Amherst Country Club v . Harleysville CV-07-136-JL 6/24/08 P

UNITED STATES DISTRICT COURT

DISTRICT OF NEW HAMPSHIRE

Amherst Country Club, Inc.

v. Civil No. 07-cv-136-JL Opinion No.

Harleysville Worcester Insurance Co.

O R D E R

After the swimming pool on its premises was destroyed in the

Mother’s Day Flood of 2006, Amherst Country Club, Inc. petitioned

the New Hampshire Superior Court for declaratory judgment that

the Club’s insurer, Harleysville Worcester Insurance Company, is

obligated to cover the loss. See N.H. Rev. Stat. Ann. 491:22

(2001). Harleysville removed the action to this court pursuant

to

and 1446 (2006).

This court has subject matter jurisdiction under

(diversity of citizenship).

The parties have filed cross-motions for summary judgment

under Federal Rule of Civil Procedure 5 6 . Each party argues that

there are no genuine issues as to any material facts, but

advocates for entirely different legal conclusions. Amherst

Country Club argues that it is entitled to coverage as a matter of law, while Harleysville argues, also as a matter of law, that

the insurance policy in question excludes coverage for the loss.

After hearing oral argument on the cross-motions, and after

reviewing the parties’ respective memoranda, objections,

affidavits, reply briefs, expert reports and depositions, the

court denies the Club’s motion for summary judgment, grants

Harleysville’s motion for summary judgment, and awards judgment

to Harleysville.

I. APPLICABLE LEGAL STANDARD

Under Federal Rule of Civil Procedure 5 6 , a motion for

summary judgment will be granted “if the pleadings, the discovery

and disclosure materials on file, and any affidavits show that

there is no genuine issue as to any material fact and that the

movant is entitled to judgment as a matter of law.” Fed. R. Civ.

P. 56(c) (2008) (amended December 1 , 2007); see Celotex Corp. v .

Catrett,

(1986) (decided under prior,

substantially identical version of the rule); Anderson v . Liberty

Lobby, Inc.,

(1986) (same). "The object of

summary judgment is to pierce the boilerplate of the pleadings

and assay the parties' proof in order to determine whether trial

is actually required." Dávila v . Corporación de P.R. Para la

Difusión Pública,

(1st Cir. 2007) (quotations

2 omitted) (quoting Acosta v . Ames Dep’t Stores, Inc.,

,

7 (1st Cir. 2004)).

Both parties have moved for summary judgment. “Cross

motions simply require [the court] to determine whether either of

the parties deserves judgment as a matter of law on facts that

are not disputed.” Littlefield v . Acadia Ins. Co.,

, 6

(1st Cir. 2004). New Hampshire law, however, which the parties

agree is controlling,1 places the burden of proof on the insurer.

“By statute, the burden is on the insurance carrier to prove a

lack of coverage.” Hudson v . Farm Family Mut. Ins. Co.,

(1997) (citing N.H. Rev. Stat. Ann. 491:22-a); 2 Union

Mut., 835 F.Supp. at 63 (“[W]hen insurance coverage is disputed,

New Hampshire law places the burden of proving that no coverage

1 See Moores v . Greenberg,

, 1107 n.2 (1st Cir. 1987) (where parties agree as to what substantive law applies, a federal court sitting in diversity jurisdiction should comply); see also Union Mut. Fire Ins. C o . v . Hatch,

9 , 62 (D.N.H. 1993) (Devine, J . ) . 2 The applicable New Hampshire statute provides as follows:

491:22-a Liability Coverage; Burden of Proof. In any petition under RSA 491:22 to determine the coverage of a liability insurance policy, the burden of proof concerning the coverage shall be on the insurer whether he institutes the petition or whether the claimant asserting coverage institutes the petition.

N.H. Rev. Stat. Ann. 491:22-a (2001).

3 exists on the insurer.”) (citing Laconia Rod & Gun Club v .

Hartford Acc. & Indem. Co.,

(1983)).

“The interpretation of the language of an insurance policy,

like any contract language, is ultimately an issue for the court

to decide.” Merchants Mut. Ins. C o . v . Laighton Homes, LLC,

(2006) (citing D’Amour v . Amica Mut. Ins. Co.,

(2005)); Ekco Group, Inc. v . Travelers Indem. C o .

of Ill.,

(1st Cir. 2001).

II. BACKGROUND

During mid-May, 2006, much of southern New Hampshire

experienced prolonged, heavy rains, which became known as the

“Mother’s Day Flood.” The parties and their respective experts

agree that the rain and flooding caused increased groundwater

levels, and that the water table in the pool area of the Amherst

Country Club was unusually high. See infra Part III(B)(1) and

n.9.

On May 18 or May 1 9 , 2006, a maintenance worker at the

Amherst Swim Club, Inc. (the entity which operated the swimming

pool located on the premises of the Amherst Country Club) drained

the water from the pool for its spring cleaning, as had been done

annually for approximately 30 years. The soil surrounding the

concrete swimming pool was saturated with groundwater, creating

4 hydrostatic pressure that, once enough water had been drained

from the pool, “floated” the pool up and out of the ground. This

disturbance of the pool’s physical position caused its structure

to crack and break, destroying i t .

The Club, which had purchased its building and personal

property insurance coverage from Harleysville, notified its

insurance agent of the loss and requested coverage. The coverage

provision of the policy’s “Building and Personal Property

Coverage Form” states:

A. Coverage

We will pay for direct physical loss or damage to Covered Property at the premises described in the Declarations caused by or resulting from any Covered Cause of Loss.

1. Covered Property

a. Building, meaning the building or structure described in the Declarations, including:

. . . .

(2) Fixtures, including outdoor fixtures.

