CENTRAL INTERNATIONAL COMPANY, Plaintiff, Appellant, v. KEMPER NATIONAL INSURANCE COMPANIES, American Motorists Insurance Company, Defendants, Appellees, 202 F.3d 372

BOUDIN, Circuit Judge. This marine insurance case, which turns on the meaning of an exclusion for “discoloration and corrosion,” begins with an ocean shipment made in December 1995. At that time, the plaintiff-appellant, Central International Company (“Central”) — a trading company doing business in Massachusetts — arranged for the transportation of galvanized steel coils from Aviles, Spain, to a customer in St. Vincent, West Indies, aboard the M/V Andrealon.

The shipment was insured under an “all risk” ocean marine open cargo insurance policy previously issued to Central by the principal defendant-appellee, American Motorists Insurance Company (“AMICO”). The policy was not limited either to this shipment or to steel products. The vessel in question encountered heavy seas en route to St. Vincent, resulting in damage to the vessel and its hatches. It is apparently undisputed that the steel coils in question were in good condition when shipped but arrived in damaged condition.

The laboratory analyses conducted on the damaged cargo showed that the coils were “severely corroded” and displayed a variety of light and dark stains. The initial direct loss, independent of consequential damages, was estimated by Central at several hundred thousand dollars. Central then brought suit on its insurance policy in a Massachusetts trial court seeking approximately $221,000 in damages to the coils plus additional related expenses. For simplicity, we treat AMI-CO as the only defendant.

AMICO removed the case to federal district court on the ground that it fell within that court’s admiralty or maritime jurisdiction, 28 U.S.C. § 1333(1). In due course, both sides moved for

summary judgment. In April 1999, the district court filed a decision granting AMICO’s motion for

summary judgment and ordering dismissal of the complaint. The district court pointed out that the policy covered all risks of “physical loss or damage [to the cargo] from any external cause” but was subject to a specific exclusion: However, as respects steel products and all metals: excluding rusting, oxidation, discoloration and corrosion; also exelud-ing bending, twisting, crimping, and end damage. Since corrosion and discoloration were the admitted damage, the district court found that the policy exclusion applied.

The court rejected Central’s effort to avoid the exclusion by saying that the corrosion and discoloration had not occurred naturally but had been primarily caused by water and by the improper stowage of powder ash, which had allegedly leaked onto the steel coils during the storm or otherwise. Central now appeals. Our review of a

summary judgment is de novo, EEOC v. Green, 76 F.3d 19, 23 (1st Cir.1996), and we begin with the choice of law. Suits on maritime insurance policies are classic examples of matters within federal maritime jurisdiction, see Gilmore & Black, The Law of Admiralty § 1-10, at 22 n. 71 (2d ed.1975) (citing Insurance Co. v. Dunham, 78 U.S. (11 Wall.) 1, 20 L.Ed. 90 (1871)).

Accordingly, Central’s brief says that such contracts are “subject to general admiralty law when there is an established federal rule and by state law when there is not.” AMICO purports to agree but says that there is enough federal precedent in analogous cases to make Massachusetts law irrelevant. Beneath this surface agreement on general principles lies an abyss of confusion. One might think that construing a maritime insurance policy, in relation to damage occurring on the high seas, would be a paradigm case for a uniform body of federal law, see Gilmore & Black, supra, § 1-17, at 49.

But the tensions in Supreme Court precedents are legendary, both with regard to the reach of state law in federal maritime law generally and its effect on cases initially brought in state court under the saving to suitors clause. Since Massachusetts law in the end is no better for Central than any federal precedent cited to us, we (like the district court) need not pursue the matter. Turning to the merits, Central must accept in light of its own affidavits that the harm to the steel coils manifested itself as corrosion and discoloration.

Conversely, at least for purposes of this appeal, we must assume the truth of Central’s claim that the anterior cause of the corrosion and discoloration was the exposure of the steel to improperly stowed powder ash as well as to sea water and that unexpected storm conditions at sea contributed to that exposure. See Green, 76 F.3d at 21. The case thus involves a generally covered risk (actually, two risks: sea peril and negligent stowage) resulting in excluded consequences (corrosion and discoloration). This is a recurring issue and, under ordinary principles of contract

interpretation, there is little doubt that the exclusion is presumptively a qualification on the risk coverage. Normally, specific language is treated as a limitation on general language; and in this case fixing the relationship of the two clauses is made even easier because the all-risk language in Central’s policy is followed by the term “[hjowever,” which introduces the exclusion. In other words, facially read, there is liability for damage to cargo from all risks including storm or accident unless the damage is corrosion or discoloration.

