BLANCHARD, J.
¶ 1 Patrick Connors alleges that a foundry where he had been employed negligently allowed bacteria to be dispersed or released into the air, that he inhaled some of these bacteria while visiting the foundry, and that this resulted in illness causing bodily injury. Connors filed this direct action against the foundry's insurer. The circuit court granted summary judgment to the insurer after concluding that a pollution exclusion provision contained in the foundry's insurance policy bars coverage for Connors' alleged injuries. The court concluded that the bacteria allegedly dispersed or released by the foundry and inhaled by Connors were "contaminants," which are defined to be "pollutants" under the pollution exclusion.
¶ 2 We conclude that the pollution exclusion in the foundry's policy, which is considerably more detailed than the standard pollution exclusion in many commercial general liability policies, is ambiguous on the question of whether the bacteria are "pollutants" in the context of the occurrence alleged here. The exclusion is ambiguous in this context because the bacteria are not obviously in the nature of the commercial or industrial products or byproducts specified in the pollution exclusion, and therefore a reasonable insured could expect coverage. Accordingly, we reverse the summary judgment and remand for further proceedings.
¶ 3 In the operative complaint, Connors makes allegations that include the following. Connors had been an employee of Grede Foundries of Reedsburg, Wisconsin, and hoped to return to work there, when he became ill and was diagnosed with pneumonia caused by exposure to the bacteria Legionella pneumophila.
¶ 4 As pertinent here, the complaint alleges that the foundry "was negligent in not properly maintaining its cooling towers, and other water sources," which "allowed for the growth of legionella pneumophila"
¶ 5 Those are the limited factual allegations in the complaint. However, the parties do not dispute that the following additional facts apply to Connors' allegations:
See Heinecke v. Aurora Healthcare, Inc., 2013 WI App 133, ¶ 2, 351 Wis.2d 463, 841 N.W.2d 52. As is commonly understood, pneumonia is an infection with potentially serious health effects that inflames air sacs in the lungs. See http://www.mayoclinic.org/diseases-conditions/pneumonia/basics/definition/con-20020032 (last visited Oct. 13, 2015).
¶ 6 The only issue on appeal is whether the pollution exclusion in the foundry's Charter Oak policy applies in the context of an occurrence of the type alleged here. We now address pertinent policy terms.
¶ 7 Under the pollution exclusion, "insurance does not apply to" bodily injuries caused by "pollutants" that have been dispersed or released from a location owned by the insured, here the foundry. More specifically, the policy excludes coverage for "[b]odily injury" "arising out of the actual, alleged or threatened discharge, dispersal, seepage, migration, release or escape of `pollutants,'" "[a]t or from any premises, site or location which is or was at any time owned or occupied by . . . any insured."
¶ 8 "Pollutants" are initially defined in the pollution exclusion to mean "any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapor, soot, fumes, acids, alkalis, chemicals and waste." "Waste includes materials to be recycled, reconditioned or reclaimed."
¶ 9 Up to this point in our summary, the Charter Oak policy pollution exclusion is identical to pertinent language contained in the standard commercial general liability pollution exclusion, which we will refer to as "the standard pollution exclusion."
¶ 10 However, an endorsement to the policy, entitled "Indiana Changes—Definition of Pollutants," replaces the definition of "pollutants" in the standard pollution exclusion with a different, more specific, definition. We will refer to this as "the endorsement." Charter Oak does not dispute that the endorsement changes the definition of "pollutants" in the policy here. Therefore, consistent with discussion by the parties, when we refer to "the pollution exclusion" at issue here we are referring to the exclusion as modified by the definition of "pollutants" contained in the endorsement.
¶ 11 The endorsement begins with the same language from the standard pollution exclusion defining "pollutants" to mean "any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapor, soot, fumes, acids, alkalis, chemicals and waste." However, the endorsement, but not the standard pollution exclusion, then lists four categories of examples of substances that are defined to be pollutants:
"Pollutants" includes:
We will refer to these four categories collectively as "the included substances categories."
¶ 12 The second set of provisions in the endorsement, but again not in the standard pollution exclusion, clarifies the definition of pollutants in the following respects:
¶ 13 Charter Oak moved the circuit court to dismiss Connors' complaint on the grounds that the pollution exclusion bars coverage, or, in the alternative to bifurcate the proceedings to resolve the coverage issue before resolving liability, if necessary. The court granted Charter Oak's motion to bifurcate the proceedings.
