Two-year-old Tiffany Todd died tragically when four-year-old Cori Smith used a Bic lighter to start a fire in Tiffany's bedroom. Tiffany's estate sued the lighter manufacturer and the district court granted summary judgment in favor of the defendants. After rehearing the appeal en banc,1 a majority of this court concluded that the warning printed on the lighter--"KEEP OUT OF REACH OF CHILDREN"--was adequate, and that summary judgment on the issue of duty to warn was proper. However, to ensure that the law we applied on the other issues in the case was genuine state law, we certified to the Illinois Supreme Court questions which examined the proper application of the consumer contemplation test and the risk-utility test under Illinois products liability law. The Illinois Supreme Court promptly declined to answer the certified questions. These issues thus return to this court sitting en banc.

Generally, this case concerns Illinois strict products liability law. In Illinois and elsewhere, strict products liability law evolved as a "special rule applicable to sellers of products",2 which went beyond the traditional boundaries of warranty and negligence law, to provide a broader method of recovery for persons injured by defective products. See William L. Prosser, Law of Torts, 641-682 (4th ed. 1971). Although the scope of strict products liability law is broader than negligence or warranty law, it is not unlimited. Only those who sell "unreasonably dangerous" products fall within its reach. Restatement (Second) of Torts Sec. 402A.

In this case we consider whether a manufacturer should be subject to liability for producing a disposable lighter which a child used to start a deadly fire. This inquiry pivots on one basic question: whether a disposable lighter is unreasonably dangerous under Illinois strict products liability law. The district court determined that a lighter is not unreasonably dangerous, and granted summary judgment for the manufacturer. The plaintiff appeals. We now affirm the district court's grant of summary judgment.

Two young families, the Smiths and the Todds, shared a house in rural Earlville, Illinois. All four adults in the house smoked, and all used disposable cigarette lighters. About a week before the fire which caused the death in this case, four-year-old Cori Smith got hold of one of these lighters and set a small fire in her parents' bedroom. The fire was quickly extinguished, and Cori's parents admonished her never to play with lighters or matches. The adults also warned the other five children who lived in the house about the dangers of fire.

The next Sunday, March 27, 1988, Cori awoke before her parents and went downstairs, where her brother was watching cartoons. She spied a green lighter on an end-table in the living room. She took the lighter upstairs into a bedroom where twenty-two month old Tiffany Todd was sleeping. Cori used the lighter to ignite some papers which were on the floor. She then took the lighter back downstairs and replaced it on the end-table. The adults did not wake in time to prevent the ensuing conflagration. Tiffany Todd was killed in the fire. Everyone else escaped unharmed.

On behalf of her estate, Tiffany's father, Rodney Todd, sued the lighter's manufacturer, Bic Corporation, claiming that the company was negligent and strictly liable for selling a defective product. Todd rested both his negligence and strict liability claims primarily on allegations that the lighter was unreasonably dangerous because it did not include a child-resistant feature, and did not provide an adequate warning. After some discovery, Bic moved for summary judgment. Bic never denied that it was possible to manufacture a lighter with child-resistant features. In fact, undisputed evidence adduced in discovery showed that at the time of the fire Bic had developed a prototype lighter which was child-resistant--because it required greater dexterity to operate--although not child proof. Bic contended, however, that the lighter Cori used was not unreasonably dangerous to consumers. Bic also vigorously defended the adequacy of the warning it placed on the lighter. The warning on the lighter which Cori found on the end-table while her parents slept, read simply, "KEEP OUT OF REACH OF CHILDREN."In a comprehensive order, the district court granted summary judgment for Bic. The court determined that an ordinary lighter, which did nothing more than provide the small flame it was intended to provide, was not defective; therefore, Bic was neither strictly liable nor negligent. The court also considered the lighter's warning and concluded that it was adequate. Finally, the court addressed the various public policy arguments Todd raised in favor of imposing liability on Bic. The court dismissed these arguments out-of-hand, noting that "public policy only requires holding manufacturers and sellers liable if their product is found to be defective or unreasonably dangerous...." Todd v. Societe Bic, No. 90 C 5487 at page 11, 1992 WL 4971 (N.D.Ill. January 7, 1992).

Todd appealed, claiming that the district court ignored questions of fact about the inherent defects in disposable lighters. He maintained that Illinois authorized two tests to determine whether a product is unreasonably dangerous: the consumer contemplation test and the risk-utility test. He argued that a disposable lighter might be considered unreasonably dangerous under either test. He also insisted that the district court's failure to even consider aspects of the risk-utility test required reversal of the summary judgment. Finally, he continued to challenge the adequacy of the warning, claiming that deficiencies in the warning, standing alone, rendered the lighter unreasonably dangerous.

A divided panel of this court agreed with Todd and reversed the district court's grant of summary judgment. 991 F.2d 1334 (1993). The full court vacated that holding, and decided to rehear the case en banc. 991 F.2d 1344 (1993). "Concern about the implications of the panel's reasoning lead the full court to vacate the panel's decision and set the case for rehearing en banc." Todd, 9 F.3d at 1218. After the rehearing, the full court was divided on what the proper disposition of this case should be. The court unanimously agreed that the district court was correct in its conclusion that the warning on the lighter was adequate as a matter of law. However, a majority of the court decided to certify certain questions about strict products liability law to the Illinois Supreme Court. Todd, 9 F.3d at 1216.

