Opinion

NEWSOM, J.

Advanced Micro Devices, Inc. (hereafter AMD), appeals from a summary judgment in favor of the respondents, Great American Surplus Lines Insurance Company and Allianz Underwriters, Inc. (hereafter respondents, Great American, or Allianz).

AMD brought suit against Great American and Allianz alleging that they had wrongfully rejected an insurance claim for costs incurred in removing a toxic contaminant from the premises of its Sunnyvale plant. The complaint concerned two insurance policies: a primary insurance policy of Great American and a related excess insurance policy of Allianz for the period August 27, 1981, to August 27, 1982, and a renewal policy of Great American for the period August 27, 1982, to August 27, 1983. Great American first secured a summary judgment on issues pertaining to the renewal policy and then moved for a summary judgment with respect to its first policy. Allianz joined in the motion on the ground that its excess insurance policy adopted the pertinent exclusion of Great American’s policy. The *794 superior court granted the joint motion and entered judgment on August 19, 1985, in favor of Great American and Allianz.

A federal regulation published on January 12, 1981, under the Resource Conservation and Discovery Act, required certain manufacturers to insure against pollution hazards. To comply with the regulation, AMD secured binders early in September 1981 for the initial insurance policies at issue in this case. Like other similar policies on the market, the insurance policies contained an exclusion for known preexisting conditions. Section IV.K.l provided: “This insurance does not apply to Loss: . . . K. arising out of: 1. the correction of any known pre-existing conditions at premises owned, occupied, rented, used or in the care, custody, or control of the Insured or over which the Insured is for any purpose exercising physical control; . . .” The Allianz policy provided in section 2 that “the coverage provided by this Policy shall follow the insuring agreements, conditions and exclusions of the underlying insurance . . . .”

Shortly after obtaining the insurance, AMD was required to submit an application form calling for various disclosures retroactive to August 27, 1981. Section X of the form asked a question obviously directed at the known preexisting conditions exclusion: “At the time of signature, are you aware of any circumstance which could give rise to a claim under this policy?” AMD answered “no.” The principal issue on appeal concerns whether AMD was required to disclose problems encountered in the Acid Neutralization System “C” (hereafter System “C”) at its semiconductor fabrication plant in Sunnyvale, California.

The following year AMD was compelled to clean up a high level of contamination by a toxic substance, chlorinated hydrocarbons, in the soil and ground water surrounding the settling tank of System “C.” To remove the toxic chemicals, it installed dewatering wells to pump out contaminated ground water and excavated the contaminated soil. When the work was completed in April 1983, more than 8 million gallons of water had been pumped from 5 dewatering wells and a soil mass measuring about 150 feet long, 40 feet wide, and 32 feet deep had been removed from an area heavily impacted with utility lines and other facilities. The total cost exceeded $1.5 million.

On October 15, 1982, AMD submitted a claim for the cost of cleaning up the System “C” contamination. In a letter dated March 29, 1983, Great American denied coverage on the ground that the loss fell within the known preexisting conditions exclusion. The letter states, “[fjurther, please be advised that in the course of our investigation we have determined that as *795 early as July 21, 1981, Advanced Micro Devices had knowledge of the contamination resulting at System “C.” . . . flf] Because of that prior knowledge, this company must respectfully decline coverage for the captioned loss.” A month later AMD filed suit.

As amended in 1982, Code of Civil Procedure section 437c, subdivision (c) imposes an affirmative duty on the trial court to grant a motion for summary judgment in an appropriate case: “The motion . . . shall be granted if all the papers submitted show that there is no triable issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” The amendment was in accord with the tendency of the courts to treat motions for summary judgment with greater favor. (6 Witkin, Cal. Procedure (3d ed. 1985) Proceedings Without Trial, § 309, p. 603.) Nevertheless, the courts—in particular, the Supreme Court—continue to enunciate traditional principles in the review of summary judgments.