The policy also includes a list of exclusions, or losses not

covered by the policy. The pertinent “exclusions” provide:

B. EXCLUSIONS

1. We will not pay for loss or damage caused directly or indirectly by any of the following. Such loss or damage is excluded regardless of any other cause or event that

5 contributes concurrently or in any sequence to the loss.

b. Earth Movement

(4) Earth sinking (other than sinkhole collapse), rising or shifting including soil conditions which cause settling, cracking or other disarrangement of foundations or other parts of realty. Soil conditions include contraction, expansion, freezing, thawing, erosion, improperly compacted soil and the action of water under the ground surface.

g. Water

(4) Water under the ground surface pressing on, or flowing or seeping through:

(a) Foundations, walls, floors or paved surfaces;

(b) Basements, whether paved or not; or

(c) Doors, windows or other openings.

Section B(1) of the exclusion provision set forth above is

referred to by the parties as the “lead-in” provision, and is

known in the industry as an “anti-concurrent causation” clause.

The parties refer to sections B(1)(b) and B(1)(g) as the “earth

movement exclusion” and the “water exclusion,” respectively.

Harleysville declined coverage based on the “water” exclusion.

This declaratory judgment action followed.

6 III. ANALYSIS

The parties agree that, although the “coverage” section of

the insurance policy makes no specific reference to swimming

pools, it applies to the pool in question as a “fixture” or

“outdoor fixture.” See supra Part I I ; Pleasant Valley

Campground, Inc. v . Rood,

6 , 88 (1980).

Amherst Country Club argues that (1) neither the “earth

movement” nor “water” exclusion applies because they are too

ambiguous, (2) in any event, the cause of the damage to the pool

was not groundwater pressure, but the draining of the pool by a

pool attendant,3 and (3) the lead-in anti-concurrent clause

doctrine, if applicable, is unenforceable as a matter of public

policy. Harleysville differs on all three points, arguing that

(1) the plain and unambiguous language of the “earth movement”

and “groundwater” exclusions releases it from the obligation to

cover the losses, (2) the cause of the loss was the sub-surface

groundwater pressure around and under the pool, as opposed to its

3 At oral argument on the motion, the Club retreated from this position, emphasizing instead that groundwater pressure could not be the “efficient proximate cause.” See infra Part III (B)(1). This shift, which did not undermine the Club’s arguments or position in any way, was likely the result of the court’s pointing out, and the Club conceding for the purposes of the motion, that neither draining the pool nor the groundwater pressure could have caused the loss alone, without the other, and that neither event could thus constitute the efficient proximate cause as a matter of law. See

7 draining, and (3) regardless of whether the groundwater pressure

was the dominant cause or a concurrent cause, the “anti-

concurrent causation” lead in clause excludes coverage, and is

enforceable under New Hampshire law.

A. The terms of policy

1. Applicable law

The New Hampshire Supreme Court, and thus this court,

“interprets an insurance policy in the same manner as any other

contract.” Hudson,

(citing Trombly v . Blue

Cross/Blue Shield,

(1980)). “The fundamental

goal of interpreting an insurance policy, as in all contracts, is

to carry out the intent of the contracting parties. To discern

the parties’ intent, we first examine the language of the

contract itself.” Bates v . Phenix Fire Ins. Co., ___ N.H. ___,

(N.H. 2008) (quoting Tech-Built 153 v . Va. Surety

Co.,

(2006)). New Hampshire courts will

“enforce a policy provision that limits the insurance company’s

liability when the policy language is clear and unambiguous.”

Merchants Mut.,

(citing Deyette v . Liberty Mut.

Ins. Co.,

(1997)). As discussed more fully

infra, however, ambiguities in the policy’s statement of coverage

8 are construed in favor of the insured. See, e.g., Hoepp v . State

Farm Ins. Co.,

(1997).

The parties’ dispute revolves in part around the meaning of

three words in the “water” exclusion--“walls,” “floors,” and

“paved surfaces,” that are not defined in the policy. If a term is not defined in the policy, the term is to be given its plain and ordinary meaning, construed “as would a reasonable person in the position of the insured based on more than a casual reading of the policy as a whole.”

Sig Arms, Inc. v . Employers Ins. of Wausau,

(D.N.H. 2000) (DiClerico, J.) (quoting High Country Assocs.

v . N.H. Ins. Co.,

9 , 41 (1994)). This familiar

standard--requiring construction of the policy “as would a

reasonable person in the position of the insured based on more

than a casual reading of the policy as a whole”--is applied by

New Hampshire courts not only to purported ambiguities, but to

all “terms of the policy.” Littlefield,

. Only

“objectively reasonable expectations of the insured,” however,

“can negate policy provisions that would negate those

expectations.” Comm. Union Assur. C o . v . Aetna Cas. & Sur. Co.,

(D.N.H. 1978).

9 2. The exclusions

a. The water exclusion

The “water” exclusion in the Harleysville policy excludes

coverage for “[w]ater under the ground surface pressing o n , or

flowing or seeping through [f]oundations, walls, floors or paved

surfaces.” The Club argues that the exclusion does not apply to

swimming pools, and specifically that the lack of internal

definitions for the terms “walls,” “floors,” and “paved

surfaces,” requires the exclusion to be construed against

Harleysville. A swimming pool, the argument goes, is not in and

of itself a wall, floor, or paved surface. And even if walls,

floors, and paved surfaces could potentially be parts of swimming

pools, an equally plausible reading is that those are parts of a

building, as opposed to a fixture like a pool. If this exclusion

were meant to apply to groundwater damage to swimming pools, says

the Club, the exclusion should make specific reference to

swimming pools, as did the policies at issue in Murray v . All

American Ins. Co.,

(1997), and Bebber v . CNA

Ins. Co.,

(2001).

