Learn About the Law
Get help with your legal needs
FindLaw’s Learn About the Law features thousands of informational articles to help you understand your options. And if you’re ready to hire an attorney, find one in your area who can help.
SHILOH CHRISTIAN CENTER, Plaintiff-Appellant, v. ASPEN SPECIALTY INSURANCE COMPANY, Defendant-Appellee.
This is an insurance case. Fear not, keep reading. In determining whether a pair of insurance policies cover losses resulting from “named windstorms,” we have to decide an important and (as it turns out) interesting question about the interpretation of written legal instruments: What is a court to do when all the surest proof of contracting parties’ subjective intentions and expectations flatly contradicts the surest indicators of an agreement's objective legal meaning?
At the risk of oversimplifying, Aspen Specialty Insurance Company, a billion-dollar insurance conglomerate, has essentially all of the subjective-intent evidence on its side: The records of the contracting parties’ course of dealing, contractual negotiations, and policy applications strongly suggest that the parties intended and expected that the policies would exclude damage caused by named windstorms. But Aspen's policyholder—Shiloh Christian Center, a small Florida church—has the text: However clear the parties’ subjective intentions or expectations, the policies do not, by their plain terms, exclude named-windstorm-related losses.
What, then? The district court found the evidence of the parties’ subjective intent overwhelming and accordingly granted summary judgment to Aspen. We hold, to the contrary, that, under Florida law—as in the law more generally—in the event of a conflict between clear text, on the one hand, and even compelling evidence of extra-textual “intent,” on the other, the latter must give way to the former Cf. CRI-Leslie, LLC v. Comm'r of Internal Revenue, 882 F.3d 1026, 1033 (11th Cir. 2018). We therefore reverse the district court's decision and remand for further proceedings.
In 2016 and 2017, respectively, Hurricanes Matthew and Irma tore through Melbourne, Florida, pummeling Shiloh Christian Center. On both occasions, the storms peeled back the church's roof, allowing rain to soak the exposed structure.1
In 2015, the year before Matthew hit, Shiloh's property-insurance policy with Aspen Specialty Insurance Company covered losses resulting from so-called named windstorms—i.e., hurricanes. In the middle of that year, though, Shiloh specifically asked Aspen to stop covering named-windstorm-related losses. Aspen agreed and issued an endorsement implementing the requested change: “THIS ENDORSEMENT CHANGES THE POLICY. PLEASE READ IT CAREFULLY․ It is understood and agreed effective 7/16/2015, the following change is made to this policy: Named Windstorm coverage is removed from this policy.” Doc. 25-4 (emphasis in original). Reflecting the amendment, Aspen reduced Shiloh's premium and even refunded its past payments for named-windstorm coverage.
In early 2016, Shiloh began negotiations to renew its policy with Aspen. An insurance broker gave Shiloh a quote for “the same coverage provided after the Return Premium endorsement was issued last year”—that is, the post-amendment coverage that “exclude[d] Named Storms.” Doc. 25-3 at 15 (email from Shiloh). In its application for the policy, Shiloh scribbled “EX wind” in the section labeled “forms and conditions to apply” for several of the covered premises. Doc. 25-7 at 4, 7. Aspen then issued a binder—which, for the uninitiated, is “a contract ․ for interim insurance” that is “effective at the date of the application and terminates at either the completion or rejection of the principal policy.” Medley Warehouses, LC v. Scottsdale Ins. Co., 39 So. 3d 440, 444–45 (Fla. 3rd Dist. Ct. App. 2010) (quotations and brackets omitted). The binder described the agreed-to scope of coverage this way: “All Risks of Direct Physical Loss or Damage excluding Flood, Earthquake and Named Windstorm.” Doc. 25-9 at 2.
Soon after, Aspen issued the 2016 policy. The cover page described the 2016 policy as a “renewal of” its 2015 predecessor. Doc. 25-10 at 1. But the two policies’ terms differed in material respects. For one thing, the 2016 policy was about $10,000 cheaper per year than the amended 2015 policy. Far more significantly here, the 2016 policy contained no exclusion for losses caused by named windstorms. It contained a detailed catalogue of other exclusions—for instance, for damage resulting from “Ordinance Or Law,” “Earth Movement,” “Governmental Action,” “Nuclear Hazard,” “Utility Services,” “War And Military Action,” and “ ‘Fungus,’ Wet Rot, Dry Rot And Bacteria”—but a “Named Windstorms” exclusion was conspicuously absent. Doc. 25-10 at 42–44 § B.
You know what happened next. In October 2016, a named windstorm—Hurricane Matthew—blew through Melbourne, ripping the roof off Shiloh's building. Rainwater poured in, aggravating the damage. Shiloh filed a claim for, in its words, “Water Damage from Roof hurricane Matthew.” Doc. 25-15. Aspen denied the claim on several grounds, including, as relevant here, that Shiloh's policy excluded coverage for losses caused by named windstorms. See Doc. 25-16 at 2.
The following year was basically a carbon copy. In early 2017, Shiloh commenced efforts to renew its policy. As in 2016, Aspen provided a quote, reminding Shiloh that the policy would exclude coverage for damage resulting from “Named Windstorms.” Doc. 25-13 at 2. As in 2016, Shiloh applied for the policy, scribbling “EX wind” into the application's “forms and conditions to apply” sections for certain buildings, Doc. 25-11 at 3–4, 7, and Aspen issued a binder reflecting the named-windstorm exclusion, Doc. 25-12. As in 2016, Aspen then formally issued a policy that described itself as a “renewal” of the 2016 policy, Doc. 25-14 at 1, but, again, whose “Exclusions” provision, while expressly carving out losses resulting from all manner of contingencies, said nothing about named windstorms, id. at 43–44.
Like clockwork, in September 2017, a named windstorm—Hurricane Irma—blew through town and, you guessed it, tore the roof off of Shiloh's building. Just as it had in Hurricane Matthew, water poured in, exacerbating the damage. Shiloh filed another claim listing the “cause of loss”—again, in its words—as “Hurricane Irma.” Doc. 25-18 at 1. And just as it had done a year earlier, Aspen denied Shiloh's claim on several grounds, among them that its policy excluded losses caused by named windstorms.
Shiloh sued Aspen for breach of contract and sought a declaration that its 2016 and 2017 policies—which we'll call the Matthew and Irma Policies—covered damages caused by named windstorms. The parties cross-moved for summary judgment, teeing up a discrete and dispositive question of law: Do the policies cover named-windstorm-related losses?
The district court granted summary judgment to Aspen. It held that “no reasonable jury” could find that the parties “intended the policies at issue to exclude named windstorm coverage.” Doc. 41 at 13 (emphasis omitted). The court acknowledged that “[t]he two policies in effect when [Shiloh's] building incurred damage do not, alone, say anything explicit concerning damage resulting from a named windstorm.” Id. at 9. But it found Aspen's evidence regarding the parties’ intent overwhelming: “[T]he explicit bargaining to remove named windstorm coverage, the reduced premiums that resulted from that bargaining, and the explicit language in the subsequent policy quotes” all proved to the district court's satisfaction that “named windstorm coverage would not be included.” Id. at 13.
This is Shiloh's appeal. Shiloh contends, as it did below, that the Matthew and Irma Policies cover losses caused by named windstorms, and it asks us to “reverse the district court's order granting summary judgment” and to “remand[ ] with instructions that [the case] be submitted to a jury.” Br. of Appellant at 25.
We reverse. For reasons we will explain, we hold that both policies cover named windstorms. The Irma Policy unambiguously covers them, and the Matthew Policy, although ambiguous, covers them by dint of the traditional contra proferentem canon of insurance-contract interpretation.
The general rules governing the interpretation of insurance policies under Florida law are clear.2 The cardinal principle is that a policy's text is paramount: “Florida courts start with ‘the plain language of the policy, as bargained for by the parties.’ ” State Farm Fire & Cas. Co. v. Steinberg, 393 F.3d 1226, 1230 (11th Cir. 2004) (citing Auto-Owners Ins. Co. v. Anderson, 756 So. 2d 29, 34 (Fla. 2000)). In particular, “[i]n insurance coverage cases under Florida law, courts look at the insurance policy as a whole and give every provision its full meaning and operative effect.” Id. (quotations omitted). To be sure, Florida law permits reviewing courts to venture outside the policy's four corners in limited circumstances—to consider, for instance, whether an insured's “application” should be understood to “amplif[y], extend[ ], or modif[y]” the policy. Fla. Stat. § 627.419(1). Florida law is clear, though, that in the event of a conflict between the policy and the underlying application, the policy controls. See Mathews v. Ranger Ins. Co., 281 So. 2d 345, 349 (Fla. 1973) (“[T]he general rule” is that “the provisions of the policy [ ] govern where conflict exists between the provisions of the application and the policy.”) (interpreting § 627.419(1)).
Beyond those basics, Florida law prescribes more particular rules for the interpretation of ambiguous and unambiguous insurance policies. The rule applicable to unambiguous policies is ruthlessly straightforward: If the policy's “language is unambiguous, it governs”—end of story. State Farm Fire, 393 F.3d at 1230; accord Allstate Ins. Co. v. Orthopedic Specialists, 212 So. 3d 973, 975–76 (Fla. 2017) (“Where the language in an insurance contract is plain and unambiguous, a court must interpret the policy in accordance with the plain meaning so as to give effect to the policy as written.”). Importantly, that is true even where extrinsic evidence contradicts the policy's terms. See Vencor Hosps. v. Blue Cross Blue Shield of R.I., 284 F.3d 1174, 1179 (11th Cir. 2002) (“It is well established under Florida law that parol evidence is inadmissible to vary or contradict the clear and unambiguous language of a contract.”). And it is “especially true when the contract contains an integration clause indicating that the parties intended the written agreement to be the entire agreement.” In re Yates Dev., Inc., 256 F.3d 1285, 1290 (11th Cir. 2001) (applying Florida law).
When confronted with an insurance policy that is facially ambiguous, Florida courts apply the familiar contra proferentem canon. Pursuant to that interpretive rule, “any ambiguity which remains after reading each policy as a whole and endeavoring to give every provision its full meaning and operative effect must be liberally construed in favor of coverage and strictly against the insurer.” Gov't Emps. Ins. Co. v. Macedo, 228 So. 3d 1111, 1113 (Fla. 2017) (quoting Wash. Nat'l Ins. Corp. v. Ruderman, 117 So. 3d 943, 949–50 (Fla. 2013) (plurality opinion)).
Importantly here, the Florida Supreme Court has clarified that facial ambiguities in insurance contracts should be resolved by reference to contra proferentem rather than extrinsic evidence of the parties’ supposed “intent.” And in fact, it did so in response to a question that we certified to it. In Ruderman ex rel. Schwartz v. Washington National Insurance Corp., 671 F.3d 1208 (11th Cir. 2012), we confronted (1) a Florida insurance policy that was ambiguous on its face and (2) an apparent split among Florida courts about how to resolve the ambiguity. One line of decisions indicated that “[a]mbiguous policy provisions are interpreted liberally in favor of the insured and strictly against the drafter who prepared the policy”; but another suggested that courts should “look[ ] to extrinsic evidence to resolve the ambiguity before construing any remaining ambiguity against the drafter of the policy.” Id. at 1211 (quotations omitted) (emphasis in original). Uncertain how to proceed, we certified several questions to the Florida Supreme Court, one of which was whether, “[i]f an ambiguity exists in this insurance policy,” a court should “first attempt to resolve the ambiguity by examining available extrinsic evidence.” Id. at 1212. In what we called a “definite response[ ],” the Florida Supreme Court “advised us that the answer” is “no”—contra proferentem controls. Ruderman ex rel. Schwartz v. Wash. Nat'l Ins. Corp., 731 F.3d 1188, 1189 (11th Cir. 2013) (citing Ruderman, 117 So. 3d at 949–50).
To the extent that the Florida Supreme Court's Ruderman decision left any doubt, its follow-on decision in Macedo resolved it. The question there was whether an insurance policy provision covering “legal expenses and court costs” included attorneys’ fees. The court concluded that the policy's language “create[d] an ambiguity”—in particular, it said, because “[r]eferring to ‘legal expenses’ in conjunction with ‘court costs’ signifies that there are ‘legal expenses’ aside from costs” that might (or might not) include attorneys’ fees. Macedo, 228 So. 3d at 1114. As already noted, for the governing interpretive principle the court quoted its earlier decision in Ruderman: “[A]ny ambiguity which remains after reading each policy as a whole and endeavoring to give every provision its full meaning and operative effect must be liberally construed in favor of coverage and against the insurer.” Id. at 1113 (quotations omitted). The court then applied that rule matter-of-factly and without further elaboration or investigation, let alone consideration of any extrinsic evidence: “[B]ecause there are multiple reasonable interpretations regarding whether attorneys’ fees are included by the terms ‘expenses’ and ‘costs,’ ” the policy “is ambiguous and must be construed in favor of coverage.” Id. at 1114 (citation omitted).
Together, Ruderman and Macedo make clear that when confronted with a facially ambiguous insurance policy, a reviewing court should simply apply the well-worn contra proferentem rule and resolve the ambiguities in favor of coverage and against the insurer. It shouldn't plumb the depths for evidence of the parties’ supposed intent.3
We now apply these principles to the two policies. First, the Irma Policy. Whatever the extrinsic evidence may suggest about the parties’ intentions or expectations, the Irma Policy unambiguously covers named windstorms. The policy opens with a provision that broadly “cover[s]” “risks of direct physical loss.” Doc. 25-14 at 43 § A. There follows a long and extremely detailed list of exclusions—set out beneath a bolded “Exclusions” header. Id. at 43 § B. That list includes exclusions for losses caused by “Ordinance Or Law,” “Earth Movement,” “Governmental Action,” “Nuclear Hazard,” “Utility Services,” “War And Military Action,” “Water” (which, to be clear, refers to floods, mudslides, etc.), and “ ‘Fungus,’ Wet Rot, Dry Rot And Bacteria.” The list conspicuously does not include “Named Windstorms,” either as a defined category of claim or in any other way, shape, form, or fashion.
Two interpretive principles confirm that the Irma Policy doesn't exclude, and therefore covers, damage caused by named windstorms. The first is the usual rule that “insurance coverage must be interpreted broadly and its exclusions narrowly.” Westchester Gen. Hosp., Inc. v. Evanston Ins. Co., 48 F.4th 1298, 1302 (11th Cir. 2022) (quotations and brackets omitted). And the second is the rule of expressio unius est exclusio alterius—i.e., that “[t]he expression of one thing implies the exclusion of others.” Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 107 (2012); see also Young v. Progressive Se. Ins. Co., 753 So. 2d 80, 85 (Fla. 2000) (using this “principle of statutory construction” to show that “[b]y failing to permit self-insured motorist policy exclusions in the list of authorized exclusions, the Legislature has ․ indicated its intent ․ not to permit self-insured motorist policy exclusions”). Here, the expressio unius canon applies with particular force because the Irma Policy's catalogue of exclusions is so detailed. See Scalia & Garner, supra, at 108 (“The more specific the enumeration, the greater the force of the canon.”).
On its face, then, the Irma Policy clearly doesn't exclude—and thus covers—losses resulting from named windstorms. We can envision only two possible responses to the policy's plain language. First, the district court pointed to both policies’ “renewal of” provisions, saying that they formed a “continuous chain” stretching back to the pre-Matthew policy, which, as amended in mid-2015, expressly excluded damages caused by named windstorms. But policies in this kind of “chain” don't invariably duplicate one other's terms. And in fact, the chained-up policies here diverge from the pre-Matthew policy in at least one material way, in that they impose different premiums. Moreover, and in any event, “[t]he general rule” in Florida is that “each renewal of an insurance policy” creates “an entirely new and independent contract of insurance.” Marchesano v. Nationwide Prop. & Cas. Ins. Co., 506 So. 2d 410, 413 (Fla. 1987) (emphasis added).4
Nor, anticipating a second response, are we persuaded that Shiloh “amplified” or “modified” the Irma Policy within the meaning of Fla. Stat. § 627.419(1) by scribbling “EX wind” on its application. Two reasons: As an initial matter, it's not self-evident that “EX wind” meant “exclusion for named windstorms”—maybe, but not necessarily. Moreover, even if it did, it wouldn't matter because, as already explained, Florida law provides that “the policy ․ govern[s] where conflict exists between the provisions of the application and the policy.” Mathews, 281 So. 2d at 349 (interpreting § 627.419(1)). And again, that rule applies with even greater force where, as here, the policy “contains an integration clause indicating that the parties intended the written agreement to be the entire agreement.” In re Yates Dev., Inc., 256 F.3d at 1290.
Shiloh concedes that, unlike the Irma Policy, the Matthew Policy is ambiguous on its face. See Oral Arg. at 5:20–5:35. We're not so sure about that, but we'll accept Shiloh's concession. The supposed ambiguity results from tension between two of the policy's provisions. On the one hand, like the Irma Policy, the Matthew Policy contains a broad coverage clause, and a detailed “Exclusion” provision that includes all manner of specific exclusions but, conspicuously, does not mention “Named Windstorms.” Doc. 25-10 at 42 §§ A–B. That juxtaposition, for reasons we have explained, strongly indicates that the Matthew Policy covers damage caused by named windstorms. See supra at 12–13. On the other hand, unlike the Irma Policy, the Matthew Policy's deductible provision specifically mentions “Named Windstorm[s]”: “DEDUCTIBLE: $5,000 Per Occurrence, except; $25,000 Per Occurrence as respects Wind and/or Hail (excluding Named Windstorm).” Doc. 25-10 at 9. On one reading, that provision's closing parenthetical could be understood to “exclud[e]” coverage for “Named Windstorm[s].” Of course, the deductible provision could also be read, in context, simply to create a “Named Windstorm” “exclu[sion]” to the $25,000 “except[ion]”—meaning only that named windstorms are subject to the usual $5,000 deductible. But again, we'll accept Shiloh's concession that the Matthew Policy's deductible provision creates a facial ambiguity.
As already explained in detail, Florida law is clear that when an insurance policy is facially ambiguous, the ambiguity is resolved in favor of coverage and against the insurer, without regard to extrinsic evidence of the parties’ supposed intentions or expectations. That's the rule of Ruderman and Macedo: “[A]ny ambiguity which remains after reading each policy as a whole and endeavoring to give every provision its full meaning and operative effect must be liberally construed in favor of coverage and strictly against the insurer.” Macedo, 228 So. 3d at 1113 (quoting Ruderman, 117 So. 3d at 949–50). Because (pursuant to Shiloh's concession) “there are multiple reasonable interpretations” regarding whether the Matthew Policy covers named windstorms, the policy “is ambiguous and must be construed in favor of coverage.” Id. at 1114. Accordingly, the Matthew Policy, like the Irma Policy, covers damage that results from named windstorms.
For all these reasons, we hold as follows:
1. Whatever the evidence of the contracting parties’ subjective intentions and expectations, the Irma Policy's plain language unambiguously covers losses caused by named windstorms.
2. Although potentially ambiguous, the Matthew Policy likewise—and, again, whatever the evidence of the parties’ subjective intentions and expectations—covers losses caused by named windstorms pursuant to the contra proferentem canon, according to which ambiguous insurance contracts are construed in favor of coverage and against the insurer.
REVERSED and REMANDED.5
1. We review the district court's grant of summary judgment de novo, viewing all facts and drawing all reasonable inferences in the light most favorable to Shiloh as the nonmoving party. See Hinkle v. Midland Credit Mgmt., Inc., 827 F.3d 1295, 1300 (11th Cir. 2016).
2. Because federal jurisdiction over this matter is based on diversity of the parties’ citizenship, Florida law governs the determination of the issues on this appeal. See State Farm Fire & Cas. Co. v. Steinberg, 393 F.3d 1226, 1230 (11th Cir. 2004).
3. We should briefly tie up one loose end. In some contexts, Florida courts have distinguished between contracts that exhibit “patent” ambiguities—i.e., those that appear on the face of the instrument—and those that contain only “latent” ambiguities—i.e., those that surface only after considering extrinsic circumstances. In those contexts, courts have permitted the consideration of parol evidence to clarify latent ambiguities but not patent ones. See, e.g., Mac-Gray Servs., Inc. v. Savannah Assocs. of Sarasota, LLC, 915 So. 2d 657, 659 (Fla. 2d Dist. Ct. App. 2005). For two reasons, though, those decisions don't affect our analysis here. First, the sole ambiguity that we find here—as we'll explain, in the Matthew Policy—is patent, not latent. Second, it may well be that the patent-latent distinction matters only for “contracts other than contracts of insurance.” Ruderman, 117 So. 3d at 950 n.3.
4. Before us, Aspen cites another passage of Marchesano for the proposition that “the parties to the renewal of an insurance contract are ‘entitled to assume that the terms of the renewed policy are the same as those of the original contract.’ ” Br. of Appellee at 20 (quoting Marchesano, 506 So. 2d at 413) (emphasis added). But that is not what Marchesano says; rather, it says that “the insured is entitled to assume that the terms of the renewed policy are the same as those of the original contract.” 506 So. 2d at 413 (emphasis added). The insurer, it seems to us, isn't entitled to make the same (if any) assumptions about the terms of a policy that it wrote.
5. In light of our analysis of the policies, Shiloh may well be entitled to summary judgment. For whatever reason, Shiloh hasn't asked us to reverse the district court's decision and render judgment in its favor; rather, it has asked us only to reverse and remand the case “with instructions that it be submitted to a jury.” Br. of Appellant at 25. We will remand the case, but not specifically with instructions that it be submitted to a jury. We leave it to the district court to decide how best to proceed.
Newsom, Circuit Judge:
Was this helpful?
Get help with your legal needs
FindLaw’s Learn About the Law features thousands of informational articles to help you understand your options. And if you’re ready to hire an attorney, find one in your area who can help.
Search our directory by legal issue
Enter information in one or both fields (Required)