“ ‘[I]ssue finding rather than issue determination’ ” is said to be “the pivot upon which the summary judgment law turns.” (Mann v. Cracchiolo (1985) 38 Cal.3d 18, 36 [210 Cal.Rptr. 762, 694 P.2d 1134].) In other words, the court inquires whether there are any triable issues of fact but refrains from weighing the evidence or making any determination of fact. (Leo F. Piazza Paving Co. v. Foundation Constructors, Inc. (1981) 128 Cal.App.3d 583, 589 [177 Cal.Rptr. 268]; Freidberg v. Freidberg (1970) 9 Cal.App.3d 754, 763 [88 Cal.Rptr. 451].) The Supreme Court continues to admonish that summary judgment is a “drastic” remedy “and should be used with caution.” (Mann v. Cracchiolo, supra, 38 Cal.3d 18, 35.) “Because summary judgment is a drastic procedure all doubts as to the propriety of granting a motion for summary judgment should be resolved in favor of the party opposing the motion.” (Palma v. U.S. Industrial Fasteners, Inc. (1984) 36 Cal.3d 171, 183 [203 Cal.Rptr. 626, 681 P.2d 893]; Miller v. Bechtel Corp. (1983) 33 Cal.3d 868, 874 [191 Cal.Rptr. 619, 663 P.2d 177].) “ ‘The moving party bears the burden of furnishing supporting documents that establish that the claims of the adverse party are entirely without merit on any legal theory.’ ” (Mann v. Cracchiolo, supra, 38 Cal.3d at p. 35.) “In examining the sufficiency of affidavits filed in connection with the motion, the affidavits of the moving party are strictly construed and those of his opponent liberally construed . . . .” (Stationers Corp. v. Dun & Bradstreet, Inc. (1965) 62 Cal.2d 412, 417 [42 Cal.Rptr. 449, 398 P.2d 785]; Mann v. Cracchiolo, supra, 38 Cal.3d at p. 35.) “All reasonable inferences are drawn in favor of the party opposing the summary judgment.” (Neinstein v. Los Angeles Dodgers, Inc. (1986) 185 Cal.App.3d 176, 179 [229 Cal.Rptr. 612]. But evidence may be so lacking in probative value that it fails to raise any triable issue. “To put the matter otherwise, the issue

*796 of fact becomes one ‘of law’ and loses its ‘triable’ character if the undisputed facts leave no room for a reasonable difference of opinion.” (Terry v. Atlantic Richfield Co. (1977) 72 Cal.App.3d 962, 967 [140 Cal.Rptr. 510].)

Discovery proceedings in this case produced a quantity of evidence on the issue of AMD’s knowledge of chlorinated hydrocarbon contamination in the form of internal AMD memoranda and depositions of Michael E. Gingrass, supervisor of environmental operations, and his two immediate supervisors, Chris Mistry, manager of deionized water, and Max L. Chancellor, director of facilities. In opposition to the motion for summary judgment, AMD also filed a series of declarations of Gingrass and other AMD officials. The evidence produced by discovery is essentially in agreement although Chancellor’s testimony differs somewhat from other evidence in emphasis and interpretation. The declarations, however, present a sharply different account of events. We will examine the facts revealed by discovery and then consider the significance of AMD’s declarations.

The manufacture of silicon wafers at AMD’s Sunnyvale plant involved the etching of wafers in acid baths. After the wafers were rinsed off, an acid solution was drained into the acid neutralization system where the pH was balanced by mixing it with anhydrous ammonia. Once the solution lost its acidic character it was dumped into Sunnyvale sewers. The Acid Neutralization System “C” was composed of three underground steel tanks, lined with fiberglass, each with a capacity of about 1,400 gallons. The first tank operated as a settling tank; the acid solution was retained for a brief period so that oils could be skimmed off the top and any heavier materials could settle to the bottom. The solution was then drawn into the other two tanks for acid neutralization.

As part of a preventive maintenance program, AMD closed System “C” on June 29, 1981, for inspection. A waste disposal firm, IT Corporation, was hired to pump out and dispose of the solution in the settling tank. But when IT Corporation drained the tank to a certain level, it found that ground water was pouring back into the tank through a large hole as fast as the acid solution could be pumped out. An AMD employee wearing protective garb entered the tank and found a large gash about one and one-half feet long and nearly one inch wide and several smaller holes. In some places the fiberglass was so thin that it could be punctured with a finger.

The next day, IT Corporation called AMD’s environmental operations supervisor, Mike Gingrass, explaining that it could not dispose of the waste at its Martinez treatment facility as planned. A laboratory test indicated that the solution taken from the settling tank had a 30 percent concentra *797 tion of chlorinated hydrocarbons which the facility was unable to process. Gingrass instructed IT to put the solution in drums and arrange for it to be buried at the Kettleman Hills toxic waste disposal site.

