don't — I think that's where the inconsistencies come from. Moreover, there were equitable issues raised and addressed by the judge, including the orphan share issue with Brown and Bryant, and including — remember, the railroads had a cross-complaint against Schell under 9613, which is a contribution analysis that would be an equitable apportionment using the Gore factors and other related factors. And, you know, there are many other places in the findings of fact where the judge talks about the law, and it's quite clear from those proportions that he understands that there's a difference between, you know, the divisibility analysis, which is causation-based, and an equitable apportionment which relies on Gore factors and the Gore factors and the other equitable considerations that the Court may want to look at. It's also notable that the governments asked this trial court to make the equitable allocation with respect to the orphan share. So to the extent they forced the government — they forced the trial court in a sense — I shouldn't say forced. They asked the trial court to look at these issues. So it's difficult for me to hear them now complain about the fact that some of that language pops up in the decision. But I think the bottom line is the fact that there may be this kind of equitable references throughout, that's irrelevant, okay? As Your Honor said, that's the judge maybe, you know — what's relevant is the fact that he looked at the evidence. Based on the evidence, he calculated Schell's volumetric contribution. He looked at the evidence, calculated other sources of spills, where he had evidence. When he didn't have evidence of other sources of spills, he didn't rule them out, or I mean he ruled them out, didn't include them, and he made a calculation. Is that a reasonable basis for the allocation, which is what the law requires? We feel it is. It's important, and I think Mr. Barg noted this, that whenever the trial judge could, he erred in the side of favoring the governments, okay? For example, he assumed that there was a three-gallon spill during each and every delivery of DD to the site. Well, the testimony by John Brown, the owner of the site, was it was maybe a couple — maybe a cupful to a gallon. Lonnie Merriman, who worked there, said maybe a quart to five gallons. Well, if the judge had adopted John Brown's one-gallon estimate, then Schell's proportionate share would have been 2 percent and not 6 percent. So the judge really, when there was — when he had a range, he tried to do — he tried to err on the side of favoring the greater liability on the part of Schell. But in the end, his numbers for Schell were based on spills, and they weren't based on any notions about what got into the — to the groundwater and why. That's correct. The judge assumed that every molecule of DD that was spilled during the delivery No. Or conversely, he assumed that DD was just as toxic as other things instead of not — instead of more toxic. No, he didn't really deal with the question of whether there could have been a disproportionate amount of the DD that got into the — into the groundwater. There was no evidence in the record that would have required him to do that, Your Honor. This was — this cleanup was primarily DD. The DD was one of the three or four chemicals of concern that was in the groundwater, vastly larger quantities. It was what the cleanup was aimed at. And cleaning that up, I'm not even sure that — I'm not even sure that the other chemicals in the groundwater reached the levels of contaminants of concern. This was a DD site. The judge was measuring or comparing apples to apples. He was comparing DD spills with DD spills. It wasn't like he was comparing some really hazardous substance. You know, there was a gallon of plutonium versus, you know, a bunch of barrels of something that wasn't quite as toxic and didn't migrate. So in comparing apples to apples and — Was there any evidence in the record as to when Shell knew that this product was so toxic that it should be taken off the market? You know, I know it was — there's evidence in the record of when it was taken off the market. I don't know that there was evidence that we left it on the market for a period of time after being told it should be taken off. As far as I know, there was no evidence one way or the other. We — we pulled it from the market and stopped selling it when it became apparent that we needed to do that. It stopped being a useful product at some point. Well, the useful product issue — and I ought to segue into the cross-appeal if Your Honors have no more questions, because I'm running out of time. Go ahead. Thank you. Okay. Is there no more questions on the — the divisibility issue? The useful product issue, Mr. Avala mentioned the totality of the circumstances test, and I'm going to get to your question, but the totality of the circumstances test, in our view, was not met here. The totality of the circumstances test, had it been applied, the judge would have considered the fact that this was a useful product at the time it was delivered. He would have considered the fact that ownership transferred upon tender of delivery at the site, and that — he would have considered the fact that Brown and Bryant owned the product after they accepted tender of delivery and at the time it was being He would have considered — Well, what does that mean, exactly? I mean, there is a — the contract gave Shell the authority to choose the carrier to bring it to the site, and when it arrived on the site, it was in the truck of this carrier, right? So whether — whether Brown and Bain owned it or didn't own it, the fact is that it wasn't in — certainly not in sole control of getting it in — to there. Here's the way we look at it, Your Honor. Brown and Bryant owned it. Common Carrier is an independent contractor. The testimony was undisputed by Mr. Brown and Gary Leary, who was the Common Carrier representative that the government paid to come and testify. Between them, they both testified that they would show up and then the Brown and Bryant representative would say either unload it over there or the Brown and Bryant person would go and do it. Who's arranging for that to happen? Shell's not making that decision. It's the Brown and Bryant people that are making the decision whether they're going to unload it, whether the independent