Appeals from the United States District Court for the Northern District of Illinois, Eastern Division. No. 94 C 5825 Suzanne B. Conlon, Judge.
Before POSNER, Chief Judge, and COFFEY and MANION, Circuit Judges.
Joseph Rizzo and his wife Margaret brought a products liability suit in federal district court under the diversity jurisdiction against the manufacturer of the "Mr. Coffee" coffeemaker and the manufacturer of the glass portion of the coffeemaker's carafe. A jury awarded damages of $182,000 to the plaintiffs (a part of the award was to Mrs. Rizzo, for loss of consortium), and the defendants appeal. The accident giving rise to the suit came about as follows. About five months after the purchase of the coffeemaker, while the Rizzos were having breakfast in their kitchen, Mr. Rizzo reached behind him for the carafe full of hot coffee to pour his wife a cup. He lifted the carafe from its perch on the warmer plate in the coffeemaker and at the exact moment when the carafe in its journey toward his wife's cup was directly over his lap the carafe broke in two, spilling the coffee (which was heated to between 170 and 180 degrees F) into Mr. Rizzo's lap and causing second-degree burns.
The plaintiffs tried to show that the carafe was defective. The defendants presented contrary evidence through their expert, a materials engineer employed by one of the defendants. They argued, and their expert testified, that Rizzo must have struck the wall with the carafe, weakening the carafe, which broke a moment later as a result. The Rizzos denied that the carafe had struck the wall and we do not understand the defendants to be arguing that their expert's evidence was conclusive on the issue, requiring the jury to find that the Rizzos were lying. The expert also testified that the Rizzos might have weakened the carafe by using abrasive cleansers to clean it; the Rizzos denied having used abrasive cleansers.
The plaintiffs tried to prove their case by two alternative routes. The first, and thoroughly conventional, was to show that the carafe had in fact a specific, demonstrable manufacturing defect. There was a small bump in the glass wall of the carafe where it had broken. The bump was a defect, and the plaintiffs tried to show that it had been the cause of the break. Alternatively, they relied on a form or variant of res ipsa loquitur that the Illinois courts recognize in products liability cases. Under this second approach, if the plaintiff can show that the product failed at a time or in a manner it would not be expected to fail merely through normal wear and tear, and can also show that the plaintiff did not misuse the product and that no "third force" (for example, a fire, an earthquake, or the mishandling of the product by a carrier or dealer, or by a house guest) -- what the Illinois courts mysteriously refer to as a "secondary cause" -- was present, the jury is allowed to infer, despite the absence of evidence of a specific defect, that the accident was indeed the result of a product defect for which the manufacturer is liable. See, e.g., Tweedy v. Wright Ford Sales, Inc., 357 N.E.2d 449 (Ill. 1976); Welge v. Planters Lifesavers Co., 17 F.3d 209, 211, 213 (7th Cir. 1994).
To establish liability under their first approach, the plaintiffs hired a 28-year-old engineer named Moore, a consultant. Moore has a master's degree in materials engineering and some experience as an analyst of material failures, including failures of fragile materials, and has taken a couple of courses on ceramics failures (glass is a ceramic). But at the time of trial his experience in analyzing glass failures was limited to one glass bottle; and he had not conducted scientifically accepted tests to determine why the Rizzos' carafe had broken. The district judge, applying the higher standard for the admissibility of expertwitness testimony that has emerged in the wake of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) --see, e.g., Braun v. Lorillard Inc., 84 F.3d 230, 234-35 (7th Cir. 1996); Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 318 (7th Cir. 1996); Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1316-18 (9th Cir. 1995); cf. Mid-State Fertilizer Co. v. Exchange National Bank, 877 F.2d 1333, 1339 (7th Cir. 1989) -- ruled before the trial that Moore was not qualified to testify about the cause of the break but that he could testify to his observations, made under a microscope, about the origin of the break -- that is, where on the carafe the fatal crack had begun. The plaintiffs' lawyer was not happy with this truncation of their only expert's evidence, and in examining Moore on direct tossed in some questions about the cause of the break, but the defendants objected and the objections were sustained. By the time Moore's direct examination ended, it should have been clear that the plaintiffs' first theory was doomed. In cross-examining him, however, the defendants' lawyers repeatedly asked him about the cause of the break and finally got him to say that the bump alone could not have caused it. On redirect, the plaintiffs' lawyer was permitted to inquire further into the question of causation and he got Moore to retract his concession that the bump could not have been the cause. Nevertheless the district judge granted a directed verdict for the defendants on the plaintiffs' first theory of liability -- proof of a specific defect that caused the accident, namely the bump. So the case went to the jury only under the second theory, the products liability variant of res ipsa loquitur, what the parties call the Tweedy theory after the leading Illinois case expounding it.