A chlorinated hydrocarbon, trichlorobenzene, was in wide use as a solvent at the plant, and certain other chlorinated hydrocarbons were also used in particular processes. In general, chlorinated hydrocarbons were collected separately, recycled, or disposed of through an independent contractor. The manufacturing process did, however, allow small quantities of chlorinated hydrocarbons adhering to silicon wafers to wash off during acid baths. Gingrass testified that this process, known as wafer drag-out, typically resulted in the presence of “something less than 20 parts per billion” of chlorinated hydrocarbons in the acid wastes. The director of facilities, Max Chancellor, described the quantities of chlorinated hydrocarbons discharged with acid wastes in the wafer drag-out process as being “diminishingly small, almost non-detectable.” The manager of deionized water, Chris Mistry, speculated that, since chlorinated hydrocarbons are heavier than water, they might collect in the bottom of the settling tank. But Gingrass thought that the solution of chlorinated hydrocarbons produced by wafer drag-out was too diluted to separate in the tank.

The discharge of chlorinated hydrocarbons was regulated by the Regional Water Quality Control Board, which limited allowable solutions to .05 milligrams per liter, and by the City of Sunnyvale, which limited the allowable discharge to .02 milligrams per liter and conducted quarterly tests at the AMD plant to assure compliance. Although the acid wash in the wafer drag-out process might conceivably produce solutions above the municipal standard, the city had never notified AMD of excessive levels of chlorinated hydrocarbons in its sewer discharge.

After receiving IT Corporation’s report of a high concentration of chlorinated hydrocarbons in the acid waste system, Gingrass searched unsuccessfully for sources of contamination other than wafer drag-out that could account for the contamination. Together with another engineer, he examined blueprints of the drainage system and physically inspected pipes used to drain chlorinated hydrocarbons to make sure that they were not connected to the acid neutralization pipes.

About the middle of July, Gingrass arranged to have an engineering consultant, Engineering Science, analyze two samples of water from System “C.” The first sample was of ground water infiltrating into the settling tank. To take it, he drained the tank, allowed water to pour back through holes in the tank, and then filled a bottle with the infiltrating ground water. The *798 second sample was taken from a valve sump, located 15 to 20 feet from the settling tank, that collected water percolating through the ground. An Engineering Science representative pointed to the settling tank as the most likely source of the water in the sump since it reached the same level as pipes leading from the tank.

On August 4 Engineering Science reported that the first sample contained 25.7 milligrams per liter of chlorinated hydrocarbons and the second sample 4.1 milligrams per liter. These concentrations were, respectively, 514 times and 82 times above the allowable discharge set by the Regional Water Quality Control Board, but they were far below—in fact, some 10,000 times below—the concentration reported by IT Corporation in the acid waste removed from the settling tank.

Gingrass reported the problems in System “C” to his supervisors in a series of five memoranda that form the heart of the respondents’ case. In a memorandum to Mistry dated July 13, 1981, he described the condition of the settling tank and reported that the laboratory of IT Corporation had found a 30 percent concentration of chlorinated hydrocarbons in liquid pumped from the tank. A later memorandum to Mistry dated July 21, 1981, concluded: “In addition to the problem of repairing the tank, I have no way of estimating the amount or type of liquid when [sz'c] may have escaped to contaminate the surrounding soil. I recommend a ground water and soil analysis from this area to determine whether a problem exists.” A few days later Gingrass noted the System “C” problem in a more comprehensive Environmental Compliance Status Report warning that the company was “far from being in compliance” with environmental regulations and faced potential exposure to heavy fines and criminal indictments. On August 4, 1981, he reported to Chancellor the results of the Engineering Science analysis of the water samples and recommended further investigation to determine “the extent of the problem.” In a further Environmental Compliance Status Report addressed to Chancellor, Gingrass noted, “Neutralization System ‘C’—The settling tank has several holes and requires immediate. action. There is no way to estimate quantities of liquid which have and are escaping from the system.”

News of the problem also circulated among AMD officials through informal channels. Gingrass talked personally to Mistry, Chancellor, and environmental counsel, Richard Lovgren. Mistry reported the problem to Chancellor and urged that a ground water monitoring system be installed. He recalls that the problem was discussed in a July meeting attended by “Richard Lovgren, [himself], Mike Gingrass, Max Chancellor, Tom Skornia, Tony Holbrook and some consultants and vendors and other people.”

*799 By all accounts, the AMD officials agreed on the need for more information; it was hard to evaluate the extent of the problem in view of the anomalous discrepancy between the laboratory tests of IT Corporation and Engineering Science and the lack of any satisfactory explanation for the source of chlorinated hydrocarbon contamination. Chancellor doubted the accuracy of the tests since the only apparent source of contamination was wafer drag-out which would probably produce much lower levels. He testified, “I felt that we should retest it.” In fact, he contacted a soil engineering firm and obtained a proposal dated August 26, 1981, for a series of test borings to study the problem. The proposal mentioned only acid waste spill, apparently reflecting Chancellor’s scepticism about chlorinated hydrocarbon contamination. But after discussing the problem further with Gingrass, Chancellor decided that the resources of his department could be best devoted to another environmental problem that appeared more urgent. He instructed Gingrass to obtain a second proposal for a soils study and deferred action on the System “C” until the next year.

In February 1982, AMD contracted for the drilling of five ground water monitoring wells on the plant premises and again retained Engineering Sciences to analyze soil and ground water samples. Samples taken from the fifth well revealed high concentrations of chlorinated hydrocarbons in the vicinity of the System “C” settling tank. At a level of five to six and one-half feet below the surface, the degree of contamination was roughly similar to that reported in Engineering Science’s earlier test of ground water infiltrating into the tank, but at a deeper level soil samples displayed the very high concentration of 320,000 parts per billion—a level of contamination, it should be added, still much lower than that of the settling tank fluids originally reported by IT Corporation.

In the face of this evidence, appellant asserts: “(3) the uncontradicted evidence establishes that until February 1982—six months after inception of coverage—no one at AMD knew of the existence of the chlorinated hydrocarbon contamination condition; (4) AMD’s investigation of the gash in the settling tank prior to inception of coverage indicated that only negligible amounts of chlorinated hydrocarbons could have entered the acid waste settling tank.”

Appellant seeks to derive some support from the testimony of Chancellor who was sceptical of the IT Corporation and Engineering Science test results. But Chancellor’s testimony cannot be pushed too far: he concluded only that the degree of contamination needed to be retested. Appellant’s briefs place most of their reliance on declarations of Gingrass and Chancellor, obviously prepared by counsel, filed in opposition to the motion for *800 summary judgment. In these declarations, they allege that during the year 1981 they were aware only of a “gash” in the acid neutralization settling tank. They believed that IT Corporation’s report of chlorinated hydrocarbon contamination was disproved by the later analysis of Engineering Science. Further investigation indicated that chlorinated hydrocarbons were present in only negligible quantities as a result of wafer drag-out. It was not until February 1982, they contend, that they acquired knowledge of chlorinated hydrocarbon contamination in the soils and water surrounding the settling tank.

The declarations are, to say the least, disingenuous. Chancellor declares, “I was later informed by Mr. Gingrass that instead of the 30% chlorinated hydrocarbons IT had earlier reported, the amount of chlorinated hydrocarbons from inside the tank was minimal.” In fact, he was informed that ground water flowing into the tank had a concentration of chlorinated hydrocarbons 514 times the allowable limit set by the Regional Water Quality Control Board. Moreover, the declaration of Gingrass strains the evident meaning of his own sworn testimony on deposition. For example, Gingrass avers, “At no time in 1981 did I know of, believe, or inform anyone at AMD that a condition of chlorinated hydrocarbon contamination existed in the soils and ground water beyond the engineered backfill immediately surrounding the tanks.”