This argument is unpersuasive, because it relies on an

overly expansive view of the concept of ambiguity in insurance

contracts, suggesting that New Hampshire courts construe even the

10 most superficial ambiguities against insurers. As the First

Circuit Court of Appeals noted in Ekco Group,

New Hampshire, like most states, tends to favor the insured where the policy is genuinely ambiguous and the choices between two plausible readings, one providing coverage and the other not. But plausibility is a matter of degree, and the policy may be unclear in some respects and clear enough in others.

(1st Cir. 2001) (emphases added) (internal

citation omitted). “New Hampshire courts . . . will not find

that a clause is ambiguous simply to interpret the clause in

favor of the insured and against the insurer.” LaSorsa v . Unum

Life Ins. C o . of Am.,

, (1st Cir. 1992)

(citing Laconia Rod & Gun Club,

). Neither the

fact that the parties “dispute the scope of a policy’s coverage,”

Titan Holdings Syndicate, Inc. v . City of Keene,

(1st Cir. 1990), nor “that the parties may disagree on the

interpretation of a term of clause in an insurance policy,”

Bates,

, renders the policy ambiguous.

“The meaning of the language must be unclear, and the

parties’ dispute based upon reasonable differences about the

language’s interpretation.” Titan Holdings,

(emphases added). The case law defining the concept of “genuine

ambiguity” in insurance policies makes reasonableness the

cornerstone of the inquiry. See, e.g., LaSorsa, 955 F.Supp at

11 147; Curtis v . Guaranty Trust Life Ins. Co.,

(1989);

Smith v . Liberty Mutual Ins. Co.,

(1987)

(Souter, J.) (citing 3 A . Corbin, Contracts § 543A (Supp. 1971);

Trombly,

; Miller v . Amica Mut. Ins. Co.,

(2007); Trombley v . Liberty Mut. Ins. Co.,

(2002).

Assessing the purported ambiguity under the standards

established by the applicable New Hampshire Superior Court

precedent, this court concludes that a reasonable person in the

position of the Amherst Country Club, undertaking a more than

casual reading of the policy as a whole, would understand that

the “water” exclusion unambiguously applies to swimming pools.

The fact that the terms “wall,” “floor,” and “paved surface” are

not defined in the policy does not require, or even suggest, a

different conclusion. “[T]he draft of a policy need not define

each word in the policy ad infinitum, but may rely on the

ordinary meanings of words.” Titan Holdings,

(citing Robbins Auto Parts, Inc. v . Granite State Ins. Co.,

(1981)). It is difficult for the Club, as the

owner of a swimming pool, to argue convincingly that the words

“walls” and “floor” are not commonly used to refer to the sides

and bottom of a swimming pool. Everyday words like “wall,”

“floor,” and “paved surface,” are particularly amenable to

12 construction according to their ordinary meanings rather than by

technical definition.

Rather than reflexively declare these undefined words to be

ambiguous and construe them against Harleysville, the court must

“look to the claimed ambiguity, consider it in its appropriate

context, and construe the words used according to their plain,

ordinary, and popular definitions.” Union Mut., 853 F.Supp. at

62 (quoting LaSorsa, 955 F.2d at 1 4 8 ) . A “wall” is defined4 as a

“vertical architectural member used to define and divide space,”

Webster’s Third New International Dictionary 2572 (2002)

(“Webster’s”), and as an “upright structure of masonry, wood,

plaster or other building material serving to enclose, divide, or

protect an area.” The American Heritage Dictionary of the

English Language 1936 (2000) (“American Heritage Dictionary”).

Likewise, the common definition of “floor” is “the lower inside

surface of any hollow structure.” Webster’s 873, and as the

“lower or supporting surface of a structure.” American Heritage

Dictionary 674.

Further indications that the plain and ordinary meanings of

the words “wall” and “floor” include the sides and bottom of a

4 See Hudson,

(Broderick, J.) (“dictionaries are of some value [in interpreting insurance policies] to the extent they inform us of the common understanding of terms”); see also Littlefield,

.

13 swimming pool are contained in the expert witness depositions

taken in this case. There, experts and counsel alike--including

counsel for the Club--made continuous, comfortable, off-handed

references to the sides and bottom of the Amherst Country Club

swimming pool as its “walls” and “floor.” (Reynolds Deposition

Transcript at 40-41 (walls), 42 (wall, floor), 73 (floor);

Cricenti Deposition Transcript at 23 (floor), 37 (wall), 43-45

(wall, floor)). The Club’s counsel sometimes objected to form

when Harleysville’s counsel included “wall” and “floor” in his

questions, (Cricenti Deposition Transcript at 23 (floor), 43-45

(floor, wall)), but objections or lack thereof to this

phraseology are not the point. At no time during the deposition,

did anyone present express the slightest bit of confusion over

attorneys’ or witnesses’ use of the terms “wall” and “floor” in

reference to the pool’s sides and bottom.5

Even if one were to assume that the existence of walls and

floors in structures other than swimming pools rendered these

terms genuinely ambiguous in the context of the policy as a

5 Amherst Country Club correctly points out that the ambiguity standard is not to be applied from the perspective of professional engineers but from that of laymen. The attorneys conducting and taking the depositions, however, were in fact laymen in the structural engineering field, and required no definitional explanations for the pool’s walls and floor, as were requested for terms like “high pressure valve” (Cricenti Deposition Transcript at 23) and “ballast” (id. at 2 4 ) .

14 whole, the same cannot be said of the term “paved surfaces.”