Curiously, this straightforward approach is not always reflected in the cases. There is language, in Massachusetts decisions as elsewhere, that purports to treat a covered risk and an excluded consequence as legitimate rivals whose priorities are to be tested by asking questions about “causation”; these discussions, often in opinions with defensible results, tend not to be very illuminating. See Keeton & Widiss, Insurance Law § 5.5(c), (d) (Practitioner’s ed.1988).

The better explanation, where the insurer is held liable, appears to be that the court has chosen to read the exclusion in context more narrowly than its literal language might suggest. A good example in Massachusetts is Standard Elec. Supply Co. v. Norfolk & Dedham Mut. Fire Ins. Co., 1 Mass.App. Ct. 762, 307 N.E.2d 11 (1973), where water from a burst pipe leaked into and damaged the basement of an adjoining building; the policy on the latter building generally covered physical damage but specifically excluded water damage in various specified forms (e.g., flood, underground seepage).

Yet recovery was allowed, seemingly because the court thought the exclusion in context was directed at natural flooding or seepage and not that resulting from an accidentally burst pipe. 307 N.E.2d at 12-13. What is confusing is the court’s reliance, later in the opinion, on an alleged “well established principle” that recovery is allowed “where the insured risk [caused] ... an excepted risk.” Id. at 13.

The smattering of authority for this “well-established” principle is not very impressive, see, e.g., 5 Appleman & Appleman, Insurance Law and Practice § 3083, at 311 (1970) (discussing fire insurance), and the cases usually- — but perhaps not always, see Jussim v. Massachusetts Bay Ins. Co., 415 Mass. 24, 610 N.E.2d 954, 955-58 (1993) — can be better explained as a limiting construction of the exclusion.

The attempt to give priority to the initial covered cause in the “train of events” over the excluded consequence is doubtful in principle even where the covered risk is narrowly defined {e.g., fire), but it becomes absurd where the initial coverage is all risk, since every excluded harm has some cause. Still, there is nothing wrong with reading policy language more narrowly (or broadly) than its literal wording might at first suggest where this better captures the reasonable expectation of the parties— the central object of all contract

interpretation. See 1 Corbin, Contracts, supra, § 1.1, at 2; see also Commercial Union Ins. Co. v. Walbrook Ins. Co., 7 F.3d 1047, 1052 (1st Cir.1993). Courts have made such adjustments on a range of bases, including not only surrounding language or other specific evidence of the parties’ intent but also — to a surprising extent — the common-sense intuition of judges as to what the parties likely had in mind. Although the readings often favor the insured, implied exceptions to coverage are not unknown. See Keeton & Widiss, supra, § 5.3(a), (c). In this case, Central does make a pair of colorable arguments for a narrowing

interpretation of the exclusion. The first is that the exclusion should apply only to naturally occurring corrosion or discoloration, say, as a result of exposure to salt air, and not to that produced by a fortuitous event like storm exposure or negligent stowage of adjacent cargo. This is not an impossible

interpretation, cf. Standard, 307 N.E.2d at 12-13 — an insurer might specially wish to prevent such liability for natural corrosion — but, of course, there is nothing in the policy to suggest such a limitation, and at least two further arguments exist against it. One is that some authority exists for excluding natural deterioration of shipped goods from coverage without a specific exclusion, see Keeton & Widiss, supra, § 5.3(c), which would make the express exclusion from corrosion redundant if it were read so narrowly.

The - other is that the full exclusion in this policy includes some excluded harms {e.g., bending of the steel) that could rarely occur “naturally” in a voyage, so Central’s argument implies that one part of the exclusion is subject to an implicit limitation that could not be applied to the other part — hardly a natural reading.

Central’s other argument for coverage is that unless the exclusion were read narrowly, the protection offered for steel shipments would be empty; its premise is that the only foreseeable damage to steel on a sea voyage would be discoloration, corrosion or some form of physical distortion covered by the above-quoted “bending” clause of the exclusion. We agree that a reading of a policy that entirely nullified meaningful protection would raise doubts that the parties intended that reading. E.g., Titan Indem. Co. v. Newton, 39 F.Supp.2d 1336, 1344-45 (N.D.Ada.1999); Employers Mut. Cas. Co. v.

Pires, 723 A.2d 295, 298 (R.I.1999). Yet it is not true that all maritime loss as to steel would trigger the literal language of the exclusion; if the ship sank in deep water, the inability to recover the cargo would create a covered loss without regard to any corrosion or discoloration that might ensue. Further, this policy was “open,” embracing various kinds of cargo that Central might ship on different voyages; the failure to provide much protection for steel did not render the policy generally useless.

