Stone Container Corporation v. Hartford Steam Boiler Inspection and Insurance Company, Cross-Appellee

POSNER, Chief Judge. The plaintiff in this insurance suit, Stone Container Corporation, is a large manufacturer of pulp, paper, and paper products. It makes the pulp in huge steel tanks called “pulp digesters.” Wood chips are placed in the tank along with chemicals. The tank is then sealed and its contents subjected to heat and pressure from steam piped into the tank, causing the chips to decompose into pulp *1159 fiber. One of the tanks in one of Stone’s plants exploded when a thin area of its steel shell ruptured during the high-pressure operation of the tank. The explosion blew a 28-ton chunk into the air; it landed more than 200 feet away with disastrous results. Besides much property damage, several workers were killed. The plant was forced to shut down for months. Stone Container incurred total losses in excess of $80 million. Stone had an “all-risks” insurance policy from Lloyd’s that, the parties agree, covered an accident of this kind. But it also had a “boiler and machinery insurance” policy from Hartford Steam Boiler Inspection and Insurance Company. Lloyd’s believed that Hartford’s policy was primary and that Lloyd’s should have to pay on its own policy only if Hartford was determined not to be legally obligated to pay for Stone’s losses. Because the accident caused liquidity problems for Stone, Lloyd’s was able to make a deal whereby Stone agreed to sue Hartford (which had denied coverage) and in exchange Lloyd’s lent Stone one-half of the insurance proceeds to which Stone would be entitled if it won the suit against Hartford. It is unclear to us what incentive Stone had to press such a suit vigorously, the dispute really being between the insurance companies; but it has done so. The Hartford policy is limited to accidents to particular enumerated “objects” in Stone’s plants. The enumeration covers a broad range of different types of machinery, but there is an exclusion for losses caused by “explosions.” There is also an exception to the exclusion. The exception is “for loss caused by or resulting from an explosion of an ‘object’ of a kind described below ...: Explosion of any: (1) Steam boiler; (2) Electric steam generator; (3) Steam piping; (4) Steam turbine; (5) Steam engine; (6) Gas turbine; or (7) Moving or rotating machinery [if the explosion is] caused by centrifugal force or mechanical breakdown.” Anything within the exception is covered by Hartford’s policy. If, therefore, either the accident to the pulp digester was not an “explosion,” and so was not within the exclusion, or it was an explosion of “an ‘object’ of a kind described” in the list in the exception to the exclusion, and so was within the exception, then Hartford’s policy covers the accident; otherwise it does not. The district judge granted summary judgment for Stone. He held that although the accident was indeed an explosion, the policy is ambiguous as to whether a pulp digester is an object “of a kind” described in the list of kinds of machinery excepted from the exclusion, and an ambiguity in an insurance contract is, under Illinois law, which governs the substantive issues in this diversity suit, to be resolved in favor of the insured. He refused to allow Hartford to present evidence to disambiguate the ambiguity. Hartford has appealed, arguing that the policy unambiguously excludes pulp digesters; in the alternative it asks for a remand to enable it to present evidence of drafting history and the like to show that the parties intended to exclude pulp digesters. Stone has cross-appealed, arguing an alternative ground for affirmance of the judgment to the district court’s ground: that what happened to the pulp digester was not an explosion. Of course Stone did not have to file a cross-appeal to argue for an alternative ground of affirmance, and in fact should not have done so. A cross-appeal is necessary and proper only when the appellee wants the appellate court to alter the judgment (the bottom line, not the grounds or reasoning) of the district court. Coe v. County of Cook, 162 F.3d 491, 497 (7th Cir.1998). We shall take up the alternative ground for affirmance first. Stone argues that “explosion” means, for exclusion purposes at any rate, a sudden and violent release of energy (which of course we have here) caused by combustion or some other chemical reaction (which we don’t have here). The qualification “for exclusion purposes” is noteworthy. Stone believes that the same word should be read narrowly when it appears in an exclusion from coverage, and broadly when it appears in an exception to an exclusion, even if the context is the same. Thus, “explosion” might mean a sudden and violent release of energy caused by combustion or some other chemical reaction in the exclusion clause but in the exception to the exclusion clause might mean any rupture. *1160 Stone offers no support for this suggestion, in cases or other recognized legal authorities, beyond the principle that ambiguities in insurance contracts should be resolved in favor of the insured: the principle of “contra prof-erentem,” well discussed in Jeffrey W. Stempel, Interpretation of Insurance Contracts: Law and Strategy for Insurers and Policyholders § 5.2 (1994), which Illinois applies even where, as in this case, the insured is a large and sophisticated firm, provided it didn’t actually negotiate over the terms of coverage. Outboard Marine Corp. v. Liberty Mutual Ins. Co., 154 Ill.2d 90, 