Although this term may more immediately evoke the notion of a

driveway or parking lot, there can be no question that the plain

and ordinary meaning of the term is broad enough to include the

Club’s swimming pool. The Club’s expert testified at his

deposition that the pool and its floor were constructed of

“concrete” that was “cast in place,” in other words, mixed and

applied directly to the surface of the excavated cavity in which

the pool sat, just like concrete paving on sidewalks, driveways,

and patios. Indeed, “paved” is defined simply as “covered with a

pavement,” Webster’s 1658; see also American Heritage Dictionary

1291 (“pave”), with a specific reference to “concrete” as such a

material. Webster’s 1658; see also American Heritage Dictionary

1291. To the extent that “surface” requires a definitional

explanation, it is “the exterior or outside of an object or body;

the outer most or upper most boundary.” Webster’s 2300; see also

American Heritage Dictionary 1741.

The Club strived mightily, both in its summary judgment

filings and at oral argument, to establish that a pool cannot be

or have a paved surface because paved surfaces are “paved” as

opposed to “poured,” composed of asphalt as opposed to concrete,

and are designed solely for smoothness and flatness to facilitate

travel or movement. Leaving aside the dictionary definitions

15 just cited, this argument runs counter to any common

understanding of the word “pavement.” Concrete streets,

driveways, and patios are no less paved surfaces because they are

comprised of concrete that is poured, as opposed to asphalt that

is laid or paved. Were “paved surfaces” intended to be limited

to asphalt in the manner the Club suggests, one would have

expected the exclusion to specify “asphalt surfaces,” rather than

use a more generic term.

Finally, use of the term “paved” elsewhere in the policy

reinforces this conclusion. The subsection immediately following

the reference to “walls, floors, and paved surfaces,” Section

B(1)(g)(4)(b), excludes groundwater damage to “[b]asements,

whether paved or not.” If a basement, which is almost always

poured concrete, is “paved” under the terms of the policy, the

same unambiguously can be said of a swimming pool. See Sig Arms,

(policy must be construed “as a whole”)

(quoting High Country Assocs., 139 N.H. at 4 1 ) .

The Club has cited no authority, binding or persuasive, for

the proposition that the terms “wall,” “floor,” and “paved

surface” are ambiguous as applied to a swimming pool in the

context of a property insurance policy exclusion. Although the

parties cited several floating-swimming-pool cases, the only one

considering whether these words are ambiguous is AGK Holdings,

16 Inc. v . Essex Insurance Co.,

(6th Cir. 2005).

In that case, which applies an identical “water” exclusion to

substantively identical facts, the insured argued “that the

groundwater exclusion does not apply to swimming pools,”

, relying on precisely the same arguments that the Club

advances here: that the terms “floor” and “paved surface” are

susceptible to more than one interpretation. Utilizing

dictionary definitions, however, the Sixth Circuit Court of

Appeals found no ambiguity in those terms, and ruled that they

commonly could be understood to refer to parts of a swimming

pool.

. Like the Club here, the plaintiff in AGK

Holdings further argued that the absence of a reference to

“swimming pools” in the exclusion, under circumstances where

specific references to pools were made in other sections of the

policy, indicated that the groundwater exclusion did not apply.

The Sixth Circuit disagreed, finding that “such an argument

ignores the clear and unambiguous language of the contract terms.

Since a ‘floor’ or ‘paved surface’ can be applied respectively to

the bottom or interior surface of the swimming pool, inclusion of

more specific language is not required.”

. 6

6 The Club argues that AGK Holdings is inapposite for several reasons, two of which merit discussion. It first argues that the AGK Holdings Court based its decision “on the principle that a policy term with several definitions must be construed in

17 So while it is true that a specific reference to “swimming

pools” in the water exclusion would have eliminated all debate in

this case, its absence creates no ambiguity. “While a claimed

ambiguity need not be apparent on the face of the policy, [this

court] will not perform amazing feats of linguistic gymnastics to

find a purported ambiguity.” Hudson,

; see also

Fed. Bake Shop v . Farmington Cas. Co.,

0 , 42 (1999).

favor of the insurance company, as opposed to the policyholder. The AGK Holdings court neither referred to nor followed any such principle. Id. at 892-93. In fact, because it found no ambiguity, it had no occasion to resolve it in either party’s favor.

The Club further distinguishes AGK Holdings because the court there, “[c]ontrary to the approach . . . under New Hampshire law . . . allowed for examination of extrinsic evidence to” resolve ambiguity, and would only construe ambiguous language against the insurer as a last resort. Although extrinsic evidence (save dictionary definitions, specifically endorsed as useful in Hudson,

(Broderick, J.)) played no role in this court’s construction of the Harleysville policy, it bears noting that New Hampshire law is no different. First, there is the familiar rule that permits New Hampshire courts to consider the “context” of the claimed ambiguity, as well as the “plain, ordinary, and popular definitions” of the words or terms in question. Union Mutual,

2 . Second, and more to the point, the rule strictly construing ambiguities –- even genuine ambiguities -– against the insured is “a rule of presumption only.” Town of Epping v . S t . Paul First and Marine Ins. Co.,

(1982). The court may consider extrinsic evidence to rebut the presumption by resolving such ambiguities,

, to ensure that “the ambiguity rule not be applied to create coverage where it is clear that none is intended.”

at 252 (quoting Robbins Auto Parts,

6 2 ) ; see also Smith,

(Souter, J . ) ) .

18 The focus is on the ordinary meaning of the policy language,

which cannot possibly provide specifically for every conceivable

circumstance. Were it otherwise, routine policies might exceed

hundreds or thousands of pages in length, rendering the “more

than casual reading by the insured” standard a complete fiction

with no relationship to the likely conduct of actual

policyholders. The “water exception” is unambiguous, applies to

the pool, and excludes coverage for the loss in question.

b. The “earth movement” exclusion

Though the “water” exclusion unambiguously excludes the loss

at issue here from the coverage of the policy, Harleysville

argues that the “earth movement” exclusion unambiguously excludes

the Club’s loss as well. The court agrees. The provision,

Section B(1)(b)(4), excludes coverage for “[e]arth sinking . . .

rising, or shifting, including soil conditions which cause . . .

disarrangement of foundations or other parts of realty.” The

exclusion defines “soil conditions” to include “the action of

water under the ground surface.”

This provision does not suffer from any of the purported

ambiguity-causing shortcomings the Club has identified in the

“water” exclusion. It expressly identifies the scope of the

exclusion with respect to the property: “foundations or other

19 parts of realty.” The Club conceded at oral argument that, as a

fixture, the swimming pool was part of its “realty.”7 The

exclusion describes precisely the type of damage sustained by the

pool: “cracking or other disarrangement.” And the definition of

“soil conditions” specifically include “the action of water under

the ground surface.”

Oral argument included an interesting discussion about

whether this definition properly treats groundwater as “part of”

the earth’s soil, subject to “movement” within the meaning of the

exclusion, or whether groundwater just “flows through” the earth

without causing it to move. Regardless of the answer to that

question in the geological sense, the exclusion applies here

because it expressly defines groundwater action as one example of

“soil conditions which cause . . . cracking or disarrangement of

. . . parts of realty.” This detailed definition eliminates any

ambiguity as to the scope of “earth movement.” Cf. Fed. Bake

Shop,

(construing policy term found to be

7 In addition, though not discussed extensively at the hearing or in the parties’ respective filings, and not necessary for this court’s ruling, the reinforced concrete pool, almost all of which sat beneath the surface of the ground, is also a “foundation.” See, e.g., Cricenti Deposition Transcript at 11 ( “ . . . the concrete that we see for the bottom and the sides is the foundation of the pool.”) This observation is also relevant to the “water” exclusion, which expressly applies to “foundations.”

20 ambiguous against insurer which “could have defined” the

ambiguous term, but “chose not to do s o ” ) .

The parties agree that groundwater pressure was at least a

concurrent cause of the pool’s floating out of the ground and

eventual destruction. A reasonable insured undertaking more than

a casual reading of the earth movement exclusion would have no

difficulty discerning that (1) “the action of water under the

ground surface” was (2) a “soil condition” that (3) “cause[d] . .

. cracking or other disarrangement of part[] of realty,” thus

meeting one of the policy’s definitions on earth movement.

It makes no difference that, as the Club points out, both

experts disavowed “earth movement” as the cause of the loss; that

testimony was elicited by questions involving was “earth

movement” in the generic sense, not as specifically defined in

the exclusion. The cases cited by the Club in its summary

judgment papers are unhelpful for similar reasons.8 None of them

construes “earth movement” exclusions which make any express

reference to groundwater movement, much less include, as the

8 See Nautilus Ins. C o . v . Vuck Builders,

(N.D. Ill. 2005); Murray v . State Farm Fire & Cas. Co.,

, 17-20 (W. Va. 1998); M&M Holdings, Inc. v . State Auto Prop. & Cas. Ins. Co., N o . 06-4031-SAC,

, at *3 (D. Kan. 2007).

21 Harleysville policy does, it as a specific type of earth movement

within the meaning of such an exclusion.

Finally, at oral argument, the Club argued that the “earth

movement” exclusion is ambiguous because, even though groundwater

movement itself was specified as a type of cause under the

exclusion, an ambiguity nonetheless results from the exclusion’s

title, “Earth Movement,” a heading in the insurance contract.

Referencing groundwater movement under an exclusion titled “earth

movement” is inherently ambiguous, the argument goes, because a

reasonable insured might not think of water as part of the soil,

and thus would not consider groundwater movement to constitute

earth movement. This argument, however, is inconsistent with New

Hampshire’s “reasonable insured undertaking a more than casual

reading of the policy as a whole” standard. Whatever general

impression the title of the exclusion might cause, the language

of the provision itself specifies “the action of water under the

ground surface” as an excluded cause. See Cross Petroleum v .

United States, 51 Fed. C l . 549, 555 (Ct. C l . 2002) (reasoning

that title of contractual provision “does not invite the court to

ignore the plain meaning of the provision’s language”); accord

N.H. Ins. Guar. Ass’n v . Pitco Frialator, Inc.,

(1998) (ruling that language of statutory provision trumped its

22 title). The “earth movement” exclusion is unambiguous, applies

to the pool, and excludes coverage for the loss in question.

B. Causation

1. Applicable law -– efficient proximate cause

The parties agree that increased groundwater levels during

and after the 2006 Mother’s Day Flood caused the pool to “float”

upward as its weight (or ballast) was decreased when a pool

attendant pumped out the pool for its annual cleaning.9 They

9 In his expert report for the Amherst Country Club, Professional Engineer Nicholas Cricenti, Jr., stated:

Cause of the Failure

When the pool is full of water and the groundwater is high the water inside acts as a ballast to keep the pool vessel from floating on the groundwater. This is evidenced by the fact that the pool did not float at any time during the previous week when the groundwater was at least as high and maybe even higher around the pool. There was standing water on the deck earlier in the week.

As the water inside the pool is pumped out, the ballast is removed and the pool is pushed upward by the pressure of the groundwater under the pool trying to equilibrate with the elevation of the groundwater. At one point during the pumping process the forces pushing up overcame the weight of the ballast remaining in the pool and the pool floated.

Likewise, the report of the Harleysville’s expert Professional Engineer, Peter Reynolds, stated:

23 disagree, however, on whether the emptying of the pool or the

groundwater beneath it was the cause of its destruction within

the meaning of the exclusions.

As Professor Couch has observed,

[t]he concept of “proximate cause” has a different meaning and applications in the area of insurance law than it has in tort law. Tort law applies the rules of proximate cause for the purpose of fixing culpability for the damage being claimed in the lawsuit. . . . In contrast, the doctrine of proximate cause as applied to insurance law bears no relationship with the determination of “culpability” or the explanation for why the injury took place. Instead, insurance law employs the concept of proximate cause for purposes of determining whether the specific type of injury caused by the specific type of physical act or event was intended to be covered under the terms of the subject policy.

Cause of incident: As the water in the pool was pumped out, the upward pressure of the groundwater eventually became more than the weight of the pool and contained water causing the deep end of the pool to float upward.

Both experts provided deposition testimony consistent with these conclusions (Cricenti Deposition Transcript at 46-50; Reynolds Deposition Transcript at 7 3 ) .

24 7 Couch on Insurance § 101:40 (Steven Plitt, et a l . , eds.,

2008). 10 Professor Stempel refers to this distinction as being

between “tort causation” and “contract causation.” 1 Stempel,

§ 7.01, 1-10.

The Club argues –- and Harleysville does not seriously

dispute -- that New Hampshire has adopted the “efficient

proximate cause” doctrine, as opposed to the more permissive

“concurrent cause doctrine,”11 as the standard for determining

so-called “contract causation” under an insurance policy.

“Efficient proximate causation is the majority rule today.”

4 Leitner, § 52.33.

The efficient proximate cause doctrine has been applied under those circumstances in which two or more identifiable causes, at least one of which is

10 Professor Couch’s view is widely held. “As one commentator observed, with some degree of understatement, [c]ausation has always been a troubling concept for lawyers.” 1 J. Stempel; Stempel on Insurance on Insurance Contracts § 7.01 (Supp. 2008) (quoting B . McDowell, Causation in Contracts and Insurance,

(1988)). “In all Anglo-American law, there is no concerpt that has been so pervasive - and yet so elusive - as the causation requirement . . . .” P. Swisher, Insurance Causation Issues: the Legacy of Bird v . S t . Paul Fire & Marine Ins. Co.,

(2002). 11 The minority concurrent cause rule “looks to whether one of the causes of a loss is covered. If s o , the loss itself is covered notwithstanding the fact that there is also an excluded cause in the chain of causation.” 4 D. Leitner, et a l . , Law and Practice of Insurance Coverage Litigation, § 52:33 (2008).

25 covered under the policy and at least one of which is excluded thereunder, contribute to a single loss. If the cause which is determined to have set the chain of events in motion, the efficient proximate cause, is covered under the terms of the policy, the loss will likewise be covered.

7 Couch, § 101:45. The New Hampshire Supreme Court applied the

efficient proximate cause doctrine,12 albeit not by name, in

Terrien v . Pawtucket Mut. Fire Ins. Co.,

(1950),

where the latter and more direct of two contributing causes was

held to be the proximate cause of the injury for purposes of

determining coverage under an insurance policy.

See also Nassif Realty Corp. v . National Fire Ins. C o . of

Hartford,

(1968) (citing Terrien and using the

phrase “the dominant and the efficient cause” to describe

proximate cause in the context of insurance coverage);

N.H. Ins. C o . v . Schofield,

(1979) (citing

Nassif).

12 Scholarly and professional authority on the subject appears to concur with the parties that New Hampshire has adopted the efficient proximate cause doctrine, see D. Wuerfel & M. Koop, “Efficient Proximate Causation” in the Context of Property Insurance Claims,

, 405 (1998) (citing Terrien and listing New Hampshire among states applying efficient proximate causation rule); 4 Leitner, supra, § 52:36 (same, also citing Terrien); 2 B . Ostrager, T . Newman, Handbook on Insurance Coverage Disputes § 21.02[c] at 1457 (2008).

26 The Club argues that the efficient proximate cause of the

loss was the draining of the pool, bringing its destruction under

the coverage of the policy nonwithstanding the exclusions. But

[e]fficient proximate cause does not apply in a case in which there is more than one cause of loss and none of the causes is sufficient by itself to cause the loss. Efficient proximate cause applies only where there are several causes of loss in a chain of causation and each cause could independently cause the loss.

4 D. Leitner et a l . , § 52:33 (emphases added). See also 7 Couch,

§ 101:45 (“Under any circumstances, in order for the efficient

proximate cause doctrine to apply, there must be at least two

potential causes of the subject loss.”); In r e : Katrina Canal

Breaches Litig.,

(5th Cir. 2007); Capitol

Indemnity Corp. v . Evolution, Inc.,

(D.N.D. 2003); Crete-Montee Sch. Dist. v . Ind. Ins. Co.,

(N.D. Ill. 2000); Pieper v . Comm. Underwriters Ins. Co.,

59 Cal. App, 4th 1008, 1020 (Cal. C t . App. 1997); see also Mattis

v . State Farm Fire & Cas. Corp.,

(Ill. C t .

App. 1983). This case does not satisfy those criteria.13 The

parties agree that the groundwater pressure would not have

floated the pool out of the ground had the pool not been emptied,

and neither argues that emptying the pool in the absence of the

The Club conceded as much at oral argument.

27 elevated water table could have itself caused the pool to float.14

Thus, neither the draining of the pool–-a covered cause–-nor

the pressure of the groundwater–-an excluded cause–-was the

efficient proximate cause of the loss. Because, as just

discussed, New Hampshire law requires the excluded cause to be

the efficient proximate cause of the loss to trigger such an

exclusion, see, e.g., 7 Couch, supra, § 101:45, it follows that

the exclusions would not apply here. But the problem with this

line of argument, as Harleysville points out, is that the policy

14 The court again notes the Club has unequivocally asserted (see supra Part III and n . 3 ) that the groundwater had no impact whatsever on the chain of causation. Even if the Club had not softened that position at oral argument, (see id.) it would have no bearing on this court’s analysis. The entire record, including the Club’s expert’s opinion, and the fact that the pool had been drained annually for approximately 30 years without incident make this argument untenable, even in a summary judgment posture where Harleysville carries a statutorily imposed burden of proof and all inferences are drawn in the Club’s favor.

The Club’s argument that this court should follow Bebber v . CNA Ins. Co.,

(N.Y. Supr. C t . 2001), ruling that emptying the pool was the cause, is rejected. Bebber is both legally and factually inapposite. The New York Supreme Court in Bebber did not apply efficient proximate cause analysis as a New Hampshire court would, but rather applied a “but for” causation test, more akin to concurrent causation analysis.

. Further, in that case, the court took pains to note that there was no evidence of increased hydrostatic groundwater pressure,

making the case dispositively distinguishable from this one.

28 itself eschews the concept of efficient proximate cause in its

anti-concurrent clause provision.

C. The anti-concurrent causation clause

The final point of contention between the parties is the

enforceability of the policy’s lead-in “anti-concurrent

causation” clause.15 If the clause -– which excludes coverage

for any loss caused “directly” or “indirectly” by any of the

specified causes, “regardless of any such cause or event that

contributes concurrently or in any sequence to the loss” –- is

enforceable, it resolves the dispute over causation without

resort to the esoteric concepts just discussed. Harleysville

argues that the parties to an insurance contract are “free to

contract around the efficient proximate cause doctrine,” that

many state and federal jurisdictions have enforced anti-

concurrent causation clauses, and that the Club cites no

authority suggesting that New Hampshire refuses, or would

refuse, to enforce anti-concurrent causation provisions.

For its part, the Club argues that anti-concurrent causation

clauses “represent[] the latest effort of insurance carriers to

avoid the effect of the efficient proximate cause test which New

15 The anti-concurrent causation clause is set forth in this entirety supra Part I .

29 Hampshire has adopted,” that New Hampshire “has not and would not

allow the enforcement of an anti-concurrent causation clause,”

and that Harleysville, which carries the burden of proof in this

case, has not sustained his burden of proving otherwise.

Both parties’ arguments rely on the same flawed assumption:

that the New Hampshire Supreme Court has yet to enforce an anti-

concurrent causation clause. This shared misapprehension is

understandable, however, in that the New Hampshire high court’s

enforcement of such a clause occurred only recently, and with

little fanfare in the text of the opinion itself, or from

insurance commentators or practitioners. Further, at least one

scholarly authority counts New Hampshire among a “number of

states which have adopted the efficient proximate cause doctrine

[but] have not had occasion to consider the enforceability of

anti-concurrent causation policy language.” 4 Leitner, § 52:36 &

n.14 (listing New Hampshire as such a state, citing Terrien).

In Bates v . Phenix Mutual Fire Insurance Co., decided in

2008, the New Hampshire Supreme Court enforced an anti-concurrent

causation clause over the express objection of the insured.

. In Bates, heavy rains in southwest New Hampshire

overwhelmed a culvert, causing the area immediately uphill to

fill with an extraordinary volume of water.

. The

flooding eventually burst through a nearby road, releasing a

30 surge of water into the downstream valley which damaged the

plaintiff’s property. The insurance policy in question covered

losses from “explosion” (the cause claimed by the insured) but

excluded coverage for water damage (the cause claimed by the

insurance company) under an exclusion identical to the “water”

exclusion in the Harleysville policy here.

. The

“lead-in” provision to the exclusion provided:

We will not pay for loss or damage caused directly or indirectly by any of the following. Such loss or damage is excluded regardless of any other cause or event that contributes concurrently or in any sequence to the loss.

This language is identical to the “anti-concurrent

causation” lead-in in the policy at issue here.

The New Hampshire Supreme Court had no difficulty upholding

the entry of summary judgment for the insurance company and

expressly enforcing the anti-concurrent causation clause. Bates,

. The court assumed, without deciding, that

the road’s collapse near the culvert was an “explosion,” a

covered event, but went on to find that “the rain-induced flood

or overflow of Warren Brook either directly or indirectly caused

damage to the plaintiff’s properties . . . . Thus, the water

exclusion would apply to preclude coverage.”

(brackets omitted). The court’s reference to water “directly or

31 indirectly” causing the loss demonstrated that it had permitted

the parties to “contract around” the efficient proximate cause

doctrine with an anti-concurrent causation clause. The court’s

willingness to assume, without deciding, that the culvert and

road failure was an “explosion” only reinforces the point;

whether any contributing cause was the efficient proximate cause

was of no consequence, because the lead-in clause excluded both

direct and indirect causes.

The Club acknowledged Bates in its summary judgment

memorandum, but dismissed it as “legally distinguishable because

in Bates plaintiff did not contest, nor did the Supreme Court

analyze, the ‘lead in’ clause.” The court disagrees on both

counts. Although the opinion did not expressly acknowledge i t ,

the parties’ appellate briefs reveal that the insured raised, and

the insurer joined, the issue of the clause’s possible

unenforceability. The insurer’s brief notes that the trial court

expressly relied on the “anti-concurrent clause,” and went on to

argue–-incorrectly, as it turned out--that the clause was

inapplicable. (Brief on Behalf of the Plaintiff-Appellant at 1 5 ,

Bates v . Phenix Mutual Fire Ins. Co., N o . 2007-0177 (N.H. Supr.