The Illinois Supreme Court declined certification. Todd, No. 7643 (November 22, 1993). That decision placed this case back before the en banc court. We will resolve the issues of state law which this case presents in accordance with the decisions of the Illinois Supreme Court. See Charles A. Wright, Arthur R. Miller and Edward H. Cooper, Federal Practice and Procedure Sec. 4507 at 89 (1982) ("federal court must determine issues of state law as it believes the highest court of the state would determine them"). Where areas of state law are not developed, we will resort to other persuasive authority in an attempt to determine what the Illinois Supreme Court would decide. See Heller Intern. Corp. v. Sharp, 974 F.2d 850, 858 (7th Cir.1992).

Strict products liability grew out of a jurisprudential movement to adjust traditional tort and warranty theories in order to facilitate recovery for consumers injured by defective products. See William L. Prosser, Law of Torts, 641-682 (4th ed. 1971). This movement spawned the notion that a manufacturer should be liable to an injured consumer regardless of the degree of care employed in producing the product. Id. at 656-657. The concept of strict products liability gained widespread acceptance in the early 1960's after the American Law Institute reduced the broad hypothesis to more definite standards when drafting the Second Restatement of Torts, Section 402A. Illinois soon joined other jurisdiction in adopting Section 402A. Suvada v. White Motor Co., 32 Ill. 2d 612, 210 N.E.2d 182 (1965).

Negligence and warranty theories are still available for plaintiffs injured by defective products. See Section 402A, cmts. a and m. In fact, Todd has made a negligence claim in this case, which we will address later. But injured plaintiffs seem to favor strict products liability, primarily because it does not include traditional warranty law's damage and privity limitations, or traditional negligence law's fault element. See Section 402A, cmts. a and m; see also Suvada, 210 N.E.2d 182 (providing history of strict products liability in relation to traditional contract and tort law). While broader than historic negligence and warranty law, strict products liability is not all-encompassing; it does not cover every injury involving a product. Some aggressive litigants, nevertheless, consider Section 402A as a mandate to compensate every consumer injured by a manufactured product. The Illinois Supreme Court has specifically criticized this approach on a number of occasions. See, e.g., Woodill v. Parke Davis & Co., 79 Ill. 2d 26, 37 Ill. Dec. 304, 309, 402 N.E.2d 194, 199 (1980) ("[s]trict liability is not the equivalent of absolute liability"); Coney v. J.L.G. Indus., Inc., 97 Ill. 2d 104, 73 Ill. Dec. 337, 340, 454 N.E.2d 197, 200 (1983) ("imposition of strict liability was not meant to make the manufacturer an absolute insurer"); Smith v. Eli Lilly & Co., 137 Ill. 2d 222, 148 Ill. Dec. 22, 42, 560 N.E.2d 324, 344 (1990) (unwarranted expansion of strict liability law is not appropriate "to achieve what is perceived as socially satisfying result.").

Therefore, although broad, strict products liability law is limited by certain formulas and boundaries. Section 402A generally sets forth these limitations:

Sec. 402A. Special Liability of Seller of Product for Physical Harm to User or Consumer

(1) One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property, if

(a) the seller is engaged in the business of selling such a product, and

(b) it is expected to and does reach the user or consumer without substantial change in the condition in which it is sold.

(2) The rule stated in Subsection (1) applies although

(a) the seller has exercised all possible care in the preparation and sale of his product, and

(b) the user or consumer has not bought the product from or entered into any contractual relation with the seller.

The comments to Section 402A also define its limits. Basically, in order to recover under that section, "[t]he plaintiffs must prove that their injury or damage resulted from the condition of the product, that the condition was an unreasonably dangerous one and that the condition existed at the time it left the manufacturer's control." Suvada, 210 N.E.2d at 188.

In this case, the district court applied Section 402A to the undisputed facts, and determined that a cigarette lighter is not unreasonably dangerous. Whether a product may be considered unreasonably dangerous is a question of Illinois law which we review de novo. Derrico v. Bungee Intern. Mfg. Co., 989 F.2d 247, 249 (7th Cir.1993); Belline v. K-Mart Corp., 940 F.2d 184, 186 (7th Cir.1991).

1. Consumer Contemplation Test.

Section 402A authorizes a simple test to determine whether a product is unreasonably dangerous: "[t]he article sold must be dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics." Section 402A, cmt. i. That test has come to be known as the consumer contemplation test. W. Page Keeton, Prosser and Keeton on Torts, 698-99 (5th ed. 1984). The Illinois Supreme Court follows the Restatement's formulation of the test almost verbatim: "[a] product is 'unreasonably dangerous' when it is 'dangerous to extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics.' " Lamkin v. Towner, 138 Ill. 2d 510, 150 Ill. Dec. 562, 570, 563 N.E.2d 449, 457 (1990).

The consumer contemplation test separates defective products from the universe of ordinary products which may be involved in causing injury. Under this test, a product is only considered defective or unreasonably dangerous if it fails to perform in a manner the ordinary consumer would expect. Therefore, a pair of rubber soled shoes are not unreasonably dangerous simply because the person wearing them slips on a wet floor, because "[i]t is a matter of common knowledge that shoes are more likely to slip when wet than dry...." Fanning v. LeMay, 38 Ill. 2d 209, 230 N.E.2d 182, 184 (1967). But if a hammer breaks when striking a steel peg, sending a metal fragment into the user's eye, the hammer is unreasonably dangerous because ordinary consumers do not expect hammers to chip on impact. Dunham v. Vaughan & Bushnell Mfg. Co., 42 Ill. 2d 339, 247 N.E.2d 401 (1969). Law reporters are full of cases illustrating the important distinction between those products that meet consumer expectations and those that do not. The Restatement's drafters provided a simple exercise to clarify this distinction:

Good whiskey is not unreasonably dangerous merely because it will make some people drunk, and is especially dangerous to alcoholics; but bad whisky, containing a dangerous amount of fuel oil, is unreasonably dangerous. Good tobacco is not unreasonably dangerous merely because the effects of smoking may be harmful; but tobacco containing something like marijuana may be unreasonably dangerous. Good butter is not unreasonably dangerous merely because, if such be the case, it deposits cholesterol in the arteries and leads to heart attacks; but bad butter, contaminated with poisonous fish oil, is unreasonably dangerous.