180 Ill.Dec. 691, 607 N.E.2d 1204, 1218-19 (Ill.1992); see generally Stempel, supra, ch. 23. Stone’s suggestion would make insurance contracts even more complex, esoteric, and inscrutable than they are already. In any event, the proposed definition of “explosion” is not only narrow, but weirdly narrow. It seems to exclude the explosion of an atomic bomb, since a nuclear reaction is not a form of combustion or a chemical reaction, at least in the usual senses of these words. It would certainly exclude volcanic explosions, as well as the “explosion” of a tire caused by a blowout, the explosion of a melon caused by a bullet, and, to take an example very close to home, the explosion of a boiler as a result of the failure of a valve to open. All these are commonplace examples of the use of the word “explosion” in ordinary speech. None involves stretching the ordinary meaning. And likewise a blast that blows 28 tons of steel and concrete more than 200 feet away is the ordinary person’s idea of an explosion, whatever the precise cause of the explosion. The Hartford policy does not define the word or scatter any clues that it is being used in other than its normal sense; even the engineering firm that Stone hired to investigate the accident called it an explosion — a “Boiling Liquid Expanding Vapor Explosion (BLEVE) of a large, steam-pressurized vessel.” And Stone has cited no ease that impresses an artificial definition on the term. The case law, fatally to Stone on this issue, gives the word “explosion” when it appears without a definition in an insurance contract its ordinary-language meaning. Pre-Cast Concrete Products, Inc. v. Home Ins. Co., 417 F.2d 1323 (7th Cir.1969) (discussing and applying Illinois law); Lever Bros. Co. v. Atlas Assurance Co., 131 F.2d 770, 775-76 (7th Cir.1942); American Casualty Co. v. Myrick, 304 F.2d 179, 182-83 (5th Cir.1962). Stone’s argument is at root that if a term is not defined in an insurance contract, the insured can impress any definition on it that will establish coverage. “You wrote it, you lose” is not the insurance law of Illinois. It is a slightly closer question whether the pulp digester is “of a kind” listed in the list of objects that are excepted from the exclusion of accidents caused by explosions, that is, that are covered by the boiler and machinery policy. Clearly the pulp digester is not any of the listed “objects.” It is closest to a steam boiler, because it employs steam under pressure. But a steam boiler creates steam by boiling water. In the pulp digester the steam is generated outside and fed into the digester; the digester does not create steam. Its function and mode of operation are thus completely different from those of a steam boiler. In engineering lingo, the steam boiler is a “fired pressure vessel,” the pulp digester an “unfired pressure vessel.” Had the pipe that carried the steam to the digester exploded, the explosion would have been covered by Hartford’s policy, because steam piping is one of the objects enumerated in the exception to the exclusion. But it is the digester itself that, as a result of the rupture in its wall, blew up. The pulp digester is, then, certainly not a steam boiler or even a kind of steam boiler; but might it be “of a kind” with a steam boiler? If so, the explosion exclusion from the policy is hopelessly ambiguous and probably illusory, for when pressed at argument for examples of explosions that Stone’s reading of the Hartford policy would exclude, Stone’s lawyer could suggest only the explosion of the gasoline tank of a truck that was in the plant. The objects covered by the policy include air-conditioning units. Suppose one exploded. Is an air-conditioning unit “of a kind” with the seven objects enumerated in the explosion exclusion? A boiler heats water; an air conditioner cools air; these could be thought analogous functions *1161 (heat exchange). An industrial air conditioner, the kind presumably found in Stone’s plants, uses water as an intermediary between the refrigerant and the air, and the water changes temperature. Even if the essential commonality of the objects embraced by the exception to the exclusion (setting aside the last and most clearly irrelevant type of object, moving machinery) is the use of steam, does this mean that an espresso machine, the radiator of a motor vehicle, a Sauna bath, a dishwasher, a steam iron, a humidifier, and a teapot are all “of a kind” with a steam boiler? We think not. Stone’s error is its refusal to read “of a kind” contextually. The term introduces a list of kinds of object. “Steam boiler” denotes a class of objects, not a single object. A class is a kind; the phrase “of a kind” introduces the various kinds or classes of object subject to the explosion exclusion. Steam boilers are one kind; steam pipes another; and so on. Pulp digesters are a kind of object, but not one of the kinds in the list. The distinction between fired and unfired pressure vessels helps to show this. These are two different kinds of pressure vessel. One includes steam boilers but not pulp digesters; the other includes pulp diges-ters but not steam boilers. One is covered by the boiler and machinery insurance policy; the other is not. This is clear enough to compel judgment for Hartford without an evidentiary hearing. We agree with Stone that it is desirable to resolve insurance disputes where possible without a trial likely to drain the resources of the insured. Rhone-Poulenc Inc. v. International Ins. Co., 71 F.3d 1299, 1305 (7th Cir.1995) (Illinois law). We also agree with Stone that ambiguities are to be resolved in favor of the insured. That is not only the rule in Illinois, but the rule everywhere. 2 Eric Mills Holmes, Holmes’s Appleman on Insurance, 2d § 6.1, p. 134 and n. 4 (1996); see also 2 Lee R. Russ, Couch on Insurance Sd § 22:14, p. 22-31 n. 23 (1997). But we do not think that “of a kind,” read in context as all contractual language must be read, is ambiguous. Read as Stone would have us read it, it would introduce radical ambiguity into an exclusion designed to be mechanically applicable and would expand coverage beyond its intended scope. The form in which the insurance policy extends coverage to boiler explosions, that is, by means of an exception to an exclusion, obscures what is at issue here. This is a boiler and machinery policy, which gives a manufacturer or other user of a narrow range of industrial equipment in which Hartford specializes additional protection for accidents involving the enumerated items, which besides moving or rotating machinery consist of steam boilers and closely related, specifically enumerated types of equipment. Stone wants to convert it to an “all risks” policy. There are, we conclude, compelling reasons to reject Stone’s reading of “of a kind” without need for a trial. But we add, lest our silence on the point mislead, that we disagree with Stone that when a term in an insurance contract is ambiguous, the insured is entitled to judgment in its favor without the insurance company’s being allowed to present evidence to disambiguate the ambiguity. As we held in Rhone-Poulenc Inc. v. International Ins. Co., supra, 71 F.3d at 1305, applying Illinois law, the rule that ambiguities in insurance contracts are to be resolved in favor of the insured comes into play only after the insurance company has had an opportunity to present evidence designed to dispel the ambiguity. Rhone-Poulenc does not, as Stone argues, stand alone. There is a legion of similar holdings. See, e.g., University of Illinois v. Continental Casualty Co., 234 Ill.App.3d 340, 175 Ill.Dec. 324, 599 N.E.2d 1338, 1345 (Ill.App.1992); Harnischfeger Corp. v. Harbor Ins. Co., 927 F.2d 974, 976 (7th Cir.1991); Harbor Ins. Co. v. Continental Bank Corp., 922 F.2d 357, 366 (7th Cir.1990) (Illinois law); In re Prudential Lines Inc., 158 F.3d 65, 77 (2d Cir.1998); United States v. Insurance Co. of North America, 131 F.3d 1037, 1042, 1043 n. 11 (D.C.Cir.1997); 12th Street Gym, Inc. v. General Star Indemnity Co., 93 F.3d 1158, 1166 (3d Cir.1996); see also Stempel, supra, § 5.5. It is also worth noting that the rule that ambiguities in insurance contracts are to be resolved against the insurer is an application of the broader rule of contract law (also called contra proferentem, though the rule *1162 has greater force in the insurance setting, see Stempel, supra, ch. 5 — as if to “insure” the insured against the risk of losing coverage because of ambiguous language in the policy) that ambiguities in a contract are to be resolved against the party that drafted the contract, and that in that setting as well the drafting party is entitled to present extrinsic evidence (that is, evidence outside the contract itself) to disambiguate the ambiguity. E.g., Howard A. Koop & Associates v. KPK Corp., 119 Ill.App.3d 391, 75 Ill.Dec. 276, 457 N.E.2d 66, 72 (1983); Baker v. America’s Mortgage Servicing, Inc., 58 F.3d 321, 327 (7th Cir.1995) (Illinois law); Residential Marketing Group v. Granite Investment Group, 933 F.2d 546, 549 (7th Cir.1991) (same). Courts in a minority of jurisdictions, it is true (but not Illinois), distinguish between “patent” and “latent” ambiguities in insurance contracts and hold that extrinsic evidence can be used only to disambiguate the latter. See, e.g., American National Fire Ins. Co. v. Rose Acre Farms, Inc., 107 F.3d 451, 457-58 (7th Cir.1997) (Indiana law); Charter Oil Co. v. American Employers’ Ins. Co., 69 F.3d 1160, 1163 (D.C.Cir.1995) (Missouri law). Stone seems to be appealing to this rule in arguing that even if “of a kind” is ambiguous, Hartford has no right to present evidence about its meaning. The distinction between the two types of ambiguity, though not the twist that the minority rule gives to it, is not peculiar to insurance law. A patent ambiguity in a contract is one that is apparent from just reading the contract. A latent ambiguity arises when, although the contract is clear “on its face,” anyone knowing the background would know that it didn’t mean what it seems to mean. AM International, Inc. v. Graphic Management Associates, Inc., 44 F.3d 572 (7th Cir.1995) (Illinois law). A latent ambiguity thus requires extrinsic evidence to establish, as well as to resolve, and only objective evidence may be used for these purposes. Id. at 575; Mathews v. Sears Pension Plan, 144 F.3d 461, 467 (7th Cir.1998); Kerin v. United States Postal Service, 116 F.3d 988, 992 n. 2 (2d Cir.1997). The minority rule for insurance contracts seems to be based on the idea that if a term in an insurance contract obviously is ambiguous, the insured should be able to rely upon that meaning within the range of possible meanings that confers coverage. As Illinois does not have this rule, and anyway neither “explosion” nor “of a kind” is ambiguous in the context of Hartford’s boiler and machinery policy, there is no need for us to opine on the merits or demerits of the minority rule. The judgment is reversed with directions to enter judgment in favor of Hartford. Reversed.