And it appears from the policy that the exclusion could have been modified, although perhaps not to provide unlimited coverage for steel, by payment of a higher premium. As for case law, neither side has cited us to maritime law cases that are closely in point, but there is no reason to believe that on this issue maritime law would differ from the ordinary application of contract rules customarily applied by state courts in insurance cases.

However, Massachusetts case law does offer a number of examples, including the burst pipe case described above, which are somewhat helpful to Central — even beyond their support for the view (with which we agree) that the presumptive literal reading is not always the one that prevails. But the precedent most closely in point from Massachusetts — whose law Central wants to govern this case' — cuts the opposite way. In Bettigole v.

American Employers Insurance Co., 30 Mass.App.Ct. 272, 567 N.E.2d 1259 (1991) (Kaplan, J.), the Court held that an insurance policy exclusion for “corrosion” negated coverage where the supporting steel in a parking facility deteriorated after exposure to ice, water and de-icing salts carried in by cars, id. at 272-73, 567 N.E.2d 1259; the court rejected the insured’s claim that the true “cause” was the “release of chloride ions, which was not named as an excluded risk, with the corrosion following as a consequence,” id. at 275, 567 N.E.2d 1259.

We cite Bettigole primarily because its facts are apt and not because we attach conclusive weight to Massachusetts case law construing a particular policy. The Supreme Court’s intermittent resort to state law to resolve maritime law issues usually involves not random state-law precedents but something closer to an affirmative state policy, typically in cases involving local maritime matters {e.g., coastal or inland waterways). Gilmore & Black, supra, § 1-17. Thus, there may be no compulsion here to treat state precedent as binding rather than merely instructive, cf. Southworth Mach. v.

F/V Corey Pride, 994 F.2d 37, 41 (1st Cir.1993), but Bettigole is certainly instructive. Central has also sought review of the district court’s refusal to allow Central leave to amend in order to add a very large claim for consequential damages over and above the value of the coils. Since we affirm the district court’s reading of the exclusion, such an amendment would be futile; whether or not consequential damages are consistent with maritime law, they can hardly be awarded under this policy to offset the further consequences of an expressly excluded loss. The judgment is affirmed.

AMICO’s motion for damages and double costs is denied. . See American Dredging Co. v. Miller, 510 U.S. 443, 452, 114 S.Ct. 981, 127 L.Ed.2d 285 (1994) ("It would be idle to pretend that the line separating permissible from impermissible state regulation is readily discernible in our admiralty jurisprudence, or indeed is even entirely consistent within our admiralty jurisprudence.”). Compare Wilburn Boat Co. v. Fireman’s Fund Ins. Co., 348 U.S. 310, 75 S.Ct. 368, 99 L.Ed. 337 (1955), and Caldarola v. Eckert, 332 U.S. 155, 67 S.Ct. 1569, 91 L.Ed. 1968 (1947), with Kossick v.

United Fruit Co., 365 U.S. 731, 81 S.Ct. 886, 6 L.Ed.2d 56 (1961). See generally Gilmore & Black, supra, §§ 1-17, 1-18. . See Restatement (Second) of Contracts § 203(c) (1979); 3 Corbin, Contracts § 24.23 (Joseph M. Perillo ed., rev. ed.1993); cf. Bordenave v. Intermedics Intraocular, Inc., 56 F.3d 703, 707 (5th Cir.1995) (similarly, in statutory construction, “lex generalis non de-rogat speciali "). . Sabella v. Wisler, 59 Cal.2d 21, 27 Cal.Rptr. 689, 377 P.2d 889 (1963); Jussim v. Massachusetts Bay Ins. Co., 415 Mass. 24, 610 N.E.2d 954 (1993); Standard Elec. Supply Co. v. Norfolk & Dedham Mut. Fire Ins.

Co., 1 Mass.App.Ct. 762, 307 N.E.2d 11 (1973); Fawcett House, Inc. v. Great Cent. Ins. Co., 280 Minn. 325, 159 N.W.2d 268 (1968); Franklin Packaging Co. v. California Union Ins. Co., 171 N.J.Super. 188, 408 A.2d 448 (1979). . Of course, as Jussim itself points out, the proximate cause concept is elastic enough to make an event at the start of a "chain” the responsible cause of one that occurs several steps later, 610 N.E.2d at 955-57, but this is hardly itself a basis for ignoring an exclusion that appears to describe the resulting loss.

Jussim's reliance on an earlier proximate cause case, which did not involve an exclusion, id. at 956 (citing Lynn Gas & Elec. Co. v. Meriden Fire Ins. Co., 158 Mass. 570, 33 N.E. 690 (1893)), suggests the source of the confusion.