Atchison, Topeka & Santa Fe Railway Co. v. Stonewall Insurance

71 P.3d 1097 | Kan. | 2003

275 Kan. 698 (2003) 71 P.3d 1097

ATCHISON, TOPEKA & SANTA FE RAILWAY CO., Appellee,
v.
STONEWALL INSURANCE COMPANY, et al., Appellants.

No. 85,794

Supreme Court of Kansas

Opinion filed May 30, 2003.

Douglas R. Richmond, of Armstrong & Teasdale, LLP, of Kansas City, Missouri, argued the cause, and William A. Larson and Timothy A. Shultz, of Gehrt & Roberts, Chartered, of Topeka, and Richard W. Bryan and Erin N. McGonagle, of Jackson & Campbell, P.C., of Washington, D.C., were on the briefs for appellants American Home Assurance Company, Audubon Indemnity Company, Granite State Insurance Company, Insurance Company of Pennsylvania, Landmark Insurance Company, Lexington Insurance Company, National Union Fire Insurance Company of Pittsburgh, Pa., New Hampshire Insurance Company, and Union Atlantique D'Assurance, S.A.

Hal D. Meltzer, of Baker Sterchi Cowden & Rice, of Kansas City, Missouri, and Michael F. Sommerville, of Centrulo & Capone, LLP, of Boston, Massachusetts, were on the brief for appellant Stonewall Insurance Company.

Eric I. Unrein, of Davis Unrein McCallister Biggs and Head, of Topeka, and Eric C. Young, of Dunham Boman and Leskera, of Belleville, Illinois, were on the briefs for appellants Employers Insurance of Wausau and Nationwide Mutual Insurance Company.

Stephen W. Cavanaugh and Todd D. Powell, of Fisher, Cavanaugh, Smith & Lemon, P.A., of Topeka, and Joseph S. Crociata, Stuart Peacock, and Michelle Oshman, of Bonner, Kiernan, Trebach & Crociata, of Washington, D.C., were on the briefs for appellants Fireman's Fund Insurance Company and National Surety Corporation.

David V. Goodsir, of Freeborn & Peters, of Chicago, Illinois, argued the cause, and Weston W. Marsh, of the same firm, and Steve R. Fabert, of Fisher, Patterson, Sayler & Smith, L.L.P., of Topeka, were with him on the briefs for appellee Atchison, Topeka, & Santa Fe Railway Company.

Karen R. Glickstein, of Shughart Thomson & Kilroy, of Kansas City, Missouri, and Laura A. Foggan and Elizabeth A. Eastwood, of Wiley Rein & Fielding LLP, of Washington, D.C., were on the brief for amicus curiae Complex Insurance Claims Litigation Association.

The opinion of the court was delivered by

ALLEGRUCCI, J.:

This is a declaratory judgment action, which was filed by the Atchison, Topeka and Santa Fe Railway Company (Santa Fe) initially against approximately 120 insurance companies that issued excess liability policies to Santa Fe between 1945 and 1986. Appellants are some of the defendant insurance companies: American Home Assurance Company, Audubon Indemnity Company, Granite State Insurance Company, Insurance Company of the State of Pennsylvania, Landmark Insurance Company, Lexington Insurance Company, National Union Fire Insurance Company of Pittsburgh, Pa., New Hampshire Insurance Company, Union Atlantique D'Assurance, S.A., Nationwide Mutual Insurance Company, Employers Insurance of Wausau, a Mutual Company, Stonewall Insurance Company, Fireman's Fund Insurance Company, and National Surety Corporation (collectively known as Insurers).

Santa Fe sought declarations that the Insurers must indemnify it for settlements of several thousand claims and lawsuits by employees who alleged hearing losses due to excessive noise in the workplace. In 1995, the district court, ruling on cross-motions it had directed the parties to file before completion of discovery, entered judgment in favor of Santa Fe. Insurers appealed the 1995 decision, and this court reversed and remanded. Atchison, Topeka & Santa Fe Railway Co. v. Stonewall Insurance Co., 1997 WL 1048134.

On remand, the parties developed the factual record. Subsequently, ruling on the parties' motions and cross-motions for partial summary judgment, the district court concluded that Kansas law applied to all issues and that the declaratory judgment sought by Santa Fe would be entered in its favor. Insurers appealed. Insurers' motion for transfer to this court was granted pursuant to K.S.A. 20-3017.

The district court's decisions that, taken together, constitute the final declaratory judgment in this case are on choice of law, which were made on November 3, 1999, March 12, 1999, and December 14, 1998, and on all substantive issues, which were made on July 24, 2000. Each memorandum decision and order was the district court's ruling on a motion or motions for partial summary judgment or summary judgment. In each memorandum decision and order, the district court set out a statement of uncontroverted facts.

Although Insurers assert in several instances (trigger of coverage, nonfortuity, and known loss) that disputed facts preclude summary judgment, in fact, their substantive arguments challenge the district court's application of law to facts rather than the district court's findings of fact. Insurers' contentions involving facts are discussed along with the issues. The trial court's findings of fact relevant to the discussion of the issues are as follows:

1. As of March 1998, Santa Fe has paid in excess of $28 million in connection with more than 3,800 claims that have been filed by its employees. The employees allege that they sustained noise-induced hearing loss (NIHL) from continuous exposure to excessive noise while employed by Santa Fe.

2. Noise is an obvious and unavoidable by-product of Santa Fe's normal railroad operations.

3. NIHL is an injury that occurs contemporaneously with unprotected exposure to excessive noise.

4. NIHL injuries continue progressively throughout the course of unprotected exposure until the exposure to excessive noise is interrupted.

5. The NIHL claimants alleged that their work environment involved a network of multiple excessive noise sources and that they were exposed to on a daily basis in different ways to noises from a variety of sources. The claimants, whether engineers, carmen, or maintenance-of-way workers, were mobile employees, changing locations and job assignments in the course of their work.

6. Santa Fe purchased millions of dollars of comprehensive general liability insurance policies throughout the years, and Santa Fe also maintained a variable level of Self-Insured Retention (SIR).

7. From at least 1945 to 1986, Santa Fe maintained a program of excess insurance coverage purchased from Insurers to protect it from "liability incurred as a result of its operations."

8. Santa Fe's self-insured retentions underlying its excess policies ranged from $1,000,000 for the period from 1956 to 1971, to $7,000,000 for the period from 1984 to 1986. In 1991, the coverage grew to $10,000,000. The self-insured coverage currently is $25,000,000.

9. In each year of coverage, Santa Fe purchased several layers of insurance.

10. Each layer has upper and lower coverage limits and each layer is composed of separate, multiple policies issued by different insurance companies.

11. Total premiums on the International Surplus Lines Insurance Company (ISLIC) policies decrease as each successive layer of coverage is reached. For the period of March 17, 1982, to March 17, 1983, the premium ranged from $170,000 for a $1,700,000 coverage of the first $10,000,000 that exceeded Santa Fe's $5,250,000 Self-Insured Retentions, to a $5,000 premium for a $5,000,000 coverage of the $50,000,000 excess over the $150,250,000 that was covered by higher layers.

12. Under the excess policies, Insurers agree, subject to all terms and conditions of their policies, to indemnify the assured for certain losses up to the limits of the policies incurred as a result of an "occurrence."

13. On December 29, 1994, Santa Fe filed its original petition for declaratory relief against the defendant insurance companies that issued general liability insurance policies for the years 1945-1986 to Santa Fe, seeking coverage for claims brought by employees of the railroad for NIHL.

14. Santa Fe is not seeking reimbursement from any policy after 1986, after which time Santa Fe began to reinsure itself and after which time the NIHL occurrence ceased.

15. Each of the pre-1974 insurance policies at issue in this case contains standard language in which the insurers agreed, subject to terms and conditions of the policies, to indemnify Santa Fe for "any and all sums" of damages arising out of an "accident or accidents" in excess of a certain amount. While the coverage limits varied from year to year, the indemnification and limits policy language was otherwise substantially similar and provided:

"To indemnify the Assured for any and all sums which the Assured shall become liable to pay, and shall pay, to any person or persons as compensation for injury or damage to persons (whether such injury or damage be fatal or nonfatal) and injury or damage to property (excluding property of the Assured or in its custody or control) arising out of any accident or accidents caused by or growing out of the Assured's Railroad operations in the United States of America and all operations incidental thereto."

16. The pre-1974 Santa Fe policies define the term "Ultimate Net Loss" with wording substantially similar to the following definition:

"the sum actually paid in cash in the settlement of losses for which the Assured is liable, after making proper deductions for all recoveries, salvages, and other insurances, and shall include all loss and legal expenses (excluding salaries of employees and office expenses) incurred in investigation, adjustment or litigation."

17. Each pre-1974 Santa Fe policy contains an "other insurance" provision. These clauses are substantially similar and provide:

"This insurance does not cover any loss or damage which at the time of the happening of such loss or damage is insured by any other existing policy or policies of valid and collectible insurance except in respect of any excess beyond the amount which would have been payable under such other existing policy or policies had this insurance not been effected."

18. Some of the pre-1974 policies define "occurrence" to mean "one or more accidents or series of accidents arising out of or resulting from one event."

19. The post-1974 Santa Fe policies at issue in this case contain the following indemnity provision, or one that is substantially similar:

"Liability: this Policy also indemnifies the Assured for any and all sums which the Assured shall become legally liable or obligated by contract (subject to Section IV-C2 hereof) to pay to any person or persons as compensation or damages for injury or damage to any person or persons (whether such injury or damages be fatal or non-fatal) and for injury or damage to property (excluding property covered under Section I-A hereof) arising out of any occurrence or occurrences caused by or growing out of the Assured's operations and/or any operations incidental thereto during the term hereof; and to indemnify the Assured for legal, investigation, and other expenses as set forth in the definition of the Ultimate Net Loss in Section V-A hereof."

20. The post-1974 Santa Fe policies also provided for a retention. While the amount of the retention varied from year to year, the policy language was otherwise substantially similar and provided:

"Retention: Underwriters shall not be liable hereunder unless the Ultimate Net Loss amounts to $3,000,000 any one occurrence involving coverages as described in Sections I-A and I-B hereof and then only for the sum in excess of $3,000,000 Ultimate Net Loss subject to the limit of $5,000,000 Ultimate Net Loss; but, there is no limit to the number of occurrences for which claims can be made, provided such occurrences take place during the term hereof."

21. The post-1974 Santa Fe policies define "Ultimate Net Loss" as

"sums actually borne by the Assured after making deductions for all recoveries, all salvages and all collectible claims upon other insurances except as respects insurances procured in accordance with Section IV-A-7, and shall include all direct wrecking and salvaging expense and Assured's additional expenses arising from the settlement of claims including investigation and adjustment expenses as well as including all legal expenses incurred with the written consent of the Underwriters, other than the Assured's usual and ordinary fixed operating and office expenses."

22. The post-1974 policies typically defined "occurrence" as

"[O]ne or more accidents or disasters and/or series of accidents or disasters arising out of or resulting from one event.

. . . .

"Each `occurrence' shall be deemed to commence on the first happening of any material damage not within the period of any previous `occurrence.'"

23. Some pre-1974 Santa Fe policies and post-1974 Santa Fe Policies provide as follows:

"Underwriters agree that in the event of a claim for damages for bodily injury or property damage being made by any Assured hereunder against any other Assured hereunder, this Policy shall cover such other Assured against whom claim is made in the same manner as though separate Policies shall have been issued to each Assured, but in no event shall there be imposed more than one deductible as set forth in Condition 4 in respect to one occurrence." (Emphasis added.)

24. Certain policies contain "other insurance" clauses which typically state:

"Notwithstanding anything to the contrary contained herein, it is further understood and agreed that where there is any other valid and collectible insurance providing coverages, as described in Section I-A and I-B hereof, this insurance shall be considered as excess insurance over and above such other insurance in effect at the time of the loss or damage except that permission is hereby granted for excess insurance over the limits of liability expressed in this Policy without prejudice to this insurance, and the existence of such insurance, if any, shall not reduce any liability under this Policy."

25. Each Santa Fe insurance policy at issue contains a notice clause. A typical "notice" provision is contained in International Surplus Lines Insurance Company (ISLIC) Policy Number 2298 (3/17/81 to 3/17/82). The pertinent language of that "notice" clause states:

"Written notice of each occurrence or accident shall be given by or on behalf of the Assured to the Underwriters as soon as practicable through Rollins Burdick Hunter Co., who are hereby authorized to instruct Messrs. Mendes & Mount, to assess the loss on behalf of Underwriters after notice has been received by an Executive Officer or the Manager of Insurance located at the Insured's Corporate Headquarters, 224 South Michigan Avenue, Chicago, Illinois 60604."

The other policies at issue have substantially similar language.

26. Each policy also contains the following clause or a substantially similar one:

"SECTION IX. PRECEDENCE OF THIS FORM

"The exact terms and conditions of this manuscript form are to be regarded as substituted for those of the certificate, cover note or policy to which it is attached and forms a part, wherever and insofar as such terms and conditions may conflict."

27. Some of the alleged policies produced by the insurers include jackets or cover notes which contain the following language or substantially similar language:

"Notice of Loss, Participation in Defense by the Company

"Notice of an occurrence which appears likely to involve this policy shall be given by or on behalf of the insured to the company or any of its authorized agents as soon as practicable. The company at its own option may, but is not required to, participate in the investigation, settlement or defense of any claim or suit against the insured."

Santa Fe disputes that the "jacket" language constitutes part of the policy.

28. Lexington Insurance Company, Fireman's Fund and National Surety Corporation, Nationwide Mutual, and Stonewall each first received notice of Santa Fe's NIHL claims in July 1989.

29. Insurers have no facts which demonstrate prejudice by the timing of the notice they received of the NIHL claims.

30. Insurers took no action to negotiate, defend, or prevent NIHL claims against Santa Fe.

31. Santa Fe's pay-outs for NIHL claims are among the lowest in the industry.

32. Of the 10 claims that Santa Fe settled from August 1965 through June 1971, nine pertained to traumatic hearing loss and only one pertained to NIHL.

33. In December 1976, E. B. Hill, a former Santa Fe engineer, filed a lawsuit against Santa Fe alleging NIHL.

34. By the time Santa Fe gave Insurers notice, Santa Fe had received notice of 247 NIHL claims (in addition to the claims noted above).

35. Steve Hanks, attorney with the law firm of Helm, Pletcher, Hogan, Bowen & Saunders, handled over 100 NIHL lawsuits and 25 claims against Santa Fe. Hanks testified in his deposition in this case:

"Q: Okay. In the noise-induced hearing loss cases, I take it from your affidavit that your belief or feeling or legal position is that the negligence was Santa Fe's failure to have a timely hearing conservation program; is that correct?

"A: It was and is our position.

"Q: Okay. Causation. What's the causation of the claims?

"A: Well, as a result of Santa Fe's failure to have a timely and appropriate hearing conservation program, the employees suffered noise-induced hearing loss while working for the railroad. I don't know exactly what you are asking.

. . . .

"Q: I'm asking you, you told me the elements of a FELA cause of action are negligence, causation, and damages. Negligence we've talked about. What do you mean by causation?

"A: Well, when you have an industry, any industry that generates noise in the process of work, then it is incumbent on the industry to provide a hearing conservation program for their employees who are working, to ensure that they do not suffer damage to their hearing.

"The railroad failed to do that. Had the railroad instituted a timely and appropriate hearing conservation program, our clients would not have suffered noise-induced hearing loss.

. . . .

"A: See, what I want to make sure you understand, because you keep talking about causation, in a FELA case, you have to prove that the defendant was negligent, you have to prove that the defendant's negligence caused the harm and then you have to prove the extent of the harm.

"So in the hearing loss cases, it didn't do us any good to prove that noise — put it this way: Just proving that their hearing had been damaged by noise did not even get us to the jury, because that did not establish causation.

"Causation that you are required to establish under the FELA is that the defendant's negligence caused, in whole or in part, damage to the plaintiffs and it was the negligence — it was the fact that their negligence caused the harm that we had to prove. The simple fact that there was noise out there doesn't establish negligence."

36. John Fabry, an attorney at the law firm of Jones and Granger, handled NIHL cases against Santa Fe. Jones and Granger filed over 1,000 claims against Santa Fe, approximately 230 of those being lawsuits. Fabry testified in his deposition regarding the Jones and Granger cases:

"Q: Can you elaborate for me as well as you can what the theory of liability has been in noise-induced hearing loss cases?

"A: There was sufficient information available for the railroad to recognize that a hazard existed to their employees and sufficient information to protect them against that hazard by way of implementing a hearing conservation program, and the railroad simply failed to do that. That failure to implement the program caused hearing loss in individual claimants, and, therefore, they were entitled to recover damages from the railroad under the FELA."

37. Plaintiff asserts that an adequate hearing conservation program requires six system-wide components: (1) noise exposure monitoring; (2) employee notification of monitoring results; (3) annual audiometric testing for affected employees; (4) availability of adequate hearing protection devices; (5) a NIHL training program; and (6) feasible administrative and engineering controls to reduce exposure to potentially hazardous noise levels as measured by appropriate testing and monitoring. Defendants controvert this fact only to state that another element is necessary to have an adequate hearing conservation program: proper management at all (including local) levels.

In addition to its Findings of Fact, the district court mentioned the following facts:

A. Counsel for an insurer conceded at oral argument on summary judgment motions that self-insurance is not insurance. (NOTE: That insurer is not among the appellants. The concession may merely have been with regard to a definition of insurance, as discussed in Issue 4.)

B. Santa Fe purchased no commercial insurance policies for the years 1952 through 1956.

C. Counsel for Santa Fe stipulated at oral argument before the district court

"that the occurrence (and thus the liability) started no earlier than 1966 (the date of Dr. Glorig's report on NIHL) and ended no later than 1984 or 1985. . . . If the occurrence (i.e. the failure to timely implement a Hearing Conservation Program) spanned 1966 through 1984 or 1985, none of the years in which Santa Fe was totally self-insured would be triggered, i.e. 1952-56, years before 1940, or years after 1986. Likewise the legal liability that Santa Fe incurred for failure to timely adopt a Hearing Conservation Program were the years 1966 through 1984 or 1985. There should have been no negligence prior to 1966 nor after 1984 or 1985 (when an HCP was implemented)."

The factual findings relevant to the district court's decisions on nonfortuity and known loss are as follows:

"1. December 22, 1970, Dr. Forrest H. Kendall of the Old Wesport Medical Association wrote a letter to Mr. F.W. Walters, Administration, Santa Fe RR advising that a Santa Fe employee to whom he had recently administered an audiogram had a rather severe type of hearing loss in both ears, the type of hearing loss associated with acoustic trauma. Dr. Kendall further advised:

`I am impressed by the number of railroad employees I see, both from your railroad and other railroads, for which my office does business, with the hearing loss associated apparently with noise exposure at work. It is our recommendation that anyone who is associated with working in an area where they cannot hear a co-worker speak at normal voice within [illegible] feet, is indeed in an area of too great a noise exposure. These people should all wear a type of protective device to reduce the possibility of acoustic trauma and irreversible hearing loss.'

"However, as Santa Fe points out, `acoustic trauma' is not necessarily noise-induced hearing loss (NIHL)."

2. During the period of August 1965 through June 1971, Santa Fe settled at least 10 employee claims involving hearing loss. However, 9 of the 10 claims pertain to traumatic hearing loss — not NIHL.

"3. July of 1971, MTS Associates, Inc. advised Santa Fe's Director of Safety, D.D. