Harry SANDERS, Executor of the Estate of Nancy A. Andersen and Assignee of John Doe, Plaintiff, Appellant, v. The PHOENIX INSURANCE COMPANY and The Travelers Indemnity Company of America, Defendants, Appellees, 843 F.3d 37

SELYA, Circuit Judge. This case begins with a tragic tale of unrequited love and morphs into a series of imaginative questions regarding the coverage available under a standard form homeowner’s insurance policy. But when imagination runs headlong into settled legal precedent, imagination loses. Recognizing as much, the court below dismissed the complaint. After careful consideration, we affirm. I.

BACKGROUND This diversity suit arises from the refusal of The Phoenix Insurance Company to defend and/or indemnify its named insured, an attorney whom we (like the court below) shall call “John Doe,” against claims advanced by Harry Sanders, suing in his capacities as executor of the estate of Nancy A. Andersen (his deceased spouse) and as Doe’s assignee. Inasmuch as the district court dismissed Sanders’s complaint for failure to state a claim upon which relief could be granted, we take as true the raw facts as alleged in the complaint. See SEC v. Tambone, 597 F.3d 436, 441-42 (1st Cir. 2010) (en banc).

Doe met Andersen in January of 2011 when she sought legal representation in possible divorce proceedings against Sanders. Roughly four months later, Doe initiated divorce proceedings on Andersen’s behalf. During this interval, Doe learned that Andersen suffered from severe depression and anxiety, had been prescribed several anti-depressant and anti-anxiety medications, and had recently attempted suicide. Nevertheless, Doe and Andersen began an on-again/off-again intimate relationship. Despite his personal involvement, Doe did not withdraw as her counsel. The relationship did not go smoothly.

As the fall of 2011 approached, Doe’s ardor cooled and he became progressively distant. Correspondingly, Andersen’s anxiety increased. Matters came to a head when, on or around October 1, 2011, Doe promised to join Andersen at her apartment. He did not do so. Distraught, Andersen wrote a suicide note lamenting Doe’s inconstancy and proceeded to drink herself to death. Doe tried unsuccessfully to contact Andersen by telephone the next day. When he could not reach her, he went to her apartment and discovered her body. Sanders was appointed as executor of Andersen’s estate.

Slightly over a year after Andersen’s death, he sent Doe a demand letter pursuant to Massachusetts General Laws

Chapter 93A,

Section 9. Cognizant that some of his meretricious interludes had occurred at his home, Doe promptly notified his homeowner’s insurance carrier (Phoenix). After looking into the matter, Phoenix denied coverage on two grounds: that Andersen’s death was not an “occurrence” covered under Doe’s homeowner’s policy (the Policy) and that the Policy’s professional services exclusion barred coverage. In a letter dated September 19, 2013, Sanders notified Phoenix that he and Doe planned to mediate their dispute and invited Phoenix to participate. Phoenix declined the invitation.

About three weeks later, Doe sought to have Phoenix reconsider its denial of coverage, informing it that Sanders was advancing a claim for negligent infliction of emotional distress; Unmoved, Phoenix reiterated its denial of coverage. Eventually, Doe, Doe’s law firm, and Sanders reached an accord: the law firm’s insurers agreed to pay Sanders $500,000 in exchange for a release of all claims against the firm.

Ancillary to the settlement, Doe agreed that his personal- liability to Sanders amounted to an additional $500,000 and assigned to Sanders all of Doe’s rights and interests under' the Policy vis-á-vis Andersen’s death and any claims that he might have against Phoenix as a result of its failure to defend and/or indemnify him. Sanders followed up by sending a

Chapter 93A demand letter to Phoenix, see supra note 2, accusing it of unfair settlement practices and demanding $500,000 (the limit of liability under the Policy). Phoenix refused the demand. Sanders repaired to a Massachusetts state court and filed this suit. Citing diversity of citizenship and the existence of a controversy in the requisite amount, Phoenix removed the case to the federal district court. See 28 U.S.C. §§ 1332(a), 1441. Phoenix then moved to dismiss. See Fed. R. Civ. P. 12(b)(6). The district court referred the motion to a magistrate judge, who recommended granting it.

Sanders objected, and the district court, undertaking de novo review, overruled his objections and dismissed the action. This timely appeal ensued. II. ANALYSIS We review a district court’s dismissal of a complaint for failure to state a claim de novo. See Artuso v. Vertex Pharm., Inc., 637 F.3d 1, 5 (1st Cir. 2011). Accepting as true all well-pleaded facts contained in the complaint, we are constrained to draw all reasonable inferences in the pleader’s favor. See id. Where relevant, we may supplement the pleaded facts with “documentation incorporated by reference-in the complaint.” Rivera-Díaz v.

Humana Ins, of P.B., Inc., 748 F.3d 387, 388 (1st Cir. 2014); see Hidalgo-Vélez v. San Juan Asset Mgmt., Inc., 758 F.3d 98, 101-02 (1st Cir. 2014). Because this case is brought in diversity jurisdiction, we must look to state law for the substantive rules of decision. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). The parties agree that Massachusetts law applies, and we readily embrace that sensible agreement. See Artuso, 637 F.3d at 5 (“In determining which state’s law applies, a diversity court is free to honor the parties’ reasonable agreement.”).

If we are unable to discern any controlling Massachusetts authority on a particular point, we must make an “informed prophecy” as to how the state’s highest court — the Supreme Judicial Court (SJC) — would rule if faced with the issue. Ambrose v. New Eng. Ass’n of Schs. & Colls., Inc., 252 F.3d 488, 498 (1st Cir. 2001). Our prediction may be “guided, inter alia, by persuasive case law from other jurisdictions and relevant public policy considerations.” Id. A, Alleged Breach of Duty to Defend.

We start with Sanders’s principal remonstrance (asserted in his capacity as Doe’s assignee): that Phoenix forsook its duty to defend Doe against the claims advanced by Sanders. In Massachusetts, the duty to defend under- an insurance policy arises “when the allegations in a complaint are reasonably susceptible of an

interpretation that states or -roughly sketches a claim covered by the policy terms.” Billings v. Commerce Ins. Co., 458 Mass. 194, 936 N.E.2d 408, 414 (2010). When determining whether an insurer has a duty to defend, an inquiring court must consider “the facts alleged in the complaint, and [any] facts known or readily knowable by the insurer that- may aid in its

interpretation of the allegations in the complaint.” Id. The precise scope of an insurer’s duty to defend is defined by the .insurance policy itself, to which we apply familiar rules of contract

interpretation. See B & T Masonry Constr. Co. v. Pub. Serv. Mut. Ins. Co., 382 F.3d 36, 39 (1st Cir. 2004); Bos. Gas Co. v. Century Indem. Co., 454 Mass. 337, 910 N.E.2d 290, 304 (2009). We interpret the words of the policy in light of their plain meaning, considering the document as a whole. See B & T Masonry, 382 F.3d at 39; Golchin v. Liberty Mut. Ins. Co., 466 Mass., 156, 993 N.E.2d 684, 688 (2013). Our construction must be consistent with what an objectively reasonable insured, reading the relevant policy language, would expect the words to mean. See Hazen Paper Co. v. U.S. Fid. & Guar.

Co., 407 Mass. 689, 555 N.E.2d 576, 583 (1990). If this analysis yields two reasonable (but conflicting)

interpretations of the-policy’s text, the insured must be given the benefit of the

interpretation that redounds to his benefit. See id. In the case at hand, the Policy (to which we add our own emphasis and remove original emphasis) states in pertinent part: If a claim is made or a suit is brought against any insured for damages because of bodily injury or property damage caused by an occurrence to which this coverage applies, even if the claim or suit is false, we will: In Phoenix’s view, the Policy therefore provides that it must only furnish counsel to defend the insured in the face of a suit but may investigate and settle a claim.

Thus, it has no obligation to provide a defense in the absence of a suit, Sanders demurs. He points to the Policy’s other references to claims or suits and asseverates that the Policy obligates the insurer to defend broadly against pre-suit claims. This asseveration lacks force. The majority of the references that Sanders identifies come from the Policy’s credit card, fund transfer card, forgery, and counterfeit money provision, which covers losses of up to $1,000 incurred due to unauthorized use of the insured’s credit cards or similar forms of financial misfeasance.

The particular subsection most loudly bruited by Sanders states: Defense: a. ... OUR OBLIGATION TO DEFEND ANY CLAIM OR SUIT ENDS WHEN THE AMOUNT WE PAY FOR THE LOSS EQUALS OUR LIMIT OF LIABILITY. b. If a claim is made or a suit is" brought "against any insured for liability under the Credit Card or Fund Transfer Card coverage, we will provide a defense at our expense by counsel of our choice.

The location of these statements within the credit card provision, coupled with the fact that the claims asserted in this case in no way implicate that coverage, persuasively indicates that the language cannot reasonably be read to support Sanders’s broader argument concerning the scope of the basic liability coverage afforded by the Policy. While an insurance policy must be read as a whole, see Golchin, 993 N.E,2d at 688, that prescription does not give a party license to transplant randomly words from one provision into the inhospitable soil of an entirely different provision.

Sanders’ next argument is no more convincing. He notes that the Policy states that Phoenix will cover “reasonable expenses incurred by any insured at [its] request ... for assisting [it] in the investigation or defense of any claim or suit.” Using this clause as a springboard, he contends that Phoenix has undertaken a duty to defend claims as well as suits. This glib reading of the quoted language does not withstand scrutiny.

Giving the language its natural meaning, it denotes no more than that — if the company does in-véstigate either a claim or a suiN-it will reimburse any reasonable expenses incurred by the insured. An objectively reasonable insured, reading this language, would surely come to this conclusion. Of course, the general rule that there is no duty to defend before the filing of a suit is not ironclad.

For example, the SJC has held that a Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) notice letter from the federal Environmental Protection Agency (EPA) to a potentially responsible party is sufficiently analogous to a suit as to trigger a duty to defend. See Hazen Paper, 555 N.E,2d at 580-81. The CERCLA letter notified Hazen Paper that it could be held responsible for the release of hazardous substances at a particular waste facility. See id. at 578. Hazen Paper asked its insurer to defend it in the CERCLA proceeding, but the insurer declined.

The SJC held that the duty to defend had been triggered, reasoning that the litigation defense insurance purchased by Hazen Paper would be “substantially compromised” if the insurer’s duty to defend was not activated by the CERCLA letter. Id. at 580. After all, the CERCLA letter itself had severe consequences: without a response, the EPA could proceed unilaterally with its administrative action, and subsequent judicial review of the EPA’s final order would be limited to the administrative record created before the EPA. See id. at 581 (citing 42 U.S.C. § 9613(j)(l)).

Moreover, the EPA’s final order would be subject to review only under an agency-friendly standard. See id. (citing 42 U.S.C. § 9613(j)(2)). In addition, merely failing to provide requested information to the EPA would have exposed Hazen Paper to monetary penalties, (independent of its potential responsibility for the hazardous substances). See id. at 580; 42 U.S.C. § 9604(e). This administrative structure, the SJC, said, left the insured with “no practical choice other than to respond actively to the letter.” Hazen Paper, 555 N.E.2d at 581-82.

To characterize such a response as “voluntary” would be “naive.” Id. ,In the last analysis, though, the SJC’s decision in Hazen Paper was quite narrow and case-specific. Among other things, the SJC took great pains to distinguish the CERCLA letter from a “conventional demand letter based on a personal injury claim.” Id. at 581. Sanders argues that this case comes within the confines of the Hazen Paper exception because of his

Chapter 93A letter to Doe. In support, he notes that failure either to respond to a

Chapter 93A letter or to make a reasonable settlement offer can expose the insured to multiple damages, attorneys’ fees, and costs. See Mass. Gen. Laws ch. 93A, § 9(3)-(4). That is true as far as it goes — but it does not take Sanders very far. A

Chapter 93A demand letter is simply not a fair congener to the CERCLA letter discussed in Hazen Paper; rather, it is more like a “conventional demand letter based on a personal injury claim” — a type of communication that the SJC said was insufficient to trigger the duty to defend. Hazen Paper, 555 N.E.2d at 581. In Hazen Paper, the failure to participate in the administrative process would have all but forfeited the insured’s case. Ignoring a

Chapter 93A demand letter, however, would not “substantially compromise[ ]” an insured’s position: a failure to respond produces a much more limited effect. Id. at 580. Significantly, Doe’s underlying liability could not have been affected by the

Chapter 93A letter — and his potential exposure to multiple" damages, attorneys’ fees, and costs would only"'come to fruition if a court established that underlying liability. See Mass. Gen. Laws ch. 93A, § 9(3X4). The short of it is that

Chapter 93A demand letters are fairly comparable to demand letters sent in anticipation of garden-variety personal injury litigation. Given the frequency with which they are used and the important differences that distinguish them from CERCLA letters, we are reluctant to widen the narrow boundaries sketched by the Hazen Paper court and hold that

Chapter 93A demand letters are a functional equivalent of a suit. In our view, such restraint is particularly appropriate given our status as a federal court predicting state law. Cf. Katz v. Pershing, LLC, 672 F.3d 64, 73-74 (1st Cir. 2012) (declining to recognize novel exception to “bright-line rule” of state law because “federal diversity courts are charged with ascertaining state law, not with reshaping it”). If more were needed — and we do not think that it is — there is substantial reason to doubt, based on the facts of this case, whether a failure to respond to "the

Chapter 93A letter would have brought about the consequences that Sanders gloomily predicts.

Chapter 93A applies to “acts or practices in the conduct of any trade or commerce.”. Mass. Gen. Laws ch. 93A, § 2(a). Here, however, Sanders alleges that Doe’s liability arises from personal— rather than professional — misconduct (presumably because the Policy excludes coverage for liability originating from professional services rendered by the insured); Sanders also submits that the mediation in which he and Doe participated was the functional equivalent of a suit and, thus, triggered the duty to defend. For this proposition, he relies on the decision in Selective Insurance Co. v.

Cherrytree Cos., 376 Ill.Dec. 159, 998 N.E.2d 701 (2013). There, the insurer denied coverage for a disputed claim, and the insured entered into a settlement agreement’with the claimant without suit having been filed. See id., 376 Ill.Dec. 159, 998 N.E.2d at 702-03. The court held that the absence of a lawsuit did not insulate the insurer from liability, explaining that “the indemnification provision in the policy ... did not require the filing of a ‘suit.’” Id. 376 Ill.Dec. 159, 998 N.E.2d at 709-10. Sanders’s relianee on Cherrytree is triply misplaced.

First, Cherrytree is an Illinois case, never adopted by the Massachusetts courts. Second, it is black-letter law that the policy language determines the scope of coverage, see B & T Masonry, 382 F.3d at 39, and the Policy on which Sanders sues contains no provision allowing indemnification without suit. Third, the Cherrytree decision did not in any way implicate the duty to defend. See Wesco Ins. Co. v. Regas, No. 14 C 716, 2015 WL 500702, at *6 (N.D. Ill. Feb, 3, 2015) (observing that, in Cherrytree, there was “no determination of whether the insurer had á duty to defend”).

Given these salient distinctions, the Cherrytree decision fails to persuade us that the SJC would conclude that the mediation conducted in this case was the functional equivalent of a suit. Bereft of any support in the case law, Sanders’s suggestion that mediation, in the circumstances of this case, should be regarded as the functional equivalent of a suit strikes us as patently unreasonable. The mediation to which he adverts .was a less formal, more ad hoc proceeding, and Doe’s involvement in it was completely voluntary.

So viewed, the mediation was markedly different from a lawsuit, which operates under established procedural rules and in which a defendant must participate to protect his interests. In the face of a lawsuit, mounting a defense is a necessity; in the face of a proposal to mediate, opting to participate is a strategic choice. In.an effort to catch lightning in a bottle, Sanders emphasizes that Doe’s desire for anonymity created an added pressure to settle before Sanders filed suit (thus making participation in mediation more imperative for fear that Doe’s name would be exposed in public court documents).

This stated concern turns a blind eye to familiar procedural protections. Sanders could have filed a motion to seal the case along with his complaint or, if he did not, Doe himself could have moved to seal. See D. Mass. R. 7.2; New Eng. Internet Café, LLC v. Clerk of Superior Ct., 462 Mass. 76, 966 N.E.2d 797, 803 (2012). In the alternative, Doe could have sought— whether by agreement or by court order— to have his identity safeguarded through the use of a pseudonym (which is precisely what transpired here).

To sum up, the Policy, fairly read, draws a clear distinction between the duty to defend (which applies to suits alone) and the right to investigate (which applies to both suits and claims). Giving force to this clear distinction, we hold — as did the court below — that Phoenix’s duty to defend was never triggered (and, thus, never breached) in the circumstances of this case. B. Other Theories of Liability. Sanders raises a gallimaufry of other theories of liability. To begin, he asserts that, even in the absence of a duty to defend, Phoenix violated its separate duty to indemnify.

In his view, the two duties are separate and distinct: the duty-to-defend analysis considers all preliminary allegations, whereas the duty-to-indemnify analysis examines the factual merits of the case, Phoenix counters that, under Massachusetts law, there can be no duty to indemnify if there is no duty to defend. As a general matter, the duty to defend is broader than the duty to indemnify. See Bagley v, Monticello Ins. Co., 430 Mass. 454, 720 N.E.2d 813, 817 (1999).

The SJC has made pellucid that “[i]f an insurer has no duty to defend, based on the allegations in the plaintiffs complaint, it necessarily follows that the insurer does not have a duty to indemnify.” Id. In other words, the broader duty (the duty to defend) swallows up the narrower duty (the duty to indemnify). Sanders resists this formulation. He lauds decisions from other jurisdictions in an effort to cast the SJC’s formulation into doubt. See, e.g., Grinnell Mut. Reins. Co. v.

Reinke, 43 F.3d 1152, 1154 (7th Cir. 1995) (suggesting that “because of the possibility that the legal theory of the underlying suit may change, a conclusion that the insurer need not defend does not [necessarily] imply that it need not indemnify”)- As federal judges sitting in diversity jurisdiction, though, we are not free to pick and choose which state’s jurisprudence is the most sound. Rather, we are duty-bound to accept controlling state law where it can be discerned. See Kassel v. Gannett Co., 875 F.2d 935, 949-50 (1st Cir. 1989).

On this issue, then, it is incumbent upon us to accept the clear statement of Massachusetts law articulated by the SJC. Here, moreover, the Policy affords us an independently sufficient reason to hold that Phoenix does not have a duty to indemnify.

The Policy states that “no action with respect to [the insured’s personal liability coverage] can be brought against [Phoenix] until the obligation of the insured has been determined by a final judgment or [an] agreement signed by-[Phoenix].” In this instance, there is neither a final judgment nor a settlement agreement executed by Phoenix; there is only a settlement negotiated between Sanders and Doe, in Phoenix’s absence. Consequently, no action lies against Phoenix for indemnification under the terms of the Policy.

Sanders asks us to overlook this Policy language and focus instead on the Policy’s general indemnification provision. That provision states that Phoenix will “[p]ay up to [its] limit of liability for the damages for which the insured is legally liable.” Sanders contends that the phrase “legally liable” does not logically require that the insured’s obligation be reflected in a final judgment. . This contention ignores the more specific provision cited by Phoenix.

It likewise ignores the logical import of that provision: that an insured’s liability will either be adjudicated by a court of law or agreed to by the insurer. Massachusetts law teaches that an inquiring-court normally should give a more specific policy provision priority over a more general policy provision, see Certain Interested Underwriters at Lloyd’s, London v. Stolberg, 680 F.3d 61, 67 (1st Cir. 2012), and Sanders has not offered a plausible rationale for abandoning that tenet here. This brings us to Sanders’s common-law claims sounding in tort and breach of contract. Those claims need not detain us.

The complaint predicates them largely on Phoenix’s failure to defend or indemnify Doe. But here — as discussed above — Phoenix did not have a duty either to defend or to indemnify in the circumstances at hand. The one remaining claim hinges on Sanders’s allegation that Phoenix is guilty of unfair and deceptive trade practices in violation of Massachusetts General Laws

Chapter 176D. Specifically, he alleges that Phoenix’s liability rests on its refusal to effectuate a prompt, fair, and equitable settlement. This claim fails -for two reasons. For one.thing, Phoenix has no duty to settle absent a duty either to defend or to indemnify. See Transam. Ins. Co. v. KMS Patriots, L.P., 52 Mass.App.Ct. 189, 752 N.E.2d 777, 783 (2001) (explaining that “[w]hen coverage has been correctly denied ... no violation of the Massachusetts statutes proscribing unfair or deceptive trade practices may be found”).

For another thing, Massachusetts law only requires an insurer to effectuate settlement once “liability has become reasonably clear.” Mass. Gen. Laws ch. 176D, § 3(9)(f); see Clegg v. Butler, 424 Mass. 413, 676 N.E.2d 1134, 1140 (1997). Doe’s liability to Sanders — as a private individual, not an attorney — was far from reasonably clear. Sanders has not identified a case in which Massachusetts (or any other jurisdiction, for that matter) has held an individual’s romantic partner responsible for the individual’s suicide. In lieu of case law, Sanders relies on the Restatement (First) of Torts

section 325, which provides: One who gratuitously undertakes with another to do

an act or to render services which he should recognize as necessary to the other’s bodily safety and thereby leads the other in reasonable reliance upon the performance of such undertaking (

a) to refrain from himself taking the necessary steps to secure his safety or from securing the then available protective action by third persons, or (

b) to enter upon a course of conduct which is dangerous unless the undertaking is carried out, is subject to liability to the other for bodily harm resulting from the actor’s failure to exercise reasonable care to carry out his undertaking. Extending the type of voluntary undertaking of another’s care that

section 325 contemplates to participants in an on-again/ off-again romantic relationship would involve mental gymnastics that we are not prepared to undertake. For present purposes, it suffices to say that such a leap in logic is far from reasonably clear. III. CONCLUSION We need go no further. For the reasons elucidated above, the judgment is Affirmed. . Sanders sued both The Phoenix Insurance Company (which issued the homeowner’s insurance policy to Doe) and The Travelers Indemnity Company of America (which, like Phoenix, is a wholly-owned subsidiary of The Travelers Companies, Inc.).

In the district court, the parties referred indiscriminately to both defendants. Phoenix now insists for the first time that The Travelers Indemnity Company of America is not a proper party. Because nothing in this case turns on any distinction between these sister companies, we refer throughout to them, collectively, as "Phoenix.” .

Section 2 of

Chapter 93A declares unlawful ‘‘[ulnfair methods of competition and unfair or deceptive acts or practices in the conduct of any trade or commerce,” Mass. Gen. Laws ch. 93A, § 2(a). The statute creates a private right of action through which consumers may seek equitable and monetary relief from businesses employing unfair or deceptive practices. See id. § 9(1). . According to Phoenix, Doe’s liability to Sanders was on a non-recourse basis; that is, the settlement agreement stipulated that the additional $500,000 could be collected only from whatever proceeds might be due under the Policy.

Sanders has not contradicted this description. . In our discussion of these rulings, it would serve no useful purpose to distinguish between the district judge and the magistrate judge. Instead, we take an institutional view and refer throughout to the district court. . The district court agreed with this conclusion, finding the

Chapter 93A demand letter insufficient to trigger the duty to defend. So, too, another district court has held that a

Chapter 2IE pre-suit demand letter did not animate the duty to defend. See Zecco, Inc. v. Travelers, Inc,, 938 F.Supp. 65, 68-69 (D. Mass. 1996) (discussing Mass. Gen. Laws ch, 2IE, § 4A; comparing that provision to

Chapter 93A; and reasoning that the Hazen Paper court had "carefully avoided opening the door” tp having an array qf pre-suit demand letters impose the duty to defend). Yet a third district court, though, has reached a contrary conclusion (but held nonetheless that various policy exclusions precluded coverage), See Cytosol Labs., Inc. v. Fed. Ins. Co., 536 F.Supp.2d 80, 88, 90 (D. Mass. 2008). The Cytosol decision contains very little in the way of reasoning on this point, and we find that court’s conclusion less convincing than the contrary conclusions reached by the court below and by the Zecco court. .

In somewhat the same vein, we note that Sanders also assigns error to the district court's ruling that the Policy’s professional services exclusion barred coverage. But the challenged ruling was simply that ”[t]o tire extent in this action that the plaintiff sought coverage of claims based on Doe’s professional misconduct under the homeowner’s policy, the professional services exclusion would be effective to preclude those claims.” Here, however, the liability that Sanders posits is premised upon Doe’s decision to embark on a volatile intimate relationship with a known vulnerable person.

Since this asserted liability is not based on a professional responsibility theory, any error in the challenged ruling is harmless. . For essentially the same reasons, we reject Sanders’s assertion that "alternative.dispute resolution proceedings [can come] within the scope of policies utilizing the term 'suit' alone,” Without exception, the few cases that Sanders cites for that proposition involve policies that explicitly include arbitration or alternative dispute resolution ■ in their

definitions of “suit,” See, e,g., Sunoco, Inc, v. Ill. Nat'l Ins, Co„ No. 04-4087, 2007 WL 127737, at *11 (E.D, Pa. Jan. 11,2007).