The defendants argue that Moore should not have been permitted to testify about anything. They say that if he wasn't qualified to testify about the cause of the break he wasn't qualified to give any expert testimony relevant to the case. Merely testifying to what he saw through a microscope would not even be expert testimony -- anyone can look through a microscope and describe what he sees -- and in any event would cast no light on any issue in the case. Admission of this evidence, the defendants argue, was bound to confuse the jury, making it think that Moore was testifying about causality, and forcing the defendants, therefore, to explore the issue of causality with him on cross-examination.
Why any of this should matter to this appeal may seem puzzling in light of the directed verdict for the defendants on the issue of proof of a specific defect. The directed verdict may seem to have wiped out everything Moore said. Not quite. The judge, while willing to give an instruction (which the defendants however failed to tender) to the jury that the bump had not been shown to be the cause of the accident, turned down the defendants' request that the jury be told not to consider any evidence that "pertain[s] to plaintiffs' claim that any bump on the carafe caused the carafe to fracture." This instruction might have been interpreted to mean (indeed the defendants themselves seem to have interpreted it to mean) that the jury could not consider evidence about the bump for any purpose. The refusal to give an instruction that could be understood in this way was proper, assuming Moore's testimony was admissible, because that testimony was relevant to the Tweedy approach. The testimony had two parts -- the testimony about the origin of the break, which Moore gave on direct, and the testimony about the cause of the break, which Moore gave on both cross-examination and redirect examination. The first part undermined the testimony by the defendants' expert that the break had originated in a crack at the bottom of the carafe, a crack that he attributed to Rizzo's having banged the carafe against the wall and that the plaintiffs attributed to the bottom having fallen onto the floor from Rizzo's lap. The second part of Moore's testimony, the part about the bump's having caused the break, corroborated the Tweedy approach by making the accident less mysterious. Even if the bump could not be proved to be the cause of the break, the fact that there was this defect which might have been the cause bolstered the inference drawn from the Tweedy factors -- the accident itself, and the absence of proof of a cause other than some unknown defect in the manufacturing process. Moore's testimony would have made the jury more comfortable about drawing the Tweedy inference.
So the jury may well have been influenced by Moore's testimony and we must therefore consider whether it was admissible. We think it was, or more precisely -- bearing in mind that appellate review of evidentiary rulings is deferential -- that the district judge was not unreasonable in deciding to admit it. The issue has two aspects --whether Moore should have been permitted to testify at all, and if so whether he should have been permitted to testify about the cause of the accident despite the judge's initial ruling on the permissible scope of his testimony. The second question is actually easier than the first, and the answer to it also answers the first. When the defendants' lawyers began questioning Moore about the cause of the accident, they waived any objection to the opposing lawyer's going into the issue on redirect examination in order to correct any misleading impressions that the cross-examination might create. The defendants' lawyers could not be allowed to ask Moore what the cause was and then, satisfied with his answer, prevent their opponents from exploring the issue further with Moore to make sure what he meant. Braun v. Lorillard Inc., supra, 84 F.3d at 238; United States v. Martinez, 988 F.2d 685, 702 (7th Cir. 1993); United States v. Touloumis, 771 F.2d 235, 241 (7th Cir. 1985). The defendants' lawyers say they had no choice but to ask Moore about cause, that the cat had been let out of the bag on direct. Not so. The judge insisted that Moore stay in the narrow groove she had planed for him. When the plaintiffs' lawyer tried to bring him out, she cut him off at the knees, as the saying goes. The defendants' lawyers, perhaps mindful that their own expert, although a distinguished fractologist, might lack a certain credibility, being an employee of one of the defendants and having testified for his employer in other products liability suits, gambled on extracting helpful evidence on causation from the plaintiffs' rather inexperienced expert. They gambled, and lost, and that is the end of it.
That still leaves the question whether Moore should have been permitted to testify about the origin of the crack. The close relation between origins and causes created a potential for confusing the jury. But one of the defendants' own lawyers, in cross-examining Moore, brought out the difference between origin and cause very clearly. The carafe fell in two, but it did not do so instantaneously. The crack that in a fraction of a second shot around the circumference of the carafe, precipitating the break and the resulting fall of the bottom of the carafe into Rizzo's lap, originated at some point on the surface of the carafe, whether that point was the bump, or the point at which the carafe banged the wall (if the defendants' theory of the accident is correct), or somewhere else in the carafe's surface. Why the crack originated at whatever point it did originate -- whether because the bump had weakened the glass, or a blow weakened the glass at the point of origin of the crack -- was a distinct question, the question of the cause of the accident. The judge may have been on thin ice (or glass) in supposing Moore qualified to testify about the origin but not about the cause. But we do not think that it can be considered an unreasonable distinction even if it were not (as we believe it is) rendered moot by the defendants' decision to go into the issue of cause with Moore on cross-examination. Moore is a materials engineer. He had studied and conducted materials failure analyses, of which identifying the origin of the failure might be considered a first step considerably less difficult than the ascription of causality. It is true that in her initial ruling the judge had seemed to confine the scope of Moore's testimony to his mere "observations" through the microscope and that later, when she directed a verdict for the defendants on the claim of a specific defect, she had disparaged Moore's qualifications. But the line between observation and inference, especially for a scientific expert, is vague. What Moore saw through the microscope was more than what a lay person would have seen, and it was implicit in the judge's allowing him to testify at all and refusing to strike his testimony that she thought he could illuminate relevant matters not obvious to the jurors. About the only such matter was the origin of the break. In testifying about that, he was offering an "observation" fairly described and authorized as an expert opinion on a relevant issue.
There is language in the trial transcript to suggest that the judge thought she was allowing Moore to testify as something other than an expert. But only an expert could testify as to the origin of the break in the carafe, and when Corning's lawyer reminded the judge that she "had found Mr. Moore qualified in some areas," the judge acquiesced, adding, "I explicitly excluded causation." We think the best interpretation is that the judge found Moore qualified to testify about the origin of the crack. This finding is not an abuse of discretion and therefore binds us.
The defendants raise other issues, but only one has sufficient merit to warrant discussion, and that is whether the case fits the Tweedy doctrine. They say that the doctrine was never intended for so fragile a product as a glass carafe, the breaking of which is likely to occur without any defect. What is true but off the point is that most carafes break not because they are defectively designed or manufactured but because, being indeed fragile, they are dropped or banged against something in shipment or use or are otherwise mishandled. When that happens, however, it happens because someone has dropped or banged or otherwise abused the carafe, which the jury was entitled to find had not occurred here, though there was some evidence that it had. The very fragility of the glass, so emphasized by the defendants, works against their argument by underscoring the unusual character of the product failure in this case. A carafe designed to be used for years, not months, breaks in half without being dropped or banged or cleaned with abrasive cleansers or damaged in a flood or fire. In these unusual circumstances the accident itself is sufficient evidence of a defect to permit though of course not compel the jury to infer a defect. Whether these were the circumstances of the accident was a jury question. The jury was not, ...