¶ 14 Charter Oak followed this with a motion for summary judgment and supporting evidence, arguing in part that the evidence establishes beyond dispute that "Legionella bacteria are contaminants, and thus `pollutants,' to which the pollution exclusion applies, barring coverage." This motion was accompanied by a report of Michael T. Osterholm, a medical school professor, addressing topics that include the prevalence of Legionella bacteria and the health hazard the bacteria pose when inhaled in mist or vapor.
¶ 15 Connors filed a brief in opposition to the motion for summary judgment, arguing in part that Charter Oak had failed to account for terms in the endorsement that diverge from the standard pollution exclusion, and that when all pertinent language is considered "a reasonable insured. . . would conclude that this particular pollution exclusion applies only to industrial or environmental pollution," not to bacteria in the occurrence alleged here.
¶ 17 The only issue presented on appeal is whether summary judgment should be granted to Charter Oak because it is undisputed that Legionella bacteria, in the context of the allegations here, are "pollutants" under unambiguous terms of the pollution exclusion.
¶ 18 We review a grant of summary judgment independently of the circuit court, applying the same methodology, based on the standards in WIS. STAT. § 802.08 (2013-14).
¶ 19 The following standards guide insurance policy interpretation:
Wilson Mutual, 360 Wis.2d 67, ¶¶ 23-25, 857 N.W.2d 156 (citations omitted).
¶ 20 Courts follow three steps in determining whether insurance coverage exists: (1) examine the allegations in the complaint to determine whether the policy makes an initial grant of coverage for the claim made in the complaint; (2) if the claim triggers a potential grant of coverage, examine whether a policy exclusion precludes coverage for that claim; and (3) if an exclusion precludes coverage, consider whether any exception to the exclusion reinstates coverage. Preisler, 360 Wis.2d 129, ¶ 22, 857 N.W.2d 136.
¶ 21 The first and third steps of coverage analysis are not in dispute. As to the first step, there is no dispute about whether the policy provides an initial grant of coverage, which arose when Connors allegedly suffered "bodily injury" caused by an "occurrence." Under the policy terms, "occurrence" "means an accident, including continuous or repeated exposure to substantially the same general harmful conditions." As to the third step, neither party directs us to an exception that could restore coverage if the pollution exclusion applies.
¶ 22 The dispute involves the second step of coverage analysis, whether the Legionella bacteria were "unambiguously" pollutants "within the policy's definition" at the time of the "occurrence," and are thereby excluded from coverage. See Wilson Mutual, 360 Wis.2d 67, ¶¶ 34, 38, 857 N.W.2d 156. This question "is evaluated from the standpoint of a reasonable insured." See id., ¶ 38.
¶ 23 We first address the suggestion by Charter Oak that we should interpret the pollution exclusion here by following the interpretations of the standard pollution exclusion contained in such cases as Wilson Mutual. See id., ¶¶ 37-38. After explaining why we consider these cases to be distinguishable, we turn to the alternative argument by Charter Oak that, in the view of a reasonable insured, the term "pollutants" in the pollution exclusion unambiguously includes bacteria in the context of an occurrence of the type alleged here.
¶ 24 Charter Oak argues that this court should not "abandon" the "well-developed framework that analyzes whether a substance is a pollutant by determining if it is `universally present and generally harmless.'" This is a reference to a body of case law culminating in Wilson Mutual and Preisler.
¶ 25 The court in Wilson Mutual held that the standard pollution exclusion unambiguously excluded coverage for a claim that liquid cow manure fouled well water after the manure was spread on farm fields as fertilizer, because the manure constituted a pollutant in this context. Wilson Mutual, 360 Wis.2d 67, 857 N.W.2d 156. In a similar vein, in Preisler the court held that the standard pollution exclusion unambiguously excluded coverage for a claim that decomposing septage fouled well water after the septage was spread on farm fields as fertilizer, because the septage constituted a pollutant in this context. Preisler, 360 Wis.2d 129, 857 N.W.2d 136.
¶ 26 Both Wilson Mutual and Preisler considered a series of prior court interpretations
See Wilson Mutual, 360 Wis.2d 67, ¶ 38, 857 N.W.2d 156.
¶ 27 The problem with Charter Oak's approach in this appeal is that the Wilson Mutual test is explicitly based on the interpretation of significantly different policy language than is at issue here.
¶ 28 Charter Oak appears to suggest that all that matters is the language that overlaps between the standard pollution exclusion and the pollution exclusion here. However, each insurance contract must be construed on its own terms. See Acuity v. Chartis Specialty Ins., 2015 WI 28, ¶ 98, 361 Wis.2d 396, 861 N.W.2d 533 ("Slightly different [insurance policy] language can have slightly different meanings."). Courts are to "interpret policy terms not in isolation, but rather in the context of the policy as a whole." Day v. Allstate Indem. Co., 2011 WI 24, ¶ 28, 332 Wis.2d 571, 798 N.W.2d 199; see also Blum v. 1st Auto & Cas. Ins., 2010 WI 78, ¶ 20, 326 Wis.2d 729, 786 N.W.2d 78 (like other contracts, insurance policies "are to be read as a whole" and "a single clause or sentence" may not "capture the essence of
¶ 29 The endorsement provides a significantly different definition of "pollutants" that cannot be treated as surplusage. Indeed, as discussed below, the additional language provides a new definition of pollutants, under which it is ambiguous whether the bacteria alleged to have been inhaled are obviously in the nature of the commercial or industrial products or byproducts specified in the pollution exclusion. For this reason, we reject Charter Oak's repeated assertion that the standard pollution exclusion is "identical" to the pollution exclusion here, and that we should therefore apply the Wilson Mutual test to the different policy language here, without regard for the more detailed definition in the endorsement.
¶ 30 We reject on similar grounds three related arguments that Charter Oak makes, based on Wisconsin appellate court precedent.
¶ 31 We also note that, even when one interprets the standard pollution exclusion, Wilson Mutual rejected a categorical approach, under which for example all bacteria in all contexts, or even all Legionella bacteria, are "pollutants." See Wilson Mutual, 360 Wis.2d 67, ¶¶ 39-43, 857 N.W.2d 156 (in interpreting the pertinent policy language, court places emphasis on the context of the type of occurrence at issue, without regard to whether the substance might be a "pollutant" in other contexts). Reading Landshire or Heinecke as adopting such a categorical approach would conflict with the rationales shared by Wilson Mutual and Preisler.
¶ 32 For essentially the same reasons, we reject Charter Oak's third related argument. Charter Oak apparently invites us to conclude that Wilson Mutual and Preisler stand for the broad proposition that all "organic substances" alleged to have caused a bodily injury "satisfy the definition of a pollutant" in any pollution exclusion provision in any insurance policy. Again, such a broad argument, if intended, would fail to account for the endorsement
¶ 33 For these reasons, we reject the argument by Charter Oak that we should interpret the pollution exclusion here by following prior case law interpretations of the standard pollution exclusion. This conclusion disposes of extensive arguments that Charter Oak makes on appeal, because Charter Oak repeatedly relies on case law interpreting the standard pollution exclusion.
¶ 34 Turning to the pertinent policy language in this case, we now explain why we conclude that the pollution exclusion language here is ambiguous on the question of whether the bacteria are "pollutants" in the context of the alleged occurrence, and why we reject specific contrary arguments that Charter Oak makes. We begin with the initial terms of the pollution exclusion, which as explained above match the terms of the standard pollution exclusion.
¶ 35 We agree with Charter Oak to the extent that it suggests that, if we were presented only with the language in the standard pollution exclusion, then coverage would be excluded, given the evidence submitted on summary judgment. It is undisputed that Legionella bacteria in aerosolized water droplets would have been pollutants under the standard pollution exclusion at the time the bacteria allegedly infected Connors, because there is no dispute that: (1) it is uncommon for people to inhale Legionella bacteria in aerosolized water droplets and when these bacteria are inhaled they pose a health hazard; and (2) a reasonable insured would view these mist- or vapor-borne bacteria as pollutants. That is, applying the two-part test in Wilson Mutual for the standard pollution exclusion, a reasonable insured would consider the bacteria to be "largely undesirable and not universally present," and that it is a pollutant in this context. See Wilson Mutual, 360 Wis.2d 67, ¶ 38, 857 N.W.2d 156.
¶ 36 There is no question that Legionella bacteria are "largely undesirable" when considered in the context of the occurrence alleged here.
¶ 37 As to the question of whether Legionella bacteria are "universally present in the context of the occurrence" here, the key is whether the Legionella bacteria are present, in any concentration, in the air, which is what would matter under the standard pollution exclusion. Connors does not argue, and could not credibly argue, that Legionella bacteria are "universally present" in the air that people typically inhale. To the contrary, the unrebutted evidence submitted on summary judgment is that Legionella bacteria are rarely present, in any concentration, in the air.
¶ 38 Connors argues that these "airborne" bacteria "remain[] waterborne," because the bacteria are suspended in water mist or vapor when they are allegedly inhaled, and therefore the question is how prevalent they are in water sources, not how prevalent they are in the air people inhale. However, we are to focus on the occurrence at issue, which involves the inhalation of aerosolized bacteria while they are in air, not consumption of the bacteria while they are in bodies or vessels of water, such as when drinking, swimming, or bathing.
¶ 39 Finally, based on these same factors, it is evident that a reasonable insured, who knows the undisputed facts about these bacteria, would view them as contaminants when inhaled as alleged here.
¶ 40 While there would be no coverage if the policy here included the standard pollution exclusion language only, we now explain why the specific, detailed endorsement language makes the difference. We address each set of provisions and explain why we conclude that the pollution exclusion here is ambiguous on the question of whether mist- or vapor-borne "bacteria" fits into any category or type of pollutants referred to in the endorsement. Ambiguity arises because the replacement definition of "pollutants" in the endorsement limits the exclusion by specifying particular types of substances that qualify as pollutants.
¶ 41
¶ 42 We would be on our own in attempting to assign precise meanings to the terms used in the included substances categories, because neither party directs us to commonly accepted or objectively undisputed definitions of any of these terms. To cite only one example, the record provides no guidance, and the parties make no attempt to explain, what a reasonable insured might think a "degradation product" of trichloroethylene is, even an insured with access to objectively undisputed evidence on pertinent topics. We are nonetheless obligated to interpret how the included substances categories would be understood by a reasonable insured, bearing in mind Charter Oak's burden of showing that the exclusion applies. See Day, 332 Wis.2d 571, ¶ 26, 798 N.W.2d 199 (insurer bears the burden of showing that exclusion precludes coverage if the insured carries initial burden of showing an initial grant of coverage).
¶ 43 Without assigning detailed meanings to each term, it is readily apparent that each category involves a type of product or byproduct that would be expected to be used in, or result from the operation of, particular types of commercial or industrial operations: petroleum-based fuels; lubricants; solvents; "manufactured gas plant . . . byproducts;" pesticides; and inorganic chemical metals. As Connors argues,
¶ 44 We do not mean to suggest that aerosolized Legionella bacteria could not possibly be considered byproducts of certain kinds of commercial or industrial operations. We also question Connors' broad assertions that the pollution exclusion does not apply because the bacteria here could not be considered "industrial pollutants," and that each category describes an "industrial chemical." Instead, to repeat, we look at the substances specified in each of the four included substances categories and conclude that there is ambiguity as to whether the bacteria alleged to have infected Connors fit into any of the four categories, or whether the bacteria are suggested by the collective meaning of the four categories.
¶ 45 Weighing in one direction are facts that include the following: no category explicitly states that bacteria cannot be a pollutant; "organic pesticides" might possibly include bacteria, at least as a component of a pesticide; and terms such as "derivatives," "additives," "chemical components," and "degradation products" are sweeping in themselves, which at least in theory could point in the direction of microbial pathogens. Weighing in the other direction are facts that include the following: neither bacteria generally nor mist- or vapor-borne bacteria in particular are specified, and arguably they are not alluded to, in any category; category d. specifies both "organic and inorganic pesticides," but includes only "inorganic contaminants"; and, as to "organic pesticides," aerosolized Legionella bacteria are not readily identifiable as a pesticide.
¶ 46 Charter Oak effectively concedes Connors' contention that the endorsement's list of numerous commercial and industrial products and byproducts limits the reach of the pollution exclusion as we have characterized it, because Charter Oak fails even to attempt to give meaning to the terms contained in the included substances categories in a way that would clearly place mist- or vapor-borne bacteria in any of the included substances categories. That is, Charter Oak gives us no reason to read any category as reasonably including bacteria in general, or bacteria as a mist- or vapor-borne pathogen in particular, or to read the categories as a group as including bacteria. Charter Oak dismissively refers only in passing to the concept that the endorsement includes "several examples of substances," leaving us in the dark as to how Charter Oak believes any term used in the categories should be reasonably understood. Charter Oak merely asserts that it is "irrelevant whether legionella bacteria are `industrial-type pollutants,'" given the Wilson Mutual test, which as we have explained does not apply here.
¶ 47
¶ 48 We conclude that the specifically identified substances provision is susceptible to more than one interpretation by a reasonable insured, and is therefore ambiguous. We first describe what we conclude is the more reasonable construction, then turn to what might be a second possible construction.
¶ 49 The more reasonable construction of the specifically identified substances provision would be to clarify that: (1) substances that are not specifically identified in one of the included substances categories may be irritants or contaminants, so long as they are similar in kind to substances that are listed in one of the included substances categories (e.g., a petroleum-related substance that is not benzene or toluene may be a pollutant); but (2) entirely new categories of substances are not contemplated. This would be a logical clarification, consistent with the interpretive maxim ejusdem generis, in light of the fact that the included substances categories each use some mixture of broad definitions (e.g., "petroleum derivatives," "inorganic contaminants") and specific identifications or descriptions (e.g., "benzene," "arsenic"). See State v. Popenhagen, 2008 WI 55, ¶¶ 46-48, 309 Wis.2d 601, 749 N.W.2d 611 (general word preceded or followed by related, specific words may be construed to embrace only items similar in nature to those enumerated).
¶ 50 Charter Oak argues that ejusdem generis can play no interpretative role here. However, Charter Oak bases this argument primarily on Wisconsin appellate court interpretations of the standard pollution exclusion. As we have explained, this argument is off topic because the pollution exclusion here is different.
¶ 51 We observe that the final phrase in the specifically identified substances provision—"such as waste from manufacturing operations"—supports our conclusion that one reasonable meaning of the provision is that, in order to be categorized as a pollutant, a substance must be similar in kind to substances listed in the included substances categories. Given the substances specified in the included substances categories, it signals an apparent ejusdem generis intent to single out "waste from manufacturing operations" as the only example. And, "waste from manufacturing operations" is not a phrase that unambiguously includes bacteria dispersed or released in mist or vapor from cooling towers.
¶ 52 In addition, this first reasonable interpretation would give meaning to the next two provisions in the second set of endorsement provisions, which we address below in ¶ 54. Read together, these two provisions suggest that the types of pollutants at issue are limited to substances of a type that fall within the included substances categories.
¶ 53 Charter Oak appears to rely on a second possible construction of the specifically identified substances provision, namely, that this provision nullifies any limiting effect of the included substances categories. Under this interpretation, all that matters is whether a substance is a pollutant under the broad definition found in the standard pollution exclusion (i.e., the first part of the pollution exclusion here). At least that is what we take Charter Oak to mean when it argues that the specifically identified substances provision "informs the insured that substances other than those specifically identified will be considered pollutants if they fall within the policy's definition of pollutants." While this may be one possible interpretation, it is the less reasonable one, because it would appear to treat the included substances
¶ 54
¶ 55
Bethke v. Auto-Owners Ins. Co., 2013 WI 16, ¶ 20, 345 Wis.2d 533, 825 N.W.2d 482. We consider this provision to be at least ambiguous, and apparently contrary to established Wisconsin law, and on this basis reject Charter Oak's unsupported assertion that this provision supports Charter Oak's argument that the pollution exclusion applies here.
¶ 56 Charter Oak asserts that it is "key" that "both the policy and the endorsement begin with the same language, defining `pollutants' as `any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapor, soot, fumes, acids, alkalis, chemical and waste." Charter Oak does not develop this assertion into an argument. In any case, we fail to see how it adds anything to Charter Oak's other arguments that this language is used to begin the definition or that it is found in both the body of the policy and the endorsement. As we have explained, the definition in the endorsement begins with broad language that is then significantly modified, notably through the included substances categories.
¶ 57 In sum, while coverage would not be available if the policy here contained the standard pollution exclusion, one reasonable
¶ 58 For these reasons, we reverse the summary judgment and remand for further proceedings consistent with this opinion.
Judgment reversed and cause remanded for further proceedings.
Wilson Mut. Ins. v. Falk, 2014 WI 136, ¶ 37, 360 Wis.2d 67, 857 N.W.2d 156; see also Preisler, 360 Wis.2d 129, ¶¶ 4, 13, 857 N.W.2d 136 (reviewing "a pollution exclusion clause commonly found in commercial general liability (CGL) policies" that excludes from coverage "harm `arising out of the actual, alleged, or threatened discharge, dispersal, seepage, migration, release or escape of "pollutants". . . .'" and defines "pollutants" as "any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapor, soot, fumes, acids, alkalis, chemicals and waste. Waste includes materials to be recycled, reconditioned or reclaimed.").