Common Carrier is going to unload it. Brown and Bryant is the one making the decision of how it's going to be unloaded. There's no one from Shell there. We provided some safe handling guidelines as instructions for them to follow to do it safely, okay? We're not making those decisions. Kagan. Here's a situation where you have a lot of very specific findings by the district court, and he has — his specific findings were essentially that as a practical matter, the Shell-designated driver was the person who was doing this and not — and not Brown and Bain, and that for the most part, and a set of other findings as well, which you have to find clear error in in order to find otherwise. I'm talking about the discrete findings, not the overall finding now, but the underlying findings. I understand what Your Honor is saying. I'm running out of time. We'll give you extra time. Go ahead. You're going to have a few extra minutes. Go ahead. The trial court here with — the primary error on the cross-appeal is a legal error. The primary error is the fact that the trial court applied the wrong legal standard to the arranger liability determination. Because the ownership's all-important, or because why? The totality of the circumstances. First of all, he didn't apply the totality of the circumstances, but I'm going to talk about a different issue, which is the United States v. Shell case. When you consider that case in consideration with the facts of this case, it becomes very clear that the trial court committed legal error by applying the wrong legal standard to finding arranger liability. Why is that? United States v. Shell. Tables were turned. The oil companies are arguing that the United States is liable as an arranger based on its involvement with the production of aviation gas during World War II. So, primary factor, the crucial factor that the court identified in United States v. Shell was control. Okay? Control. The other factor that the court talked about in United States v. Shell is knowledge. That is, did the United States have knowledge that these wastes were being produced as concomitant to making the aviation gas? Control is a crucial factor. You have knowledge. What do you have in this case? Well, you have in this case is somebody who, as the statute required, owned the product at the time it arranged for someone else to bring it to the site, knowing, the judge found quite specifically, that there was going to be a certain amount of spillage when that happened, and not doing anything or doing some things to prevent that, but not adequately. Right. There's a big difference, Your Honor, between an arrangement, arranging to deliver a useful product. A useful product and some waste that you know is going to occur as a result. That's the knowledge factor again. This is why I'm saying the judge didn't apply the totality of the circumstances test, because he disregarded that we arranged for delivery of a useful product. Now, there's knowledge. Okay. Let me get back to the United States v. Shell, because that case says quite clearly that actual control is required. Actual control is required. If you only have the ability to control or an opportunity to control without actual control, then you are not an arranger. Here, this judge made a specific finding that Shell did not have actual control over the unloading process. He nonetheless imposed arranger liability. The United States v. Shell, and I just want to read a brief quote from that decision because here's what the Court said in the United States v. Shell. Am I not correct that in the United States v. Shell, the United States did not own that product at any relevant time? Oh, they did. Only at the end. Only afterwards. What's the difference between owning a useful product before and owning a useful product afterwards? There's a big difference. The statute talks about arranging it by the owner or the person who owned it at the time. Okay. Well, I arranged to get my useful product to you. You arranged to have it manufactured so I can buy it from you. I don't see that as a, you know, that's a distinction, but in my view, not a significant distinction. The Court said if authority to control the oil company's waste disposal were sufficient Authority to control is not legally sufficient, even where there is knowledge, because the United States in the U.S. v. Shell case, remember, had knowledge. The United States had opportunity to control and knowledge. In our case, Shell had the opportunity to control by way of safe handling instructions where it was trying to prevent spills. And the Court found it had knowledge, although that was hotly disputed, and I'm not going to concede that even today because we've been fighting that fight a long time. We didn't have knowledge that this stuff was going to contaminate the environment. It would have been very easy to prevent these spills from contaminating the environment. Well, you knew that there was going to be spill when delivered, correct? What's that? Shell knew that there would be spills when the product was delivered. Correct. Shell may have known that there was an opportunity for spills sometimes or that spills might occur sometimes. Shell didn't know that those spills would not be caught in a bucket, would not be caught on an impervious concrete pad, would not have been cleaned up if they did get under the soil. Those are the things that our instructions recommended Brown and Bryant do. If they don't do that, if they don't follow our recommendations, how can that possibly be a basis for imposing arranger liability on us? We're trying to prevent it. If we had told them you have to unload the product this way and the way we told them to do it actually created the spills, well, then that's a much different story. Well, isn't the whole thing about arranger liability is that it assumes really a lack of control in the statute. The statute is set up so that somebody who arranges for someone else to dispose of hazardous waste is liable apparently without regard to whether they have any idea what the process is. There's no question. The cases talk about, you know, Circle being a strict liability statute. But it's clear that when they drafted this, they had a certain intent. So what does it mean to say that control is all important? In the paradigm arranger, I was having trouble trying to understand this because in the paradigm arranger situation as I understand it, I call up, you know, ex-hazardous waste company. And I say, ex-hazardous waste company, I have some hazardous waste. Come take my hazardous waste and dispose of it. And that's all I do and I'm liable. And I have no idea what they're doing. I have no control over what they're going to do with it. Is that correct? That's the decision that the legislature made. And then what the courts have done is taking that language in light of the other court decisions that say it ought to be applied liberally. They've stretched that. They've stretched it from direct arranger liability to the broader theory of arranger liability. The question is, how far can you stretch this before the statute loses all meaning? And that's where this case is. There is no case that has addressed these facts where someone has arranged to ship a product by common carrier and there are leaks and spills when it's unloaded that has imposed arranger liability. There are two cases that have considered those exact facts. And what did they conclude in those two cases? They concluded there was no basis for arranger liability. Kagan. Which were those, please? One of them is the AMCAST decision. Sorry, AMCAST? AMCAST. Oh, geez. The other one, I believe, is the ESROB. Actually, when you say there were two cases, you know what two cases you're talking about. I wrote them down. Yeah. I'm sorry. I've got the cites right here. ESROB 912F Sup 1476. It's a Northern District, Alabama case, 1995. And AMCAST v. Detrex 3F 3rd 746, 751. That's a Seventh Circuit decision from 1993. Both of those are cited in our papers. Well, this is why the useful product question is a very interesting one. When did this cease to be a useful product? It is. And it goes again. It sort of circles back to the totality of the circumstances. Are we going to look at all the circumstances in light of the intent behind the statute? I always have the same problem with totality of the circumstances test. Totality of the circumstances as to what? We need a standard about what? What are we trying to find out with the totality of the circumstances? Well, and that's where the language of this. You interpret the language of the statute based on what it says. If I call someone and have them haul off my waste, that's a range of liability. Courts have stretched it, which is fine. Formulator cases, we've looked at the pesticide formulator cases. The byproduct cases where you sort of have a disguised sale of a waste or byproduct that has some commercial value. You know, the U.S. v. Shell decision says that we need to look at the facts of our case, compare them with the existing decisions as the starting point of our analysis. You haven't asked my question. I'm sorry. Totality of the circumstances as to what standard? What are we trying to find out the totality of the circumstances about? Totality of the circumstances test is essentially a way to look beyond the label that's put on the transaction to find if a transaction that is described as something other than an arrangement for disposal is, in fact, an arrangement for disposal. So in the formulator cases, they labeled the transactions, formulation of pesticides. The court considered the totality of the circumstances to get beyond that label to find that really there was an arrangement for disposal in there. Same thing that they did with the byproduct cases. They're labeled as a sale of a product when in reality they are, you get behind the label and you find that, oh, this is a waste or byproduct that has some nominal commercial value that they're selling, but it's not really a sale of a useful product. What that really is is a way to dispose of some waste that they need to get rid of. So that's what the totality of the circumstances test is designed to do. You don't need it in the direct arranger theory. You need it in the broader arranger theory. I hope that makes sense. Okay. Thank you very much. Okay. Thank you, Your Honors. Just a couple of points, Your Honors. Under the totality of the circumstances, you look to see if there's been arrangement for disposal of a hazardous substance. Here, Shell did that. Counsel relied on U.S. v. Shell, but in that case, the U.S., as the court pointed out, was the end purchaser. It didn't own any of the products. I think that's where the control question came in. It's a question of whether someone who didn't at the time of arrangement own the material could still follow in the statute. Again, in Shell, the U.S. never owned any of the raw materials or the intervening products. The court found that the U.S. did not contract out a crucial waste-producing intermediate step in the manufacturing process, and the U.S. never exercised any actual control or have direct ability to control the waste. Those facts are not present here, as the district court found. The district court made detailed findings of fact on that question. And a couple of other things, just on the divisibility point. Your Honor hit the nail right on the head with the two-thirds contaminants with respect to the railroads. That comes out of thin air. Counsel talked about DD being volatile. Of course, the district court, when it was doing the divisibility analysis for the Shell portion, found that there were numerous spills from DD rigs which were parked on the railroad parcel. So there's just no basis for arbitrarily lumping off a third of the contaminants here. One other point. But then he essentially added it back in anyway, because he said two-thirds, and then he said, but I'm going to increase it by 9 percent to 9 percent. Well, and again, the 50 percent error rate, I mean, could be 100 percent error rate. But this is the hard question. Is there any doubt that the railroads were responsible for, A, less than 100 percent, and on any fair view of the record, less than 50 percent? Let's just say that much. The record is clear that they are responsible. And under CERCLA, the joint lawyer said that. I understand that. I'm not asking you that. I'm asking you whether any fair view of the record would say that they are responsible for less than 100 percent and less than 50 percent. I don't think so, Your Honor, because there is no evidence saying that the relative activities on the railroad parcel compared to the B&B parcel makes it such that only less than 100 percent of their contaminants made it to the contaminated groundwater. No, but that's not the question. The question is, even if 100 percent of the railroad's contaminants made it to the groundwater, there was still less than 100 percent of the contaminants. Well, it could be all of their contaminants. The record doesn't display that. I mean, there's, because all three chemicals were, the DD, Nemagon, and the Dynaseb, were present on the railroad parcels at various times. And that's another point I wanted to bring up. There's no evidence in the record that says that absent the Dynaseb, the remediation at this site would not be the same. And my point being that to arbitrarily say that you can lump off a third of the contaminants, as the district court did, doesn't mean that that actually is a third of the harm. Even without the DD, there's no evidence in the record that says absent the DD, the remediation effort would have been something else. Or absent the Dynaseb contamination, the remedial effort and the contamination that would have required that remediation effort would have been different. Is there any way the railroad could have proved this? Let's say, aside from the fact that they weren't running the place and were not physically present, suppose they had somebody there every day watching. Could they have ever proved it? Could they? I'm sorry? Could they have ever proved? What could they have done to prove in a satisfactory way what one intuitively thinks, which is that they weren't responsible for too much of it? They could have brought in expert testimony to show. And they tried to do that. And their theory of the case was that none of our contaminants reached our present in the groundwater. And the district court rejected it. They could have taken a different theory of the case and tried to support what the district court did here. They could have said, here's an expert who explains why it's appropriate to say that X amount of activity took place on the railroad parcel and Y percent of the groundwater. They didn't do that. They said the answer was zero. Unless the court has any other questions. I have a question. Yes. Did you folks ever go through the mediation process of our court? We did. You all had hard heads. After this argument, would you like to renew that process? The United States is always willing to engage in mediation. Anything is fair. Is that it? You might wish to consult all of you and advise the court as to whether you want to return to mediation. Thank you, Your Honors. I would also like to say that for the State of California, we are also always interested in participating in mediation and look to try to resolve matters as much as possible through that process as opposed to necessarily this one. You just can't tell what these judges might do. That's right. Your Honors, I just wanted to make a couple of brief comments with regard to the divisibility issues that have been raised by Schell and the railroads. First of all, I think the railroads suggested that with regards to divisibility, it's enough for them to show that they might be liable for more than their share. I mean, that cannot be the standard for finding divisibility under CERCLA. Finding divisibility under CERCLA under a – with respect to a government's 107 cost recovery claims is the exception, not the rule. And what the railroads are suggesting would basically be, well, you know, it's more than what we should otherwise pay. You should find divisibility. What's unusual about this case is that we have three differently situated PRPs. We have a landowner, an arranger, and an operator. And the operator's out because he's defunct, who's the person who's most obviously liable. So in that circumstance, what case law is there? There's the Third Circuit case about landowner liability. And is that it in terms of the divisibility and landowner liability? Well, I think in terms of what the landowner would have to do in the situation such as the railroads would have to show that there is, in fact, some basis based upon causation to show that there's a volumetric contribution coming from one parcel versus a volumetric contribution coming from the other. Well, to begin with, it can't be a bad causation because the whole statute is a non-causation statute with regard to the landowners, right? They don't have to have caused anything to be liable. So it doesn't make any sense to start asking them to prove causation under divisibility. Well, I think it does. And what it does is if you're going to establish liability under Section 107, that's one thing. But then you're going to say, oh, by the way, I don't – I'm not going to be jointly and severally liable for the contamination for the entire site. I have to provide some reasonable basis in which to demonstrate that there can be some kind of division of that liability. And I believe what the Court has suggested is the way to do it is a causation-based causation. Causation in this sense meaning not that the railroad caused anything, but that the – its land, there was something – there was some connection between its land and certain contaminants. Correct. Not really causation, but okay. Go ahead. And I just wanted to mention, with regard to that, what the trial court said was that there is no evidence to quantify the differences in the volume of releases from the railroad parcel as compared to the Brown and Bryant parcel. That fact doesn't exist. So therefore, the railroads have not been able to demonstrate divisibility based upon the applicable case law. I think that if you look at the Kim Nuclear v. Bush case that we cited in our papers, that is instructive on the issue as to whether or not you can substitute theories, estimates, assumptions for actual foundational facts. And that's something the railroads didn't do. Likewise for Shell, the trial court said Shell did not present evidence on how its products' contribution to the contamination at the Arvin facility can be apportioned. The same kind of situation applies to Shell. Thank you very much. Thank you very much. Thank you, counsel. I'm sorry. You seem to be standing up. Oh, I see. All right. Thank you. I thought that somebody else thought he was arguing. But as long as he doesn't. You used all your time by quite a lot. By about ten minutes extra. Thank you very much. Thank you. This court, this session stands adjourned.