In reviewing motions for summary judgment, the courts have long tended to treat affidavits repudiating previous testimony as irrelevant, inadmissible, or evasive. (Bauman, California Summary Judgment: A Search for a Standard (1963) 10 UCLA L.Rev. 347.) In D’Amico v. Board of Medical Examiners (1974) 11 Cal.3d 1, 21 [112 Cal.Rptr. 786, 520 P.2d 10], the Supreme Court first approved the rule that the declaration of facts by affidavit contrary to deposition testimony does not constitute “substantial evidence of the existence of a triable issue of fact.” It explained, “As the law recognizes in other contexts (see Evid. Code, §§ 1220-1230) admissions against interest have a very high credibility value. . . . Accordingly, when such an admission becomes relevant to the determination, on motion for summary judgment, of whether or not there exist triable issues of fact (as opposed to legal issues) between the parties, it is entitled to and should receive a kind of deference not normally accorded evidentiary allegations in affidavits.” (Id. at p. 22.) The rule has been followed in a line of subsequent decisions. (Leasman v. Beech Aircraft Corp. (1975) 48 Cal.App.3d 376, 382 [121 Cal.Rptr. 768]; Gray v. Reeves (1977) 76 Cal.App.3d 567, 573-574 [142 Cal.Rptr. 716]; Mikialian v. City of Los Angeles (1978) 79 Cal.App.3d 150, 161 [144 Cal.Rptr. 794]; Girard v. Ball (1981) 125 Cal.App.3d 772, 781 *801 [178 Cal.Rptr. 406]; Shapero v. Fliegel (1987) 191 Cal.App.3d 842, 849 [236 Cal.Rptr. 696].)

AMD places much reliance on the rules requiring strict construction of insurance contracts against the insurer. (For precise statement of the rules, see Reserve Insurance Co. v. Pisciotta (1982) 30 Cal.3d 800, 807-808 [180 Cal.Rptr. 628, 640 P.2d 764]; Continental Cas. Co. v. Phoenix Constr. Co. (1956) 46 Cal.2d 423, 437 [296 P.2d 801, 57 A.L.R.2d 914]; 1 Witkin, Summary of Cal. Law (8th ed. 1973) §§ 535-540, pp. 456-464.) Interpreting narrowly the exclusionary clause relating to known preexisting conditions, AMD argues that it required proof of actual knowledge of these conditions, not merely proof that AMD “might have known” or “should have known” of the conditions.

Respondents counter by citing Garcia v. Truck Ins. Exchange (1984) 36 Cal.3d 426, 438 [204 Cal.Rptr. 435, 682 P.2d 1100], and Fireman’s Fund Ins. Co. v. Fibreboard Corp. (1986) 182 Cal.App.3d 462, 466-468 [227 Cal.Rptr. 203], which hold that the rules of strict construction against the insurer do not apply where the insurance contract is the product of joint drafting between parties of comparable bargaining strength. As a major manufacturer of semiconductors, AMD was a sophisticated insurance purchaser that enjoyed at least some power to negotiate changes in policy provisions. While it did not possess the same degree of bargaining strength as the insureds in Garcia and Fireman’s Fund, it was represented by a national insurance broker and an experienced risk manager; before procuring environmental impairment liability insurance, it obtained a sophisticated analysis of four proposals; and on objecting to another exclusionary clause, it was able to obtain a change in the language. The respondents argue that the holdings of Garcia and Fireman’s Fund should be extended to the facts of the present case.

The issue is difficult and significant, but it does not need to be resolved on this appeal. “Words used in an insurance policy are to be interpreted according to the plain meaning which a layman would ordinarily attach to them. Courts will not adopt a strained or absurd interpretation in order to create an ambiguity where none exists.” (Reserve Insurance Co. v. Pisciotta, supra, 30 Cal. 3d 800, 807.) Under any interpretation consistent with common usage, the language of the preexisting knowledge exclusion applies to the facts of the case. The undisputed evidence reveals that AMD had actual knowledge of a problem of chlorinated hydrocarbon contamination in System “C” before inception of the policy; it was uncertain only of the extent of the problem. If we dismiss the self-serving declarations prepared by counsel, the evidence most favorable to AMD indicates that the

*802 soil and ground water surrounding System “C” needed to be retested for an accurate evaluation of the contamination. As Gingrass asserts, it may not have appeared “near as big a problem as it turned out to be.” But AMD knew that fluids from the acid neutralization system freely entered the soil and ground water through holes in the settling tank, and the test results of IT Corporation and Engineering Science put it on notice that these fluids contained chlorinated hydrocarbons in concentrations far above the limits permitted by environmental regulations. At the time its insurance coverage began, AMD could not reasonably doubt that it faced a costly cleanup project.

We conclude that the trial court correctly ruled that exclusion K.l applied to the first policy. That holding disposes also of the issue of liability under the renewal policy. Since the language of the exclusion was identical under each, the summary judgment was properly granted.

The judgment of the trial court is affirmed.

Racanelli, P. J., and Elkington, J., concurred.