Full opinion text
MEMORANDUM OPINION AND ORDER JENKINS, District Judge. This matter is now before this court on cross-motions for summary judgment filed by plaintiff Quaker State Minit-Lube, Inc. (“Quaker State”), and defendants Liberty-Mutual Insurance Company (“Liberty Mutual”), Fireman’s Fund Insurance Company, American Insurance Company, National Surety Corporation (collectively “Fireman’s Fund”) and Unigard Insurance Company (“Unigard”). The parties have submitted a series of lengthy memoranda, supported by voluminous appendices of exhibits, documents, excerpts of deposition testimony, and an array of unpublished court decisions, articles and texts. The motions were heard on March 29, 1993. Since the hearing, the parties’ moving papers have been augmented by a series of supplemental citations and copies of cases forwarded by counsel. The Court has reviewed and considered the materials submitted and the arguments made by counsel, and now rules as follows: FACTUAL BACKGROUND AND PROCEDURAL HISTORY This action arises out of the efforts of business, firmly prodded by public mandate and administrative enforcement action, to attempt to clean up the mess left by an industrial enterprise, now defunct, which engaged in the discharge of contaminated oils and toxic chemicals upon land. It arises as well out of the efforts of insurance carriers, who profit by reason of their paid-for promises to assume the risks of others, to write those promises as narrowly as possible. It is a case about cost, about consequence, and about who will, as a matter of public policy, ultimately bear the economic burden of the wrongful conduct of others, now defunct. This action concerns a 6.6-acre industrial oil refining facility located at 1628 North Chicago Street, Salt Lake City, Utah, which until its closure in 1988, was operated purportedly for the purposes of re-refining and recovery of used automobile and industrial oils. The facility was originally owned and operated as an oil refinery by O.C. Allen Oil Company from 1953 to 1968. In 1968, Flinco, Inc. purchased and began operating the facility. In 1978, it was purchased by Axel Johnson, Inc. and was operated by Ekotek, Inc., a Delaware corporate subsidiary of Axel Johnson, Inc. See EPA Administrative Order on Consent for Emergency Surface Removal, dated August 1, 1989 (Docket No. CERCLA VIII-89-25), at 4 (annexed as Exhibit 4 to the Memorandum of Defendants Fireman’s Fund Insurance Company, American Insurance Company and National Surety Corporation in Support of Motion for Summary Judgment, dated January 15, 1993 (hereinafter “Fireman’s Fund Mem.”)). Following the purchase of the property by Steven Self and Steven Miller in 1981, Ekotek, Inc., a Utah corporation (“Ekotek”), operated the facility until its bankruptcy in 1987. The facility was last operated by an entity known as Petrochem Recycling Corporation, which also purported to engage in the oil recovery/re-refining and recycling business, until February 1988, when all operations ceased. Id. Quaker State owns and operates a series of “convenience automobile service centers,” which provide simple vehicle maintenance services, including engine oil changes. Beginning in 1977 and continuing through April of 1985, Quaker State sold drain oil collected at its service centers to Ekotek. In the ordinary course of dealing between Quaker State and Ekotek, Ekotek trucks would collect drain oil from storage tanks located at the Quaker State service centers and transport it to the Ekotek re-refining facility where it would be transferred to large storage tanks for later processing. See “Statement of Facts,” Plaintiff’s Memorandum Brief in Support of Motion for Partial Summary Judgment, dated January 15, 1993 (“Quaker State Mem.”), at ¶¶ 5-6. Prior to the commencement of remedial action by the United States Environmental Protection Agency (“EPA”) at the 1628 North Chicago Street site (hereinafter referred to as the “Ekotek Site”), it was observed that an estimated 500,000 gallons of liquid “containing varying concentrations of hazardous substances” was held in approximately 60 above-ground storage tanks ranging in size from 2,900 to 87,000 gallons, including twelve 20,000-gallon tanks located on the site, along with another 475 drums and approximately 1,500 smaller containers found in five warehouse buildings. See EPA Administrative Order on Consent for Emergency Surface Removal (Exhibit 4 to the Fireman’s Fund Mem.), at 6. “Approximately 200,000 gallons of flammable liquids ... [we]re contained in 32 tanks and 69 drums.” Id. at 9. Also found on the Ekotek Site were three surface impoundment areas, “numerous piles and pits of waste material,” underground tanks and an underground drain field. “EPA Findings of Fact” at ¶ 11, EPA Administrative Order on Consent for Remedial Investigation/Feasibility Study, dated July 10, 1992, (Docket No. CERCLA (106) VIII-92-21) (annexed as Exhibit 5 to the Fireman’s Fund Mem.), at 4. Contaminants associated with these on-site sources include a wide range of organic substances such as chlorinated solvents and other volatile organic compounds (acetone, vinyl chloride, 1, 1-dichloroethane, 1, 1, 1-trichloroethane), polynuclear aromatic hydrocarbons (2-methylnapthalene), phthalates, pesticides (chlordane, endrin, 4, 4-DDE, 4, 4-DDT) PCBs (Aroclor 1260), dioxin (2, 3, 7, 8-TCDD) and furans. Arsenic, chromium, lead, and mercury are also present in on-site primary and secondary sources. Id. Quaker State’s initial summary judgment memorandum details a number of incidents in which oil or toxic materials (such as acid sludge produced in the re-refining process) were discharged or released upon the land or into the water at the Ekotek Site. See Quaker State Mem. at 3-9, ¶¶ 7-22. [O]n two separate occasions, accidents in connection with loading and unloading of Union Pacific Rail Road ears with used oil led to 12,000 gallons of oil being spilled on the property____ The great majority of this oil went underground in a deep trench designed to collect rainwater runoff from the rail siding at the property; the amount recovered is unknown. [R]efinery accidents over the course of the years contributed very large volumes to the spillage total____ Volumes lost as the result of tank overflows from operator error and spills from trucks resulting from driver error were substantial____ Prior to 1967, acid sludge produced in the rerefining process was discharged into a pit on the property. The pit was covered over in 1967, but no effort was made at the time to remove sludge residue remaining in the pit____ The rerefining process used clay as a filtration substance to remove impurities from the oil. Oil and clay were mixed together, and the oil was then squeezed out of the clay in a rotary vacuum filter. The spent clay was accumulated on the ground in the refinery area and periodically carted to landfills. Witnesses differed on the amount of oil remaining in the spent clay and whether oil leaked out of the pile____ Id. at 10-11, 13, 15 (citations omitted). Deposition testimony of former refinery employees related numerous incidents of the discharge of oil and other materials onto the Ekotek Site property: James Blaser recalls that in early 1985 there was a large spill when a manhole was left off a tank and product flowed from the tank all the way to the west end of the site. He also recalls two or three incidents when tank 52 accidently overflowed approximately 