Section 402A, cmt. i.

The consumer contemplation test is an important aspect of Section 402A, because it prevents imposing absolute liability on manufacturers. Without this test, injured plaintiffs would be heard to argue that the fact of their injury establishes a product's dangerous propensities. The test heads off this type of after-the-fact rationale by recognizing that virtually any product can cause injury when put to certain uses: "ordinary sugar is deadly poison to diabetics, and caster oil found use under Mussolini as an instrument of torture." Section 402 A, cmt. i. The consumer contemplation test assures that products are not considered unreasonably dangerous simply because they have conceivable unsafe uses. The test focuses not in hindsight on the injury which occurred, but on the ordinary consumer's expectations when purchasing and using the product.

This discussion brings us to the seminal question in this case: what does an ordinary consumer expect when he purchases a lighter? Quite obviously, the consumer expects the lighter, when activated, to provide a flame. That is exactly how the lighter in this case performed. But the lighter was used to do more; the small flame it produced was used to ignite some papers, causing a deadly fire. The ordinary consumer expects that if a lighter's flame is put to some other combustible object, a larger fire ensues. This makes a lighter dangerous. But it does not make a lighter unreasonably dangerous under the consumer contemplation test. As the Illinois Supreme Court has stated, "[v]irtually any product is capable of producing injury when put to certain uses or misuses.... Injuries are not compensable in products liability if they derive merely from those inherent properties of a product which are obvious to all who come in contact with the product. The injuries must derive from a distinct defect in the product, a defect which subjects those exposed to the product to an unreasonable risk of harm." Hunt v. Blasius, 74 Ill. 2d 203, 23 Ill. Dec. 574, 578, 384 N.E.2d 368, 372 (1978). The fact that a lighter starts a fire is one of its inherent properties. That fact does not breach consumer expectations; really it fulfills them. Therefore, a lighter which does nothing more than provide a small flame is not unreasonably dangerous under the consumer contemplation test.

a. Foreseeable user versus ordinary consumer.

The plaintiff argues that instead of looking to the expectations of the ordinary consumer, we should look to those of the foreseeable user--in this case a child--when assessing liability under the consumer contemplation test. This expansive view is required, the plaintiff contends, because a child ultimately used a lighter in this case.3 Some of the leading academic commentators in the area of strict products liability support a reformulation of the consumer contemplation test which measures the expectations of foreseeable users, rather than ordinary consumers. See W. Page Keeton, Prosser and Keeton on Torts, 698-99 (5th ed. 1984). But we need not concern ourselves with academic speculation when Illinois law is clear. Both the Restatement and the Illinois cases which have adopted it, speak in terms of the "ordinary consumer ... with the ordinary knowledge common to the community as to [the product's] characteristics." See Lamkin, 150 Ill. Dec. at 570, 563 N.E.2d at 457. We therefore look to the expectations of the ordinary consumer, and not the foreseeable user, when assessing liability under the consumer contemplation test.4

Reason dictates that children are not ordinary consumers under the consumer contemplation test. Unlike ordinary consumers, children lack knowledge common to the community regarding consumer products. This lack of knowledge makes children particularly unfit subjects for any test meant to measure expectations. Consider the absurd results if courts were required to look to the expectations of children when gauging a product's dangerousness. The pair of shoes in Fanning, which the Illinois Supreme Court considered not defective because "[i]t is a matter of common knowledge that shoes are more likely to slip when wet than dry", would be considered defective under any formulation of the consumer contemplation test which measured a child's expectations. Fanning, 230 N.E.2d at 184. Children do not share the community's knowledge about a shoe's propensity to slip when wet.

Nor do children perceive the dangers inherent in just about any product. We all had the experience as children of putting a hand on a hot pot or stove. Recall the story about the unfortunate child who expected his Superman cape to allow him to fly. Children engage in these dangerous activities because, in their limited experience, they cannot comprehend the possibility of injury. Any reformulation of the consumer contemplation test which requires a court or jury to assess liability based on a child's product expectations, removes well-designed limitations from Section 402A in all cases where a child is injured. Under such a test, manufacturers would face absolute liability every time a child was injured by a product. The child's attorney would march into court and claim the product failed to perform in the manner the child expected. The attorney would be right, but the failure would be due not to a deficiency in the product, but to the natural deficiency children have in their knowledge of consumer products.

In any event, no Illinois court authorizes an expansion of the consumer contemplation test to the realm of foreseeable users. Rather, Illinois courts, as well as the Restatement, confine the test to the ordinary consumer. It works an absurdity to abandon the ordinary consumer standard for the foreseeable user standard, especially where children are concerned. Children lack knowledge about consumer products. Therefore, children should not be the standard to measure consumer expectations. In the case presently before the court, the ordinary consumer--if not a four-year-old child--is deemed to expect that lighters cause fires. Therefore, lighters are not unreasonably dangerous.5

2. Risk-Utility Test.

Under the Restatement, the consumer contemplation test was the only standard authorized to determine whether a product was unreasonably dangerous. But tort commentators soon began to express dissatisfaction with the test's restrictions on manufacturer liability. These criticisms evolved into case law. See Cronin v. J.B.E. Olson Corp., 8 Cal. 3d 121, 104 Cal. Rptr. 433, 501 P.2d 1153 (1972), and Barker v. Lull Engineering Co., Inc., 20 Cal. 3d 413, 143 Cal. Rptr. 225, 573 P.2d 443 (1978), and articles cited therein. The California Supreme Court, which earlier had pioneered the movement away from traditional liability based on contract and negligence, to a theory of strict products liability, see Greenman v. Yuba Power Products, Inc., 59 Cal. 2d 57, 27 Cal. Rptr. 697, 377 P.2d 897 (1963), generated a new test--the risk-utility test--to be used in defective design cases. This innovation began in Cronin and culminated in Barker, where the court squarely confronted what it perceived as the limitations of the consumer contemplation test.

Barker involved a construction worker who was seriously injured when the lift truck he was operating became unbalanced on a steep grade, and dropped the lumber it was carrying. The worker jumped out of the cab of the truck when it started to tip. Unfortunately, he jumped directly into the path of the falling lumber. He sued the truck manufacturer claiming, among other things, that the truck should have been equipped with "outriggers"--large mechanical arms extending from the sides of the machine--to provide stability on steep grades. The manufacturer responded that the truck was not meant for steep terrain--that cranes and not lift trucks were designed to be used on steep terrain.

At the trial, the court instructed the jury under the consumer contemplation test, and the jury returned a verdict for the manufacturer. Essentially, the jury found that the ordinary consumer would expect the lift truck to tip on steep terrain. The California Supreme Court reversed, on the ground that the consumer contemplation test represented "undue restriction on the application of strict liability principles." Barker, 143 Cal. Rptr. at 233, 573 P.2d at 451. The court explained that "[i]n Cronin ... we flatly rejected the suggestion that recovery in a products liability action should be permitted only if a product is more dangerous than contemplated by the average consumer, refusing to permit the low esteem in which the public might hold a dangerous product to diminish the manufacturer's liability for injuries caused by the product." Id. To deal with this perceived inadequacy, the court fashioned a second test to gauge a manufacturer's liability for designing a defective product. The court specifically held that "a product may be found defective in design, even if it satisfies ordinary consumer expectations, if through hindsight the jury determines that the product's design embodies 'excessive preventable danger,' or, in other words, if the jury finds that the risk of danger inherent in the challenged design outweighs the benefit of such design." Id. at 236, 573 P.2d at 454. By introducing this element of hindsight into the strict liability equation, the so-called risk-utility test fundamentally changed that equation. Under the new test, juries would no longer be asked merely to view the foresight of the ordinary consumer. Instead, juries would be asked, in hindsight, to weigh the risk inherent in the product's design.6

The California Supreme Court decided Barker on January 16, 1978. The case did not gain immediate acceptance in other jurisdictions. It represented a significant expansion upon the Restatement approach, which itself was a significant expansion upon traditional contract and tort law.7 Within a year after Barker was decided, the Illinois Supreme Court continued to cling exclusively to the Restatement approach, declaring in a decision on December 4, 1978 that "[a]n action is not maintainable in products liability merely because the design used was not the safest possible." Hunt, 23 Ill. Dec. at 578, 384 N.E.2d at 372. But within two months the court seemed to change course. In Anderson v. Hyster Co., 74 Ill. 2d 364, 24 Ill. Dec. 549, 551, 385 N.E.2d 690, 692 (1979), the court declared, without citation, that a design defect may be proved "by evidence of the availability and feasibility of alternative designs at the time of its manufacture, or that the design used did not conform with the design standard of the industry, design guidelines provided by authoritative voluntary association or design criteria set by legislation or governmental regulations." Later Illinois cases which have applied variations of the risk-utility test have cited Anderson and Barker as authority for the test. See Lamkin, 150 Ill. Dec. at 570, 563 N.E.2d at 457; Palmer v. Avco Distributing Corp., 82 Ill. 2d 211, 45 Ill. Dec. 377, 381, 412 N.E.2d 959, 963 (1980); Kerns v. Engelke, 76 Ill. 2d 154, 28 Ill. Dec. 500, 505, 390 N.E.2d 859, 864 (1979).

Thus did the risk-utility test gain root in Illinois strict products liability law. How deeply it is rooted remains another question. True, the Illinois Supreme Court has sanctioned variations of the test, in certain cases, to determine whether a product is unreasonably dangerous. See, e.g., Lamkin, 150 Ill. Dec. at 570, 563 N.E.2d at 457; Palmer, 45 Ill. Dec. at 382, 412 N.E.2d at 964; Kerns, 28 Ill. Dec. at 506, 390 N.E.2d at 865; and Anderson, 24 Ill. Dec. at 551, 385 N.E.2d at 692. But in other cases, the court has summarily determined that certain products simply are not unreasonably dangerous, without reference to the test. See West v. Deere & Co., 145 Ill. 2d 177, 164 Ill. Dec. 122, 123-24, 582 N.E.2d 685, 686-87 (1991); Dubin v. Michael Reese Hospital, 83 Ill. 2d 277, 47 Ill. Dec. 345, 347, 415 N.E.2d 350, 352 (1980); Hunt, 23 Ill. Dec. at 578, 384 N.E.2d at 372. The idea that the test is a sweeping pronouncement of strict products liability law, meant to apply to all products, comes from some broad language in Lamkin, where the court stated: "[a] plaintiff may demonstrate that a product is defective in design, so as to subject a retailer and a manufacturer to strict liability for resulting injuries, in one of two ways: (1) by introducing evidence that the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner or (2) by introducing evidence that the product's design proximately caused his injury and the defendant fails to prove that on balance the benefits of the challenged design outweigh the risk of danger inherent in such designs." Lamkin, 150 Ill. Dec. at 570, 563 N.E.2d at 457.

Does this pronouncement require that we mechanically apply the risk-utility test to all design defect claims, or may we--as the Illinois Supreme Court has done on occasion--summarily determine that a product is not defective? In Scoby v. Vulcan-Hart Corp., 211 Ill.App.3d 106, 155 Ill. Dec. 536, 569 N.E.2d 1147 (1991), an Illinois Appellate Court grappled with this question in a case similar to this one, and its decision is instructive. There, the plaintiff sued a deep fat fryer manufacturer for burns he sustained when he slipped at work and his arm became submerged in the hot oil contained in the fryer. He argued that the fryer should have had a simple cover, which the parties agreed would have prevented the injury. The trial court granted summary judgment for the fryer manufacturer and the appellate court affirmed, determining that the fryer without a cover was not unreasonably dangerous. The appellate court considered whether Lamkin's broad authorization of the risk-utility test (the appellate court called this the danger-utility test), meant that the test should be applied in all cases. The court determined that in certain cases, the risk-utility test should not be applied, stating:

We do not deem Lamkin or other cases applying aspects of the danger-utility test intend that all manufacturers of products ... should be subject to liability depending on the trier of fact's balancing under that test, when suit is brought by one injured by such a product. Somewhere, a line must be drawn beyond which the danger-utility test cannot be applied. Considering not only the obvious nature of the danger here but, also, the simple nature of the mechanism involved, we conclude that the circuit court properly applied only the consumer-user contemplation test. Under that test, summary judgment for the defendant was clearly proper.

Scoby, 155 Ill. Dec. at 540, 569 N.E.2d at 1151.

The court in Scoby alluded to the absurdity of applying the risk-utility test to simple but obviously dangerous products. Consider, as the defendant in Scoby proposed, the ordinary kitchen knife. Suppose someone cut by such a knife sued its manufacturer for failing to design a permanent retractable sheath for the knife. Should that person be able to avoid summary judgment simply by presenting statistics showing the feasibility of a permanent retractable sheath? The logical answer is no. As the court in Scoby recognized "many manufactured products have potential to cause substantial injury and most can be made safer by different design." Scoby, 155 Ill. Dec. at 540, 569 N.E.2d at 1151. For certain products, at least, the manufacturer should not be subject to liability just because it failed to choose the safest (and probably more expensive) alternative design.

So how then do we read the broad statement in Lamkin ? Really, the statement says nothing more than Illinois authorizes the risk-utility test in products liability cases. The statement does not address whether this authorization extends to simple but obviously dangerous products. To the extent that one might read the statement as covering simple but obviously dangerous products, the statement is mere dictum. Indeed, Lamkin had nothing to do with such products; therefore, whether or not the risk-utility test is appropriate for such products was not a question presented to the court. If the court answered that question--as those who would apply the risk-utility test to this case suggest--the answer was "not addressed to the outcome of the ... case and therefore perhaps not as fully considered as it would have been if essential to the outcome." United States v. Crawley, 837 F.2d 291, 292 (7th Cir.1988). In Northwestern Nat. Ins. Co. v. Maggio, 976 F.2d 320, 323 (7th Cir.1992), we discussed the force we should give to state supreme court dicta in a diversity case:

Judicial opinions are frequently drafted in haste, with imperfect foresight, and without due regard for the possibility that words or phrases or sentences may be taken out of context and treated as doctrines. We shouldn't like this done to our opinions and are therefore reluctant to do it to the opinions of other courts. No court, even a federal court in a diversity suit, is obliged to treat a dictum of another court (or, for that matter, its own dicta) as binding precedent.

No Illinois court has applied the risk-utility test to a simple but obviously dangerous product. We need not expand upon the Illinois Supreme Court's already broad statement in Lamkin, and apply the test to such a product. If the question before us was, simply, "Does Illinois authorize the risk-utility test to determine whether a product is unreasonably dangerous?" the answer would be clear. Certainly, the Illinois Supreme Court has authorized use of the test in some cases. But the question before us is slightly different: "Does Illinois authorize use of the risk-utility test for a simple product which poses an obvious danger?" The answer to this question is not so clear. The Illinois Supreme Court has never applied the test to such a product. An intermediate appellate court has determined that it never will. Scoby, 155 Ill. Dec. at 540, 569 N.E.2d at 1151. Our task in a diversity case is to predict what the state's highest court would do if presented with the same issue. See Heller, 974 F.2d at 858; see also Wright, Miller and Cooper, Sec. 4507 at 94 ("If a state's highest court has not ruled on an issue, intermediate appellate court decisions constitute the next best indicia of what state law is.").8 Based on our examination of past and current Illinois case law, we agree with the appellate court which decided Scoby, that the Illinois Supreme Court would not apply the risk-utility test to simple but obviously dangerous products. Accord Mason v. Ashland Exploration, Inc., 965 F.2d 1421, 1428 (7th Cir.1992) (treating Scoby as authoritative reading of Lamkin ).

In summary, whether Illinois authorizes use of the risk-utility test is not disputed; clearly it does. See Lamkin, 150 Ill. Dec. at 570, 563 N.E.2d at 457. Whether this authorization requires that courts apply the test to all products liability cases is a point of contention. See Scoby, 155 Ill. Dec. at 540, 569 N.E.2d at 1151. When given a choice between an interpretation of Illinois law which reasonably restricts liability, and one which greatly expands liability, we should choose the narrower and more reasonable path (at least until the Illinois Supreme Court tells us differently). The dictum in Lamkin notwithstanding, we may assume, as an Illinois appellate court has done, that the Illinois Supreme Court would not apply the risk-utility test to a simple but obviously dangerous product. An ordinary disposable lighter is such a product. Therefore, the district court was correct to determine that such a lighter is obviously dangerous, but not unreasonably dangerous, without reference to the risk-utility test.

Two final issues remain for our consideration. First, Todd has appealed the district court's decision to grant summary judgment for Bic on his negligence claim. In his complaint, Todd claimed that Bic was negligent for breaching its duty to manufacture a safe lighter. It is true that manufacturers have a legal duty, under Illinois negligence and strict products liability law, to produce reasonably safe products. Coney, 73 Ill. Dec. at 343, 454 N.E.2d at 203. If the manufacturer breaches this duty, it is strictly liable; if the manufacturer breaches this duty because it failed to employ the requisite degree of care, it is also negligent. Id.; see also Phillips v. United States Waco Corp., 163 Ill.App.3d 410, 114 Ill. Dec. 515, 519, 516 N.E.2d 670, 674 (1987) ("the breach of duty is the same in both a negligence and a strict products liability claim, but the key distinction between a negligence claim and a strict liability claim lies in the fault concept."). We have already determined that the lighter in question was not unreasonably dangerous. It follows that Bic did not breach its duty to produce a reasonably safe product.

Secondly, in his appellate brief, Todd claimed that the warning Bic placed on the lighter was insufficient, and that the insufficiency made the lighter unreasonably dangerous. In the course of certifying this case to the Illinois Supreme Court, however, we decided the warning issue. We agreed with the district court that the warning Bic placed on the lighter was adequate as a matter of law. See Todd, 9 F.3d at 1219 ("Bic accordingly is entitled to summary judgment on the warning branch of this case."). We incorporate the sum and substance of that decision into this opinion. Certainly, the plain admonition "KEEP OUT OF REACH OF CHILDREN" provided adequate warning that the cigarette lighter should have been kept out of the reach of four-year-old Cori Smith.

The district court was correct in its conclusion that, as a matter of law, an ordinary disposable cigarette lighter is not unreasonably dangerous. Because Bic did not produce an unreasonably dangerous product, it was neither strictly liable nor negligent. Therefore, the judgment of the district court is

AFFIRMED.

CUMMINGS, Circuit Judge, dissenting.

I dissent for the reasons expressed by Judge Kaufman in his majority opinion in Todd v. Societe Bic, 991 F.2d 1334 (1993), and by Judge Flaum in his dissent herein.

CUDAHY, Circuit Judge, dissenting.

I join Judge Flaum's dissenting opinion. And, while I have the highest regard for my colleagues who may see the matter differently, I certainly agree with Judge Ripple that this question of Illinois products liability law has occupied a disproportionate share of this court's attention. This is, after all, a diversity case. And so far as I can recall during my tenure here, this is the first such case to receive en banc review. Nonetheless, and despite Judge Ripple's cogent arguments to the contrary, I suppose one bad turn deserves another. Having descended this far and at this cost into the morass, we might as well finish the job and put one more opinion of dubious authority in the books.

In my view as a federal observer of a state process, Judge Ripple's concurring opinion when we first visited this matter, see Todd v. Societe Bic, S.A., 9 F.3d 1216, 1224 (7th Cir.1993) (Ripple, J., concurring), along with Judge Flaum's opinion today, successfully demonstrate why the foreseeable user standard and the risk-utility test may properly apply here on the basis of the relevant Illinois cases. With respect to the foreseeable user standard, the majority goes to some pains to show the "absurdity" of using the knowledge and expectation of children as a measure of the "unreasonable dangerousness" of the product. Since children are unknowledgeable about danger, says the majority, and innocent of the ways of causation, they would contemplate little if any danger in almost any product. This would result in virtually absolute liability for the use or misuse of products by children. But the fact is that, in the case of products like lighters and knives, injury to some children is perhaps inevitable. The issue ought to be what party is best able to bear and distribute the cost of this inevitable injury--the child or the manufacturer. While the majority points to Illinois cases that suggest that it does not intend to make the manufacturer an insurer against injuries that result from the use of its product, Op. at 1405-06, cases like Doser v. Savage Mfg. & Sales, Inc., 142 Ill. 2d 176, 154 Ill. Dec. 593, 568 N.E.2d 814 (1990), and Lamkin v. Towner, 138 Ill. 2d 510, 150 Ill. Dec. 562, 563 N.E.2d 449 (1990), appear to point the other way. It is not clear to me whether employing a regime of tort law is preferable to requiring potential victims (or their parents) to go into the marketplace and buy insurance to guard against the cost of unavoidable accidents. While this question, in my view, should drive the policy analysis, our duty is not so much to evaluate policy as simply to predict (or guess at) the course of the Illinois courts.

With respect to the risk-utility test, again I am not sure that it is "absurd" to apply this measure to simple but obviously dangerous products like lighters and knives. Again, injury seems inevitable and the issue is whether it is properly part of the cost of manufacture or better borne by the victim. The point of strict products liability, after all, is "to minimize the costs of accidents and to consider who should bear those costs." Note, Strict Products Liability and the Risk-Utility Test for Design Defect: An Economic Analysis, 84 Colum.L.Rev. 2045, 2049-50 (1984) (quoting Suter v. San Angelo Foundry & Mach. Co., 81 N.J. 150, 406 A.2d 140, 151 (1979)). Perhaps it is patronizing to deny consumers the opportunity to purchase "riskier" products at a lower price. This is not our choice to make, but I see nothing irrational in Illinois' choosing the course it thinks most sensible and socially appropriate. I therefore respectfully dissent.

FLAUM, Circuit Judge, with whom CUMMINGS, CUDAHY and ILANA DIAMOND ROVNER, Circuit Judges join, dissenting.

I cannot accept the court's judgment and opinion. Our charge here is not to determine the best, most logical, or even the most efficient product-liability rules for the state of Illinois. If it were, I would be joining the court's persuasive opinion. However, Article III of the Constitution only affords us the power to ascertain the law of Illinois and apply it to the litigants before us as if we were an inferior state court. Because I remain convinced that the law of Illinois required the federal district court to apply the risk-utility doctrine in this product liability case, I conclude that our inquiry in this case should go no further.

When this court sits in diversity, federalism requires us to enforce the substantive law of the forum state, even when we conclude we see a more enlightened path. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S. Ct. 817, 822, 82 L. Ed. 1188 (1938) ("Except in matters governed by the Federal Constitution or by acts of Congress, the law to be applied in any case is the law of the state.... [a]nd whether the law of the state shall be declared by its Legislature in a statute or by its highest court in a decision is not a matter of federal concern."); see also Guaranty Trust v. York, 326 U.S. 99, 110, 65 S. Ct. 1464, 1470-71, 89 L. Ed. 2079 (1945) ("Erie R. Co. v. Tompkins has been applied with an eye alert to essentials in avoiding disregard of State law in diversity cases. A policy so important to our federalism must be kept free from entanglements with analytical or terminological niceties."), overruled on other grounds by Hanna v. Plumer, 380 U.S. 460, 85 S. Ct. 1136, 14 L. Ed. 2d 8 (1965). Federalism proscribes unwarranted federal judicial meddling in state matters because such interference would "prevent the informed evolution of state policy by state tribunals." Moore v. Sims, 442 U.S. 415, 429-30, 99 S. Ct. 2371, 2380, 81, 60 L. Ed. 2d 994 (1979), citing Trainor v. Hernandez, 431 U.S. 434, 445, 97 S. Ct. 1911, 1919, 52 L. Ed. 2d 486 (1977).

While I appreciate the sound and persuasive policy reasons for limiting access to a risk-utility jury test when a plaintiff has freely undertaken to use a product with obvious risks, the Illinois Supreme Court, in cases such as Lamkin, 138 Ill. 2d at 528, 150 Ill.Dec. at 570, 563 N.E.2d at 457, and Doser, 142 Ill. 2d at 197-98, 154 Ill.Dec. at 603, 568 N.E.2d at 824, appears to disagree. In Lamkin, the Illinois Court defined a defective product as one that fails the risk-utility test. See Lamkin, 138 Ill. 2d at 528, 150 Ill.Dec. at 570, 563 N.E.2d at 457. In Doser, the Illinois Court confirmed the appropriateness of the risk-utility test for products with patent dangers. 142 Ill. 2d 176, 154 Ill.Dec. at 598, 568 N.E.2d at 819. When analyzing unreasonably dangerous products in Illinois, "[t]he balancing of the likelihood and gravity of harm must be weighted against the burden of the precaution which would be effective to avoid the harm." Id., citing Burke v. Illinois Power Co., 57 Ill.App.3d 498, 511, 15 Ill. Dec. 670, 683, 373 N.E.2d 1354, 1367 (1st Dist.1978). As the Illinois Supreme Court explained, in Illinois, "[w]hether a product is unreasonably dangerous for failure to incorporate safety devices is ordinarily a question of fact which the jury should resolve." Doser, 142 Ill. 2d 176, 154 Ill.Dec. at 599 & 603, 568 N.E.2d at 820 & 824.

Neither Lamkin nor Doser hint at any limitation of the risk-utility doctrine in Illinois. While Scoby, a decision from the Illinois Fourth District Appellate Court, may have rejected the risk-utility test as a means of analyzing whether lid-covers are required on cooking pots,1 211 Ill.App.3d 106, 155 Ill.Dec . 536, 540, 569 N.E.2d 1147, 1151 (1991), Scoby cannot in any way constrain Lamkin and Doser. In a hierarchical judiciary, judges of inferior courts may not limit decisions of superior courts, despite their desire to improve the law. See Gacy v. Welborn, 994 F.2d 305, 310 (7th Cir.1993). It is a "fundamental principle of judicial construction that the lower judicial tribunals in Illinois are bound by decisions of the Illinois Supreme Court, and it is the duty of such tribunals to follow those decisions in similar cases." Greenlee v. Shedd Aquarium, 36 Ill.App.3d 924, 925, 344 N.E.2d 788, 790 (1st Dist.1976) citing Agricultural Transp. Ass'n. v. Carpentier, 2 Ill. 2d 19, 116 N.E.2d 863 (1953). Definitive pronouncements of the State's Supreme Court, such as Lamkin and Doser, are binding upon inferior courts applying State law, "contrary [Illinois] appellate court decisions notwithstanding." Greenlee, 36 Ill.App.3d at 925, 344 N.E.2d at 790. Whatever the preferences of this court, the Illinois Supreme Court's adoption of the risk-utility doctrine should not be refashioned by the interpretive prowess of this tribunal. Thus, I stand by my initial read of Illinois Supreme Court precedent, that Illinois' law appears to require a risk-utility analysis in product liability cases, and therefore I dissent.

RIPPLE, Circuit Judge, dissenting.

I doubt that few, if any, cases that have been decided during recent years have consumed the number of judge-hours that this case has cost this court. After full consideration by a panel of judges, full en banc consideration by the entire court, an attempt at certification, and the drafting and full consideration of a second en banc opinion, we remain divided on the content of Illinois law as well as the appropriate application of that law to the facts of this case. Today's decision resolves the litigation before us, but the bench and bar are provided with a decision that, as a practical matter, does nothing to clarify the law and therefore expedite the course of later litigation. Indeed, all that is decided is that an ordinary disposable cigarette lighter is such an obviously dangerous product that, when district courts of this circuit are confronted with such a case again, they ought not apply the risk utility test. I assume that the court's decision implicitly permits a different result even in that case if, in the interim, the courts of Illinois indicate that they would reach the contrary result. This minimal yield from such a significant effort on the part of so many judges ought to induce a great deal of reflection and a careful self-examination of our use of en banc procedures.1

Courts of appeals have wide latitude in determining whether to hear a case en banc. The Federal Rules of Appellate Procedure provide some, but minimal, assistance in this regard. Rule 35(a) provides in pertinent part:

(a) When Hearing or Rehearing In Banc Will be Ordered. A majority of the circuit judges who are in regular active service may order that an appeal or other proceeding be heard or reheard by the court of appeals in banc. Such a hearing or rehearing is not favored and ordinarily will not be ordered except (1) when consideration by the full court is necessary to secure or maintain uniformity of its decisions, or (2) when the proceeding involves a question of exceptional importance.

When the court decides to devote its time to hearing a case en banc, it is determining that such a significant investment of time and energy is necessary to accomplish legitimate jurisprudential goals beyond decision in the individual case. Such "legitimate jurisprudential goals" include, as our rules note, the resolution of conflicts between this circuit and other courts of appeals. It also includes the resolution of novel and important issues of federal law and recurring procedural issues of enduring importance to the administration of justice in the federal courts. On the other hand, few would maintain that "legitimate jurisprudential goals" include forsaking the normal course of common law adjudication in order to impose ideological discipline on fellow federal judges. Nor do such legitimate goals include attempting to set the vectors for the future course of state law development, a prerogative left by the Constitution to the state courts.

The adjudication of diversity cases affords little opportunity for the accomplishment of the legitimate goals of an en banc proceeding. Indeed, beyond rendering a decision in the particular case at hand, little of enduring value can be accomplished by hearing a diversity case en banc. In any diversity case, this court's decision has no precedential effect on the state courts. Diginet, Inc. v. Western Union ATS, Inc., 958 F.2d 1388, 1395 (7th Cir.1992) ("State courts are not bound by federal courts' interpretations of state law."); Cohn v. Checker Motors Corp., 233 Ill.App.3d 839, 175 Ill. Dec. 98, 103, 599 N.E.2d 1112, 1117 (1992) (stating that Illinois courts have no obligation to follow decisions of federal courts interpreting state law). While the decision is binding on the federal trial courts--at least as long as the decision's underpinnings are left intact by subsequent state court activity--it is rarely possible, given the difficulty2 in making "Erie Guesses,"3 to provide, in any principled fashion, a great deal of meaningful guidance for the resolution of future cases. Moreover, federal court pronouncements on the content of state law inherently involve a significant intrusion on the prerogative of the state courts to control that development. As Chief Judge Jon Newman of the Second Circuit has pointed out, federal court pronouncements on state law in the exercise of its responsibilities under the Erie doctrine have the potential of disrupting the normal pattern of development of state law. "One distinct shortcoming of diversity jurisdiction is the interruption of the orderly development and authoritative exposition of state law occasioned by sporadic federal court adjudications." Factors Etc., Inc. v. Pro Arts Inc., 652 F.2d 278, 282 (2d Cir.1981), cert. denied, 456 U.S. 927, 102 S. Ct. 1973, 72 L. Ed. 2d 442 (1982). When the federal court's "Erie Guess" is wrong, the resulting impact on state law developments is, of course, especially significant. As Chief Judge Sloviter has written:

Until corrected by the state supreme court, such incorrect predictions inevitably skew the decisions of persons and businesses who rely on them and inequitably affect the losing federal litigant who cannot appeal the decision to the state's supreme court; they may even mislead lower state courts that may be inclined to accept federal predictions as applicable precedent.

Dolores K. Sloviter, A Federal Judge Views Diversity Jurisdiction through the Lens of Federalism, 78 Va.L.Rev. 1671, 1681 (1992).

All of these limitations and dangers in the adjudication of diversity cases are magnified when the court decides to hear such a matter en banc. As noted earlier, by their nature, en banc proceedings are intended to decide more than the case before the court and to marshall the intellectual and moral authority of the entire court on the issue presented. Yet, these goals are, in the context of the diversity case, frustrated ab initio. Whatever the court's decision, it can have no binding authority on the state courts and, in future cases, lower federal courts must still take into consideration subsequent state caselaw developments. An erroneous decision by the en banc court--or its gratuitous comment on what state law ought to be--is a particularly unwarranted intrusion into the prerogative of the state to control the development of its own law.

These considerations counsel that en banc adjudication in diversity cases ought to be undertaken only in the most unusual circumstances. The circumstances surrounding this litigation certainly make the exercise of our en banc authority questionable. We have labored greatly and produced a decision that does little, if anything, to promote a legitimate goal of an en banc proceeding. The result is no doubt pleasing, from an ideological point of view, to the majority of the court and it can now impose that ideological position, at least temporarily, on the district courts. Yet, whether the majority has interpreted properly Illinois law remains, at best, a rather weak "Erie Guess." The imprudence of the course chosen by the court is, in my view, particularly underlined by the fact that it appears that BIC's concession in the district court that the child was a foreseeable user creates a genuine issue of triable fact with respect to the consumer contemplation test. See Doser v. Savage Mfg. & Sales, Inc., 142 Ill. 2d 176, 154 Ill. Dec. 593, 603, 568 N.E.2d 814, 824 (1990).4 The court's expounding on the risk utility test is therefore unnecessary and a significant intrusion into the prerogatives of the state. Accordingly, I would vacate the order granting rehearing en banc as having been improvidently entered.