Ct. 2008)). In its brief, the insurer made an even lengthier

argument on this point. (Brief of the Appellee at 13-14, Bates

v . Phenix Mutual Fire Ins. Co., N o . 2007-0177 (N.H. Supr. C t .

32 2008))). These arguments were reactions to the superior court’s

detailed analysis of the anti-concurrent causation clause, which

concluded “that New Hampshire would favor the majority rule

permitting parties to freely contract out of the efficient

proximate cause doctrine.” (Order on Cross-Motions for Summary

Judgment, Bates v . Phenix Mutual Fire Ins. Co., slip op. at 10-

1 2 , N o . 06-E-0046 (N.H. Sup. Ct., Feb. 7 , 2007)).

The court also rejects the Club’s argument that the New

Hampshire Supreme Court did not analyze the clause. It quoted

the entire anti-concurrent causation provision in the main text

of its opinion, Bates,

, and later quoted the

clause again, making specific reference to the words “directly or

indirectly” and “regardless of any other cause or event that

contributes concurrently or in any sequence to the loss.”

. In the same paragraph, the court noted its assumption

without deciding that an “explosion” was a contributing cause to

the loss, and its finding that water “directly or indirectly

caused damage to the plaintiff’s properties.”

(brackets

omitted). If these passages -– applying contract language to

facts, and doing so against the backdrop of an applicable

33 doctrine of law –- do not constitute “analysis,” this court is at

a loss to describe what does.16

It is true that the Bates court’s analysis does not

expressly address the clause’s enforceability as a matter of

public policy. This is not troubling, however, in light of the

court’s longtime recognition that consistency with public policy

is a prerequisite to the enforceability of any provision of an

insurance policy. See, e.g. Charest v . Union Mut. Ins. Co.,

(1973). Moreover, the superior court expressly

ruled that the clause was enforceable, and the superior court did

16 Counsel for the Club –- who should be credited with citing the Bates decision at all, as Harleysville did not –- also argues that Bates is distinguishable because the Bates plaintiff was viewed as “apparently conceding” that water was at least an indirect cause of the loss in that case, whereas the Club makes no such concession regarding the causal effect of the groundwater pressure here. See

. This argument overestimates the importance of “concession” as a distinguishing factor. The Bates court made a finding that water was at least an indirect cause of the loss; whether that finding was based on a concession or on some other evidence is of no consequence to the enforceability of the anti-concurrent causation clause. Further, the Bates plaintiff’s concession does not appear to have been an express, unequivocal concession, but rather an “apparent” one, Bates,

, much like the Club’s concession in this case, for purposes of summary judgment, that although the groundwater was not the efficient proximate cause of the loss, it was at least an indirect, concurrent cause in the sense of tort (as opposed to contract) causation. Finally, whether or not the Club is willing to make the concession that groundwater pressure contributed to the loss, its expert physical engineer agrees with Harleysville’s expert: the groundwater pressure, in combination with the draining of the pool, caused the pool to float and, thus, the ensuing loss. See supra Part III(B)(1)and n.9.

34 not so much as question, let alone overrule, that ruling on

appeal.

Other available data strongly suggests that, even if this

court had not expressly enforced an anti-concurrent causation

clause four months ago in Bates, there is little doubt that it

would enforce the clause at issue here. “In the absence of a

definitive ruling by the highest state court,” this court has

previously recognized that it “may consider analogous decisions,

considered dicta, scholarly works, and any other reliable data

tending convincingly to how the highest court in the state would

decide the issue at hand.” United Mutual,

-64

(quoting Redgrave v . Boston Symphony Orchestra, Inc.,

(1st Cir. 1988) (en banc)).

First, the New Hampshire rule has long been that “[a]bsent

statutory provision or public policy to the contrary, an

insurance company is free to limit its liability through an

exclusion written in clear and unambiguous language.” Miller,

(2007); Charest,

. New

Hampshire has not enacted legislation limiting the rights of

parties to insurance contracts to contractually limit coverage as

it has, for example, in the context of automobile liability

insurance. See Wegner v . Prudential Prop. & Cas. Ins. Co.,

(2002). Almost a decade ago, our sister district

35 of Massachusetts noted that the “vast majority of states” enforce

such clauses, and scholarly authorities confirm that this is

still the case. Preferred Mutual Ins. C o . v . Meggison,

(D. Mass. 1999); 4 Leitner, § 52:9; see M.

Wuerfel at 407; 7 Couch, § 101:45. Given this legacy, the Bates

court’s enforcement of the anti-concurrent causation clause was a

predictable application of its time honored approach to insurance

coverage, deferring where appropriate to “the intent of the

contracting parties.” Tech-Built 153,

. The

court rules that under New Hampshire law, the anti-concurrent

cause provision is enforceable.

IV. CONCLUSION

For the reasons set for above, the court finds that coverage

for the loss in question is excluded by the “water” exclusion or

the “earth movement” exclusion of the Harleysville policy, if not

both; that groundwater pressure, though not the efficient

proximate cause of the loss, was at least an indirect cause of

the loss bringing it under the policy’s “anti-concurrent cause”

exclusion, and that the policy’s “anti-concurrent causation”

clause is enforceable under New Hampshire law. Accordingly, the

Club’s motion for summary judgment is DENIED, Harleysville’s

motion for summary judgment is GRANTED, all other motions are

36 DENIED AS MOOT. The clerk shall enter judgment accordingly and

close the case.

SO ORDERED.

K /^tff6™^ eph N. Laplante ted States District Judge

Dated: June 2 4 , 2008

cc: Timothy G. Kerrigan, Esq. Andrew J. Piela, Esq. Lawrence A . Dugan, Esq. Ralph Suozzo, Esq.

37