500 gallons caused by water in the hot oil resulting in the oil foaming out the top of the tank____ Jim Blaser also states that he does not believe that there was any year when there wasn’t an accidental spill/mishap at the site____ Scott Adair recalls that in 1980 or 1981 there were times when there were run-overs in the processing area large enough that they went across the street____ He also recalls that there were occurrences between 1978 and 1983, happening once or twice a year, of overflows during the loading of acid sludge onto trucks. These spills, on occasions, were large enough that they flowed down the road and would have involved 500 to 2,000 gallons of oil and acid mixture.... Alex Bloomfield recalls that there were significant tank spills in each and every year he was at Ekotek, 1978 through 1985.... Id. at ¶¶ 19-21 (citations omitted); cf. Quaker State Minib-Lube’s Response in Opposition to Defendants’ Motions for Summary Judgment, dated February 26, 1993 (“Quaker State Opp.Mem.”) at 3-42; Quaker State Minib-Lube’s Reply Memorandum in Further Support of Motion for Partial Summary Judgment, dated March 10, 1993 (“Quaker State Reply Mem.”) at 3-11, ¶¶2-16. The defendants’ summary judgment memoranda set forth additional facts detailing similar events: Scott Adair who started work in 1978 testified ... that during cold weather people draining sludge from a tank to a truck would get cold and leave to go inside and allow the sludge to drain into the truck and sludge could overflow the truck.... He testified that ... “when all of the sludge drained out then it was just thin oil, acid treated oil ... and it would fill that semi truck up in just about a minute or two.” ... It would then overflow “and it would be a river running along that roadway.” Joint Memorandum of Defendants in Opposition to Plaintiffs Motion for Partial Summary Judgment, dated February 26, 1993 (“DefsJoint Mem.”) at 6-7, ¶ 4(g). Ekotek’s wastewater treatment system was fraught with a history of oil[-]eontaminated effluent discharged into the Salt Lake sewer system.... Beginning in November, 1980, acid sludge from the oil dehydration process was discharged directly on the ground in a large, unlined, earthen pit north of the plant’s Administration building____ Over a period of years, sludge was hauled daily to the acid sludge pit and mixed with lime to neutralize the sludge---- The acid sludge/lime mixture could sit onsite from anywhere up to a one-month period, until it was transported off-site to a landfill facility____ Ekotek continued this cycle of dumping and then disposing of acid sludge until at least 1985____ Sludges remained onsite, however, until removal procedures were instituted by the EPA in 1988____ The plate and frame filters would regularly “squirt oil all over the place” in the course of day-to-day operations____ Former process operators testified that the inferior and outdated equipment used to recycle the waste oil contributed heavily to the persistent discharge of contaminants onto the site____ Broken or leaking process equipment was not repaired or replaced unless absolutely necessary____ In fact, leaking pumps were a facet of everyday operations at the plant____ Plant operators and supervisors have testified at length concerning the ongoing failure of [facility] personnel from 1967 to 1988 to keep the site clean by failing to: empty buckets full of oil; pump oil out of the catch basins designed to contain runoff and rainwater; and clean up spilled oil and oil leaking onto the ground from trucks, pumps and plant machinery____ Oily water regularly and routinely ran over from tanks, valves, dump trucks, catch basins and earthen berms____ Back-ups of oily water in the east tank area and the wrecking yard to the extreme far north of the site occurred systematically.... [T]wo large retention areas in the northwest portion of the Ekotek site experienced frequent overflows of oil that contaminated the soil____ Fireman’s Fund Mem. at 9-18, ¶¶ 18, 20-21, 28, 35-36, 38-39 (citations omitted). Accord, Defendant Liberty Mutual Insurance Company’s Memorandum in Support of Motion for Summary Judgment, dated January 14, 1992 (“Liberty Mutual Mem.”) at 11-27, ¶¶ 8-47. It is uncontroverted that as a consequence of numerous such incidents occurring during the years the Ekotek facility was in operation, the soil at the Ekotek Site as well as surface water and ground water on and near the Ekotek Site have become significantly contaminated with oil and other toxic substances: The volatile organic data indicates that soils and groundwater are contaminated with both petroleum and chlorinated volatile compounds.... The oil and grease analysis likewise indicates a pattern of contamination with oil that extends site[-]wide and beyond____ Soils, liquids and sludges over much of the site are contaminated with petroleum and petroleum[-]derived volatile and semi-volatile compounds. The soil, sludges and groundwater at the site are also contaminated with compounds not typically associated with petroleum hydrocarbons including chlorinated volatile organic compounds (especially TCE and PCE) and PCBs (Aroclor 1260)____ Office of Emergency & Remedial Response, U.S. Environmental Protection Agency, Final Report for Petrochem Site Investigation, Salt Lake City, Utah (annexed as Exhibit 10 to the Fireman’s Fund Mem), at §§ 4.0, 5.0. “Contaminated soils and sediments, as well as contaminants within the shallow ground water aquifer, remain at the Site.” EPA Region VIII, Superfund Program Fact Sheet: Petrochem Recycling Corp./Ekotek, Inc. Superfund Site, at 1 (September 1992). See Quaker State Mem. at 4, ¶¶ 8-9. Acting pursuant to the Comprehensive Environmental Response Compensation and Liability Act of 1980 (“CERCLA”), 42 U.S.C. §§ 9601-9657, the EPA commenced “response” activities involving the Ekotek Site beginning in 1988. The EPA has taken the position that the Ekotek Site qualifies as a CERCLA “facility” (§ 9601(a)) because of its contamination with “hazardous substances” (§ 9601(a)(14)) and that “responsible parties” (§ 9607(a)), including Quaker State, are liable for “response” costs incurred through cleanup efforts at the Ekotek site. See EPA Administrative Order on Consent for Remedial Investigation/Feasibility Study, dated July 10, 1992, (Docket No. CERCLA (106) VIII-92-21) (annexed as Exhibit 5 to the Fireman’s Fund Mem.), at 8. By February 1992, the EPA had formally identified 470 entities as “potentially responsible parties” (“PRP”s) in connection with the Ekotek Site, 129 of which had agreed to participate in response activities in accordance with the EPA’s Administrative Order on Consent (AOC). EPA Region VIII, Superfund Program Fact Sheet: Petrochem Recycling Corp./Ekotek, Inc. Superfund Site 1 (September 1992). Responsible parties may be held strictly, jointly and severally liable under CERCLA for the costs of response and clean-up. See generally, Barr, CERCLA Made Simple: An Analysis of Cases Under the Comprehensive Environmental Response, Compensation and Liability Act of 1980, 45 BusXiAW. 923, 968-83 (1990). Further, pursuant to section 105 of CERCLA, 42 U.S.C. § 9605, the Ekotek Site was proposed for placement on the EPA’s National Priorities List, or “Superfund List,” on July 29, 1991, (see 55 Fed.Reg. 35844 (1991)); “Petrochem Recycling Corp./Ekotek, Inc.” has since been placed on that list, along with seven other “Superfund” sites found in Utah. 40 C.F.R. Part 300, App. B, at 208 (1993). Quaker State and a number of other businesses identified by EPA as potentially responsible parties under CERCLA formed the Ekotek Site Remediation Committee, which pursuant to the Administrative Order on Consent entered into with EPA, has funded clean-up activities at the Ekotek Site. By affidavit, Quaker State indicates that as of January 15, 1993, the Committee had expended approximately $10,000,000.00 to pay costs incurred in removing waste materials, equipment and machinery from the Ekotek Site. Affidavit of Shane Smoot, dated January 15, 1993, at ¶ 7. Quaker State estimates that the cost of further investigation, planning and total cleanup of the Ekotek site will exceed $60,000,000.00. Id. at ¶¶8-9. At the times pertinent to this action, Quaker State has maintained an array of policies of comprehensive general liability insurance and primary garage liability insurance, as well as excess, or umbrella policies, all purchased from one or more of the defendants. Not surprisingly, Quaker State is looking to its liability insurance coverage to defray Quaker State’s share of the actual and expected costs of cleaning up the Ekotek Site, for which Quaker State assumes liability as a “responsible party” under CERCLA. Quaker State suggests that it had no actual knowledge of the activities at the Ekotek Site which resulted in the contamination. Quaker State Mem. at 9, ¶23. Since it has not engaged in the knowing or intentional pollution of the Ekotek Site, Quaker State argues, its strict statutory liability under CERCLA for response and clean-up costs cannot fall within any exclusion or limitation of coverage under the insurance policies purchased from the defendants which otherwise would operate to deny coverage arising from deliberate contamination of the environment. Id. at 34-57. Quaker State commenced this action to enforce defendants’ obligations to indemnify and defend Quaker State under the terms of their policies. The summary judgment motions address a series of issues concerning the existence and extent of insurance coverage available to Quaker State: (1) whether “damages” covered under defendants’ policies extends to costs of the kind incurred by Quaker State in connection with the cleanup of the Ekotek Site (see Part II, infra); (2) whether the hazardous waste contamination at the Ekotek Site resulted from an “occurrence” within the meaning of defendants’ policies (see Part III, infra); (3) whether the defendants’ duty to defend under their liability policies is triggered by administrative action taken by the United States Environmental Protection Agency (EPA), where no civil action, or lawsuit, has been filed in court (see Part IV, infra); (4) whether coverage is excluded by the standard pollution exclusion clause in defendants’ comprehensive general liability policies, or whether the discharge of drain oil, acid sludge and other materials by Ekotek comes within the “sudden and accidental” exception to that exclusion (see Part V, infra); and (5) whether defendants’ “garage policies” extend coverage to liability for cleanup costs arising out of discharge of waste oil at third-party-owned sites outside of the garage premises (see Part VI, infra ). I. SUMMARY JUDGMENT. The Tenth Circuit summarized the principles governing summary judgment in a recent case, Jensen v. Kimble, 1 F.3d 1073 (10th Cir.1993): Rule 56(c) permits summary judgment when “there is no genuine issue as to any material fact and ... the moving party is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c). In analyzing whether summary judgment is appropriate, we must view the evidence in the light most favorable to the non-moving party. Deepwater Invs., Ltd. v. Jackson Hole Ski Corp., 938 F.2d 1105, 1110 (10th Cir.1991). Id. at 1076. “Of course, a party seeking summary judgment always bears the initial responsibility of informing the district court of the basis for its motion, and identifying those portions of ‘the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,’ which it believes demonstrate the absence of a genuine issue of material fact.” Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 2552, 91 L.Ed.2d 265 (1986). The initial burden of production imposed by Rule 56(c) requires the moving party to make a prima facie showing that it is entitled to summary judgment. 10A Charles A. Wright & Arthur R. Miller & Mary K. Kane, Federal Practice and Procedure § 2727, at 121 (2d ed. 1983). If the moving party will bear the burden of persuasion at trial on the claim addressed by the motion, “that party must support its motion with credible evidence — using any of the materials specified in Rule 56(c) — that would entitle it to a directed verdict if not controverted at trial.” Celotex, 477 U.S. at 331, 106 S.Ct. at 2556 (Brennan, J., dissenting). Where the nonmoving party bears the burden of persuasion at trial on the claim or issue addressed by the motion, Celotex instructs that the moving party may meet its initial burden “by ‘showing’ — that is, pointing out to the district court — that there is an absence of evidence to support the nonmoving party’s case.” 477 U.S. at 325, 106 S.Ct. at 2554 (emphasis added); Bacchus Indus., Inc. v. Arvin Indus., Inc., 939 F.2d 887, 891 (10th Cir.1991). “[W]e find no express or implied requirement in Rule 56,” the Court explained, “that the moving party support its motion with affidavits or other similar materials negating the opponent’s claim.” 477 U.S. at 323, 106 S.Ct. at 2553 (emphasis in original). Once the moving party has met its initial burden, “the burden shifts back to the non-moving party to show that there is a genuine issue of material fact. Bacchus, 939 F.2d at 891. To discharge its burden, the nonmoving party must “go beyond the pleadings and by her own affidavits, or by the ‘depositions, answers to interrogatories, and admissions on file,’ designate ‘specific facts showing that there is a genuine issue for trial.’ ” Celotex, 477 U.S. at 324, 106 S.Ct. at 2553 (quoting Fed.R.Civ.P. 56(e)).” Jensen v. Kimble, 1 F.3d at 1077. As the Court explained in Celotex: In our view, the plain language of Rule 56(c) mandates the entry of summary judgment, after adequate time for discovery and upon motion, against a party who fails to make a showing sufficient to establish the existence of an element essential to that party’s case, and on which that party will bear the burden of proof at trial. 477 U.S. at 322, 106 S.Ct. at 2552. Under Utah law, “[t]he interpretation of an unambiguous contract is a question of law to be determined by the court and may be decided on summary judgment. As a general rule, “The construction of an insurance policy is a matter of law.” Grimes v. Swaim, 971 F.2d 622, 623 (10th Cir.1992). If the policy language is clear and unambiguous, the court must construe it according to its plain and ordinary meaning.” Utah Power & Light Co. v. Federal Ins. Co., 983 F.2d 1549, 1553 (10th Cir.1993) (citations omitted). Treated as a matter of law, the construction of an insurance contract can be resolved by the Court in the context of a motion for summary judgment. Adams-Arapahoe Joint School District v. Continental Ins. Co., 891 F.2d 772, 774 (10th Cir. 1989). Contract construction remains a question of law even though the parties may disagree about the meaning of the contract, or even though one party may claim that the contract is ambiguous. Gomez v. American Electrical Power Service Corp., 726 F.2d 649, 651-52 (10th Cir.1984). II. RESPONSE AND REMEDIATION COSTS UNDER CERCLA AS “DAMAGES” UNDER DEFENDANTS’ CGL POLICIES. Several of the insurance policies at issue in this case are comprehensive general liability, or CGL policies. CGL policies “are standard insurance policies developed by insurance industry trade associations, and these policies are the primary form of commercial insurance coverage obtained by businesses throughout the country.” Dimmitt Chevrolet, Inc. v. Southeastern Fidelity Ins. Corp., 636 So .2d 700, 702 (1993). In their policies, the defendant insurers promise to “pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of ... property damage____” “Damages” is not specially defined by the policies. The defendants now suggest that their promise to pay “damages” does not extend to Quaker State’s share of the environmental clean-up costs incurred at the Ekotek Site because those costs are not true legal “damages.” Conversely, Quaker State asserts that the payment EPA seeks under CERCLA, viz., the costs of restoring real property to its natural state, has long been recognized as a measure of legal damages, citing among other authorities, Ault v. Dubois, 739 P.2d 1117, 1120 (Utah Ct.App.1987), and Thorsen v. Johnson, 745 P.2d 1243, 1244-45 n. 1 (Utah 1987). Policy terms, Quaker State asserts, should be given their plain and commonly understood meaning in ordinary usage, and “damages” as plainly and ordinarily understood, encompasses CERCLA remediation costs as compensation or “estimated reparation in money.” Quaker State Mem. at 59-60 (quoting Webster’s Third New International Dictionary 571 (1986)). No distinction should be made, Quaker State contends, between common-law damages and payments made in compliance with an equitable remedy. Id. at 61 (citing AIU Ins. Co. v. Superior Court, 51 Cal.3d 807, 274 Cal.Rptr. 820, 826 nn. 11, 12, 799 P.2d 1253, 1259 nn. 11, 12 (1990)). See also Aetna Casualty & Sur. Co. v. Pintlar Corp., 948 F.2d 1507, 1512-13 & n. 4 (9th Cir.1991). Nor should any distinction be made between judicial and administrative remedies. Id. at 62. Quaker State cites to one unpublished Utah district court opinion and a string of state and federal cases from other jurisdictions standing for the proposition that “damages” for the purpose of CGL coverage includes environmental response costs. See also Fireman’s Fund Ins. Co. v. Ex-Cell-O Corp., 662 F.Supp. 71, 75 (E.D.Mich.1987) (“‘damages’ include money spent to clean up environmental contamination”). Defendants rely on cases such as Continental Ins. Co. v. Northeastern Pharmaceutical & Chem. Co., 842 F.2d 977 (8th Cir.) (en banc) (applying Missouri law), cert. denied, 488 U.S. 821, 109 S.Ct. 66, 102 L.Ed.2d 43 (1988) (“NEPACCO ”), and Maryland Casualty Co. v. Armco, Inc., 822 F.2d 1348 (4th Cir.1987) (applying Maryland law), cert. denied, 484 U.S. 1008, 108 S.Ct. 703, 98 L.Ed.2d 654 (1988) (“Armco ”), as expressing the “better reasoned approach” construing “damages” as distinguishing legal claims for monetary damages from costs incurred in response to equitable remedies, including environmental response costs under CERCLA. See “Defs. Joint Mem.” at 32 & n. 16. In Armco, the Fourth Circuit postulated that “[b]lack letter insurance law holds that claims for equitable relief are not claims for damages under liability insurance contracts,” and opined that it would be “a great step, and a dangerous one, for courts to begin to construe insurance policies to encompass costs of compliance with injunctive and reimbursement relief.” 822 F.2d at 1351, 1353. As used “in the insurance context,” the Fourth Circuit concluded, the term “damages” means “damages in the legal sense.” Id. at 1354. In NEPACCO, the Eighth Circuit followed Armco: In the insurance context, ... the term “damages” is not ambiguous, and the plain meaning of the term “damages” as used in the insurance context refers to legal damages and does not include equitable monetary relief. See Maryland Casualty Co. v. Armco, Inc., 822 F.2d [1348] at 1352 [ (4th Cir.1987) ]. The CGL policies require Continental to “pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of ... property damage to which this insurance applies caused by an occurrence.” (Emphasis added.) “The obligation of the insurer to pay is limited to ‘damages,’ a word which has an accepted technical meaning in law.” [Aetna Casualty & Surety Co. v.] Hanna, 224 F.2d [499] at 503 [ (5th Cir.1955) ]. Although not defined in the CGL policies, “[t]he word ‘damages’ is not ambiguous in the insurance context. Black letter insurance law holds that claims for equitable relief are not claims for ‘damages’ under liability insurance contracts.” 842 F.2d at 986 (quoting Maryland Casualty Co. v. Armco, Inc., 643 F.Supp. 430, 432 (D.Md.1986)). Further, defendants point to the separate enumeration of “all costs of removal or remedial action” and “any other necessary costs of response” and “damages for injury to, destruction of, or loss of natural resources ...” found in § 107(a)(4)(A), (B) and (C) of CERCLA as evidencing purposeful distinction between “damages” and other CERCLA remedies. Defs.Joint.Mem. at 36-37. In NEPACCO, the Eighth Circuit took pains to differentiate the various remedies available under CERCLA, explaining that “[t]he type of relief sought is critical to the insured and the insurer, because under the CGL policies the insurer is liable only for legal damages, not for equitable monetary relief, such as cleanup costs.” 842 F.2d at 987. Defendants also contend that environmental response costs cannot be treated as “damages” under Utah law, particularly where those costs exceed the diminution in the market value of the property caused by the contamination. They cite to two cases relied upon by Quaker State, Ault v. Dubois, 739 P.2d 1117 (Utah Ct.App.1987), and Thorsen v. Johnson, 745 P.2d 1243 (Utah 1987), as standing for the proposition that “costs of restoration may not be used as a measure of damages if such costs exceed the loss of value (ie., damages) sustained by the injured party.” Defs. Joint Mem. at 38-39. The rule of Ault and Thorsen has been summarized as follows: Generally, the measure of damages for tortious injury to real property is the difference between the value of the property immediately before and immediately after the injury. Thorsen v. Johnson, 745 P.2d 1243, 1244-45 (Utah 1987); Ault v. Dubois, 739 P.2d 1117, 1120 (Utah App.1987); see also Pitts v. Pine Meadow Ranch, Inc., 589 P.2d 767, 769 (Utah 1978). An alternative measure is the cost of restoration, provided that restoration costs do not exceed diminution in value. See Thorsen, 745 P.2d at 1244-45 n. 1; Ault, 739 P.2d at 1120. Henderson v. For-Shor Co., 757 P.2d 465, 471 (Utah Ct.App.1988). Of course, Ault and Thorsen are speaking of damages in the legal sense. The argument advanced by the defendants simply begs the question whether “damages” as used in their CGL policies, read in light of Utah law, refers to damages in that sense. Moreover, in relying on Ault and Thorsen, defendants logically may be forced to concede that at least up to the point that the diminution in value of the property is exceeded, costs of restoration recoverable by EPA under CERCLA are legal damages within their own theory of coverage. See, e.g., Federal Insurance Company v. Susquehanna Broadcasting Company, 727 F.Supp. 169 (M.D.Pa.1989); T. Myers, Insurance Coverage for CERCLA Cleanup Costs: Resolving the Intercircuit Conflict, 45 Ark. L.Rev. 747 (1992). In NEPACCO, the Eighth Circuit noted that “from the viewpoint of the lay insured, the term ‘damages’ could reasonably include all monetary claims, whether such claims are described as damages, expenses, costs, or losses.” 842 F.2d at 985. From this language, Quaker State argues that NEPACCO, if anything, indicates the Utah Supreme Court would not adopt the narrow, technical reading of “damages” urged by the defendants. Quaker State Opp.Mem. at 85. Quaker State points to cases such as LDS Hospital v. Capitol Life Ins. Co., 765 P.2d 857 (Utah 1988), which in considering the use of the term “accident” in policy language, said: “[W]e are guided by the principle that it is the common understanding of the term which must be used and not its technical meanings. The insurance company may, of course, insert in its policy any definition of ‘accident’ it chooses but, in the absence of doing so, it must accept the common understanding of the term by the ordinary member of the purchasing public.” 765 P.2d at 861 (quoting Harbeintner v. Crown Life Ins. Co., 46 Or.App. 579, 612 P.2d 334, 335 (1980)). Thus, Quaker State contends, absent a specific definition in the defendants’ CGL policies, “damages” should be construed according to “the common understanding of the term by the ordinary member of the purchasing public” and that as NEPACCO suggests, “damages” would “include all monetary claims, whether such claims are described as damages, expenses, costs, or losses,” including CERCLA remediation expenses. In Independent Petrochemical Corp. v. Aetna Casualty & Sur. Co., 944 F.2d 940 (D.C.Cir.1991), cert. denied, - — U.S. -, 112 S.Ct. 1777, 118 L.Ed.2d 435 (1992), the United States Court of Appeals for the D.C. Circuit departed from NEPACCO’s reading of Missouri law on the issue of covered “damages.” Noting that the Missouri cases, inter alia, mandate that “[m]eaning of words or terms in an insurance contract is tested by common understanding,” and that a conflicting technical meaning of a word shall not be applied “unless it plainly appears that the technical meaning is intended,” the D.C. Circuit observed: Our difficulty with NEPACCO is that it fails to apply these Missouri law principles ____ Liability for environmental cleanup costs quite naturally fits this common and ordinary understanding of damages---- [W]ith the exception of NEPACCO, in every case in which the operative state’s rules of insurance contract interpretation required — -as Missouri’s does — resort to the common and ordinary understanding of language, the word “damages” has been construed to cover reimbursement for environmental response costs____ Decisions construing the term differently were apparently governed by state rules of interpretation under which the technical or legal meanings of language controlled. 944 F.2d at 946 (citations omitted). The D.C. Circuit concluded that under Missouri law, “damages” as commonly understood “includes costs the insured is legally obligated to pay ... as reimbursement for [state and federal] activities in remedying environmental harm.” Id. at 947. More recently, Morton International, Inc. v. General Accident Insurance Co. of America, 134 N.J. 1, 629 A.2d 831 (1993), reached the same conclusion, applying New Jersey law: “The clear weight of authority, however, among both federal and state courts adopts the view that the undefined term ‘damages’ in CGL policies should be accorded its plain, non-teehnical meaning, thereby encompassing response costs imposed to remediate environmental damage____” 134 N.J. at 25-26, 629 A.2d at 845 (citations omitted). Utah law approaches the reading of insurance policies as follows: Generally, the interpretation of insurance policy language presents a question of law to be decided by the trial judge using accepted methods of construction. Specifically, the terms of insurance contracts, as well as all contracts, are to be interpreted in accordance with their usually accepted meanings and should be read as a whole, in an attempt to harmonize and give effect to all of the contract provisions. To protect against overreaching insurers and because courts construe contracts against their drafters, ambiguities in the policy are resolved in favor of coverage. Policy language is ambiguous if it is not “ ‘plain to a person of ordinary intelligence and understanding, viewing the matter fairly and reasonably, in accordance with the usual and natural meaning of the words, and in the light of existing circumstances, including the purpose of the policy.’” Nielsen v. O’Reilly, 848 P.2d 664, 665-66 (Utah 1992) (footnote omitted). Accord, United States Fidelity & Guar. Co. v. Sandt, 854 P.2d 519, 522 (Utah 1993) (“ambiguous or uncertain language in an insurance contract that is fairly susceptible to different interpretations should be construed in favor of coverage”). The “usually accepted meanings” of the words in an insurance policy are the ones found in common usage: ‘“If a policy of insurance is clear and unambiguous, the words are to be taken and understood in their plain, ordinary and popular sense, as an average or reasonable person with ordinary understanding would construe them.’ ” Perkins v. Great-West Life Assurance Co., 814 P.2d 1125, 1128-29 (Utah Ct.App.1991) (quoting Draughon v. CUNA Mutual Ins. Soc’y, 771 P.2d 1105, 1108 (Utah App.1989)); accord, United States Fidelity & Guar. Co. v. Sandt, 854 P.2d at 523 (“the language of an insurance contract must be interpreted as an ordinary purchaser of insurance would understand it”). Read in this fashion, legalistic distinctions between “legal” and “equitable” remedies appear to be immaterial. As the Ninth Circuit explained in Aetna Casualty & Sur. Co. v. Pintlar Corp., “Any definition of ‘damages’ which is grounded upon the ancient division between law and equity — such as the definition now proffered by the insurers — would hardly be an ‘ordinary and accepted meaning in the eyes of a reasonably prudent lay person.” 948 F.2d at 1513 (citations omitted). Similarly, in Bausch & Lomb, Inc. v. Utica Mutual Ins. Co., 330 Md. 758, 625 A.2d 1021 (Ct.App.1993), the Court of Appeals of Maryland said: Absent an express provision in the document itself, insurance policy-holders surely do not anticipate that coverage will depend upon the mode of relief, i.e., a cash payment rather than an injunction, sought by an injured party. Policy-holders will, instead, reasonably infer that the insurer’s pledge to pay damages will apply generally to compensatory outlays of various kinds, including expenditures made to comply with administrative orders or formal injunctions. The ordinary person understands “damages” as meaning money paid to make good an insured loss. In this context, environmental response costs fall within that definition. 330 Md. at 782, 625 A.2d at 1033 (footnote omitted). The Maryland court correctly points out that “damages,” as commonly, plainly or ordinarily understood, is not without limits: “The payment of damages — even broadly defined — remains distinct from other expenditures, such as fines, penalties, or assessments, which would not be covered under the CGL policy.” Id. To the same effect, see Independent Petrochemical, 944 F.2d at 947. As of the date of this memorandum opinion, the Utah courts have not yet addressed, in a reported case, the question of CERCLA response costs as “damages.” Nevertheless, following the reasoning in cases such as Independent Petrochemical, Morton, and Bausch & Bomb in light of prior pronouncements by the Utah courts concerning insurance policy language, it appears that the Utah courts would construe “damages” as used in defendants’ CGL policies to encompass CERCLA environmental response costs. To the extent the policies’ reference to “damages” is ambiguous, the ambiguity must be resolved in favor of coverage. As the Utah Supreme Court explained in LDS Hospital v. Capitol Life Ins. Co.: [T]his Court, similar to courts in many jurisdictions, has long subscribed to the view that any ambiguity or uncertainty in the language of an insurance policy must be resolved in favor of coverage. Also, since the policy is drawn by the insurer, ambiguities are construed against that party. One acknowledged rationale underlying this sound determination is the need to afford the insured the protection he or she endeavored to secure by paying premiums ____ 765 P.2d at 858 (footnotes omitted). See United States Fidelity & Guar. Co. v. Sandt, 854 P.2d at 522 (“ambiguous or uncertain language in an insurance contract that is fairly susceptible to different interpretations should be construed in favor of coverage”); American Casualty Co. v. Eagle Star Ins. Co., Ltd., 568 P.2d 731, 734 (Utah 1977) (“[I]f an insurance policy is ambiguous or uncertain, so that it is fairly susceptible to different interpretations, any doubt should be resolved in favor of insurance coverage.” (citations omitted)); Perkins v. Great-West Life Assurance Co., 814 P.2d 1125, 1128 (Utah Ct.App.1991). The proper query concerning ambiguity, as framed by the court in LDS Hospital, is “ Would the meaning [of the language of the insurance contract] be plain to a person of ordinary intelligence and understanding, viewing the matter fairly and reasonably, in accordance with the usual and natural meaning of the words, and in light of existing circumstances, including the purpose of the policy[?].’” Id. at 858-59 (quoting Auto Lease Co. v. Central Mutual Ins. Co., 7 Utah 2d 336, 339, 325 P.2d 264, 266 (1958)). See Dawson v. Dawson, 841 P.2d 749, 751 (Utah Ct.App.1992). Posing that query in the context of this action, the Court is satisfied that the meaning of “damages,” viewed as LDS Hospital instructs, would be “plain to a person of ordinary intelligence and understanding.” Its plain meaning to such a person would not carry the more subtle, fine-tuned legal distinctions now urged herein by the defendant insurers; it would, as NEPACCO suggests, “include all monetary claims, whether such claims are described as damages, expenses, costs, or losses.” 842 F.2d at 985 (emphasis added). If, as Quaker State insists, “an insured is entitled to the broadest protection he could have reasonably understood to be provided by the policy,” (Fuller v. Director of Finance, 694 P.2d 1045, 1047 (Utah 1985)), it follows that Quaker State should be entitled to the broadest protection that reasonably flows from the language of defendants’ CGL policies, including coverage for “damages” in the form of expenses incurred in remedial action undertaken consistent with CERCLA requirements. III. WHETHER RELEASES OF HAZARDOUS WASTE RESULTING IN PROPERTY DAMAGE AT THE EKOTEK SITE CONSTITUTE ONE OR MORE COVERED “OCCURRENCES.” A. “Occurrence.” Defendants submit that under the insurance policies issued to Quaker State, coverage is available only in the event of an “occurrence,” which is defined in the policies as: an accident, including continuous or repeated exposure to conditions, which results in bodily injury or property damage neither expected nor intended from the standpoint of the insured. Fireman’s Fund Mem. at 27 (quoting id. at Exhibit 19, Fireman’s Fund “Master Policy: Minit-Lube” #MXP 5670727 (Sept. 1, 1983—Sept. 1, 1985)). They further contend that “[t]he intentional discharge of contaminants at the Ekotek Site in the regular course of business demonstrates that there was no ‘accident’ and hence, no ‘occurrence,’ ” regardless of whether the property damage was “neither expected nor intended” from Quaker State’s standpoint. Defs. Joint Mem. at 12. Quaker State contends that as defined, “occurrence” excludes coverage “only for bodily injury or property damage actually expected or intended by the insured,” (Quaker State Mem. at 21 (emphasis in original)), and that defendants have failed to carry their burden under Rule 56 to show that coverage is excluded under the definition of “occurrence” because they have failed “to point to any evidence in the record that [Quaker State] expected or intended to cause property damage at the Ekotek [S]ite ...” Quaker State Reply Mem. at 12. “ ‘The occurrence clause provides coverage when the damage was unexpected and unintended, though caused by an intentional act____’ ” Id. at 13 (quoting New Castle County v. Hartford Accident & Indem. Co., 970 F.2d 1267, 1269 (3d Cir.1992)). B. History and Rules of Construction. The adoption of the term “occurrence” to define coverage of a loss represents one step in the continuing evolution of standard CGL insurance policies: After 1966, the term “occurrence” replaced the term “accident,” in the standard CGL policy as the event triggering coverage. The “occurrence” language is generally understood to offer broader coverage than that offered by the former term, “accident.” 11 G. Couch, Couch on Insurance 2d § 44:285 at 437 (1982). The change to occurrence-based coverage was made in response to the need for coverage against damages resulting from continued or repeated exposure____ United States Fidelity & Guar. Co. v. Morrison Grain Co., 734 F.Supp. 437, 443 (D.Kan. 1990); accord, 1 R. Long, The Law of Liability Insurance § 1.21, at 1-88 (1990); 7A J. Appleman, Insurance Law and Practice §§ 4492-93 (Berdal ed. 1979). Adopting the concept of “occurrence” in order to broaden the available coverage was wholly consistent with the overall purpose of a CGL policy: The primary purpose of a comprehensive general liability policy is to provide broad comprehensive insurance. Obviously the very name of the policy suggests the expectation of maximum coverage. Consequently the comprehensive policy has been one of the most preferred by businesses and governmental entities over the years because that policy has provided the broadest coverage available. All risks not expressly excluded are covered, including those not contemplated by either party. See The Applicability of General Liability Insurance to Hazardous Waste Disposal, 57 S.Cal.Law Review 745 (1984). James Graham Brawn Foundation, Inc. v. St. Paul Fire & Marine Ins. Co., 814 S.W.2d 273, 278 (Ky.1991). Moreover, “[c]ourts and commentators alike are in agreement that the term ‘occurrence’ is to be broadly and liberally construed in favor of extending coverage to the insured.” Id. (citing Buckeye Union Ins. Co. v. Liberty Solvent & Chem. Co., Inc., 17 Ohio App.3d 127, 477 N.E.2d 1227 (1984)). As a general rule, “the insured is entitled to all the coverage he may reasonably expect under the policy. Only an unequivocal, conspicuous and plain and clear manifestation of the company’s intent to exclude coverage will defeat this expectation.” Id. at 277 (citations omitted); accord, United States Fidelity & Guar. Co. v. Sandt, 854 P.2d at 522-23. C. Burden of Proof. As the insured, Quaker State bears the burden of proving that its claim comes within the broad meaning of “occurrence” and thus comes within the coverage under an insurance policy. See, e.g., Chemical Lea-man Tank Lines, Inc. v. Aetna Casualty & Sur. Co., 817 F.Supp. 1136, 1143-44 (D.N.J. 1993) (definition of “occurrence” in CGL policy constitutes grant of basic coverage under policy, and insured bears the burden of proving all elements of the definition); see generally, Blair v. Metropolitan Life Ins. Co., 974 F.2d 1219, 1221 (10th Cir.1992) (Oklahoma law); McGee v. Equicor-Equitable HCA Corp., 953 F.2d 1192, 1205 (10th Cir.1992); E-Z Loader Boat Trailers, Inc. v. Travelers Indem. Co., 106 Wash.2d 901, 906, 726 P.2d 439, 443 (1986). See also Morris v. Farmers Home Mutual Ins. Co., 28 Utah 2d 206, 500 P.2d 505 (1972). The burden does not shift to the insurer to prove that an occurrence was intended or expected, though the insurer may interpose evidence of insured’s knowledge to prevent the insured from meeting its burden. Cf. Adams-Arapahoe Joint Sch. Dist. No. 28-J v. Continental Ins. Co., 891 F.2d 772, 778-79 (10th Cir.1989) (all-risk policy; Colorado law); Queen City Farms, Inc. v. Central National Ins. Co. of Omaha, 64 Wash.App. 838, 827 P.2d 1024, 1040-41 (1992); but see Broderick Inv. Co. v. Hartford Accident & Indem. Co., 954 F.2d 601, 606 (10th Cir.1992) (“The district court properly assigned [insurer] the burden of proving [insured] knew its disposal practices would cause environmental damage.”) (Colorado law). D. “Neither expected nor intended ...” Citing a string of authorities extending back to Cardozo, Quaker State argues that “occurrence” depends upon whether the property damage was expected or intended, disputing the defendants’ contention that there was no “occurrence” at the Ekotek Site because the discharge of pollutants was intended or expected by the operators at the Ekotek Site. Quaker State also argues that the question of point of view is decisive: emphasizing the language defining “occurrence” in terms of “property damage neither expected nor intended from the standpoint of the insured,” Quaker State submits that “[environmental damage to third-party property that occurs unbeknownst to the insured, and which the insured did not intend or expect, ... constitutes a covered ‘occurrence.’ ” Id. at 12 (emphasis added in memorandum). In effect, Quaker State contends that defendants have failed to carry their burden under Rule 56 to “point to any evidence in the record that [Quaker State] expected or intended to cause property damage at the Ekotek site,” or for that matter, intended any specific acts which resulted in property damage. Id. Courts in other jurisdictions, considering the “neither expected nor intended” language in other factual contexts, are divided on the degree to which the insured must “intend” the harm, viz., whether specific intent to cause the particular injury is required, or whether natural and probable consequences, or foreseeability of the harm is sufficient. See Annotation, Construction and Application of Provision of Liability Insurance Policy Expressly Excluding Injuries Intended or Expected by Insured, 31 A.L.R.4th 957, 983-999 (1984). While courts disagree as to the reading of “intended,” they more consistently read “expected” to require “a high degree of certainty or probability.” Id. at 999-1002. The Utah Supreme Court has not yet addressed the meaning of “occurrence” in the context of a CGL policy. Recently, however, in State Farm Fire & Casualty Co. v. Geary, 869 P.2d 952 (Utah Ct.App.1994), the Utah Court of Appeals considered the meaning of “occurrence” in the context of a loss involving intentional conduct resulting in allegedly unintended harm. The issue arose under a homeowners liability policy providing coverage “for damages because of bodily injury or property damage ... caused by an occurrence.” The policy defined “occurrence” as “an accident, including exposure to conditions, which results in” bodily injury or property damage. Under this definition, Geary said, coverage extends to “only those injuries to property or person resulting from an ‘accident,’ and not an intentional occurrence.” Id. at 954. With “accident” as the starting point, the existence of a covered “occurrence” is determined first with reference to the conduct or event causing the injury (in Geary, the discharge of a shotgun in the victim’s direction) rather than solely the injury itself (victim struck in the head, neck and chest with 132 pellets). Where the act causing injury is deliberate, there is no “accident” and therefore, no “occurrence;” it does not matter that the resulting injury was not intended or expected. Id. Following the reasoning of Geary, intentional or deliberate conduct is not an “accident” and cannot constitute an “occurrence.” That reasoning is decisive here, however, only if Quaker State cannot come forward with any evidence from which a reasonable fact finder could find the existence of one or more “accidents” that may constitute a covered “occurrence.” Where there is significant probative evidence in the record from which a fact finder may conclude that Quaker State did not engage in intentional discharge of hazardous waste at the Ekotek Site, defendants are not entitled to summary judgment under Geary. Nor can the term “accident” be read in total isolation from the remainder of the policy definition of “occurrence.” After all, “occurrence” was incorporated into CGL policy language to replace — and broaden — coverage which previously had been defined solely in terms of “accident.” See Part III.B supra. “Occurrence,” as used in a CGL policy, should be construed more broadly than Geary's reading of a homeowner’s policy might suggest. Significant case authority supports the proposition that to be excluded from the definition of “occurrence,” the injury must be intended or expected, rather than merely the specific conduct which causes the injury. Several courts have held that “expected” and “intended” as used in the definition of “occurrence” are nearly synonymous, so that for all practical purposes the evidence must establish an insured’s subjective intent to cause harm in order to defeat coverage. See, e.g., City of Johnstown v. Bankers Standard Ins. Co., 877 F.2d 1146, 1150-51 (2d Cir.1989) (“Recovery will be barred only if the insured intended the damages, ... or ... the insured knew that the damages would flow directly and immediately from its intentional act, ...” [citations omitted]); Queen City Farms, Inc. v. Central National Ins. Co. of Omaha, 64 Wash.App. 838, 827 P.2d 1024, 1033-40 (1992); United Services Auto Ass’n v. Elitzky, 358 Pa.Super. 362, 517 A.2d 982 (1986); Patrons-Oxford Mutual Ins. Co. v. Dodge, 426 A.2d 888, 890-91 (Me.1981); State Farm Fire & Casualty Co. v. Muth, 190 Neb. 248, 207 N.W.2d 364, 366 (1973) (“The term ‘expected’ when used in association with ‘intended’ carries the connotation of a high degree of certainty or probability ... ”); 7A J. Appleman, Insurance Law and Practice § 4492.02, at 26-38 (Berdal ed. 1979). See also Annotation, Liability Insurance Coverage for Violations of Antipollution Laws, 87 A.L.R.4th 444, 502-15 (1991). The Court is satisfied that insofar as CGL policy coverage of “occurrences” is concerned, a broader reading is appropriate, and Quaker State “is entitled to coverage under its policies unless it has specific and subjective intent to cause the pollution giving rise to the CERCLA claims.” Brown Foundation, 814 S.W.2d at 278. See also 7A J. Appleman, Insurance Law and Practice § 4492.02 (Berdal ed. 1979) (expected and intended injuries flowing from intentional acts are excluded from coverage). “The ‘expected or intended’ exception is inapplicable unless the insured specifically and subjectively intends the injury giving rise to the claim.” Brown Foundation, at 278. The Court has examined the defendants’ citations to materials in the record (see Fireman’s Fund Mem. at 44-50) which purport to show that the discharge of hazardous waste and resulting property damage at the Ekotek Site were the product of “intentional and, to a large extent, criminal conduct.” Defs. Joint Mem. at 13. The referenced “intentional ... conduct” is almost exclusively that of the operators of the Ekotek Site, particularly principals in Ekotek, Inc., not Quaker State or its principals. There was no “accident” in Geary because the insured himself intentionally fired the shotgun, causing the injury and loss. Nothing in defendants’ memoranda points to any incident at the Ekotek Site in which, in effect, Quaker State deliberately squeezed the hazardous waste discharge “trigger,” causing the property damage at the Ekotek Site. Nor have defendants raised a genuine issue of material fact concerning whether the property damage at the Ekotek Site was “expected” or “intended” from the standpoint of Quaker State. While Quaker State may be held legally accountable for the actions of the Ekotek Site operators under CERCLA, it does not necessarily follow that the operators’ intent or expectations are imputed to Quaker State for purposes of liability insurance coverage: “One who is statutorily vicariously liable for the intentional act of another cannot be denied coverage since the defendant does not possess the requisite intent to do injury.” Brown Foundation, 814 S.W.2d at 277 (citing 1A R. Long, The Law of Liability Insurance § 5.06, at 5-54 (1990)); accord, Morrisville Water & Light Dept. v. United States Fidelity & Guar. Co., 775 F.Supp. 718 (D.Vt.1991) (while insured’s delivery of PCBs to site operator was an intentional act, operator’s mishandling of waste was not “expected”); Centennial Ins. Co. v. Lumbermens Mutual Casualty Co., 677 F.Supp. 342 (E.D.Pa.1987). CERCLA imposes strict liability jointly and severally upon “responsible parties,” imposing legal responsibility upon hazardous waste generators regardless of their subjective intent. Based upon the present record, it does not appear that a reasonable fact-finder could find that Quaker State “ha[d] specific and subjective intent to cause the pollution giving rise to the CERCLA claims” at the Ekotek Site, or that Quaker State expected with “a high degree of certainty or probability” (31 A.L.R.4th at 999), that “the damages would flow directly and immediately from” Quaker State’s own conduct in the disposition of its hazardous waste under contract with the operators of the Ekotek Site. City of Johnstown v. Bankers Standard Ins. Co., 877 F.2d at 1150-51. E. “Occurrences” Within Periods of Coverage. “Occurrence” is important to the question of coverage for another reason: defendants contend that no “occurrence,” properly defined, took place during the periods of coverage under their respective policies. See Defs. Joint Mem. at 51-64. The “occurrence” concept applies not only to identify what kind of event may represent a covered loss, but also to identify when the loss is deemed to have “occurred” for purposes of determining coverage. Quaker State bears the burden of showing that an “occurrence” triggering coverage under defendants’ policies took place during the policy period. See New Hampshire Ins. Co. v. Martech USA, Inc., 993 F.2d 1195, 1199-1200 (5th Cir.1993) (Texas/admiralty law). Courts around the country have formulated various tests for determining when, in the context of claims for indemnification for costs of environmental clean-up, an “occurrence” has taken place within the meaning of a liability insurance policy. See Annotation, Liability Insurance Coverage for Violations of Antipollution Laws, 87 A.L.R.4th 444, 515-27 (1991); F. Powell, Insuring Environmental Cleanup: Triggering Coverage for Environmental Property Damage Under the Terms of a Comprehensive General Liability Insurance Policy, 71 Neb.L.Rev. 1194 (1992). These tests include (1) the “exposure” trigger, (2) the “actual injury” or “injury-in-fact” trigger; (3) the “manifestation” or “discovery” trigger, and (4) the “continuous” trigger. See, e.g., Dow Chem. Co. v. Associated Indem. Corp., 724 F.Supp. 474, 478 (E.D.Mich.1989). The Utah courts have not yet adopted one of these tests to determine when an “occurrence” has taken place under a CGL insurance policy. 1. The “exposure” trigger. According to courts adopting the “exposure” trigger, liability under a CGL policy is triggered at the time that the property is exposed to the hazardous substance. F. Powell, supra; see, e.g., Continental Ins. Cos. v. Northeastern Pharmaceutical & Chem. Co., 811 F.2d 1180, 1190 (8th Cir. 1987), aff'd in part, rev’d in part on other grounds en banc, 842 F.2d 977, 984 (8th Cir.), cert. denied, 488 U.S. 821, 109 S.Ct. 66, 102 L.Ed.2d 43 (1988) (“NEPACCO ”). The “exposure” trigger is thought to reflect a commonly held view that environmental “property damage” occurs “at the moment that hazardous wastes are improperly released into the environment ... [Wjhere improper disposal of hazardous wastes immediately results in release into the environment, this trigger is justified. Hazardous wastes are by definition harmful, and exposure and injury in fact occur simultaneously in such cases.” F. Powell, supra. In NEPACCO, exposure occurred when 842 F.2d at 979. “Under the specific facts presented,” the Eighth Circuit continued, “the crucial events — the improper disposal of the hazardous wastes (wrongful act), the release of hazardous wastes into the environment (exposure), the contamination of the environment (injury-in-fact), — all happened virtually simultaneously.” Id. at 984. NEPACCO disposed of about eighty-five 55-gallon drums of hazardous wastes by burying them in a trench on a farm near Verona---- Many of the drums had deteriorated and were in poor condition at the time of disposal; many broke open when they were dumped into the trench. A strong chemical odor persisted in the immediate area of the Denney farm site for several months thereafter. In 1971 or 1972 NEPACCO hired Independent Petrochemical Corp. (IPC) to dispose of more hazardous wastes containing dioxin---- [IPC’s employee] allegedly transported and sprayed the hazardous wastes, mixed with waste oil, as a dust suppressant on the grounds of the Bubbling Springs Stables in Fenton, Missouri, and on the roads of Times Beach, Missouri____ 2. The “actual injury” trigger. The “actual injury” or “injury-in-fact” trigger invokes coverage at the point when actual injury or damage to the person or property occurs. “Under this theory, an actual injury must occur during the time the policy is in effect in order to be indemnifiable.” F. Powell, supra. The “actual injury” or “injury-in-fact” trigger is premised upon the literal language of the standard CGL policy’s definition of “occurrence:” The plain language of the definition of “occurrence” used in the CGL policy requires exposure that ‘results, during the policy period, in bodily injury’ in order for an insurer to be obligated to indemnify the insured. The unambiguous meaning of these words is that an in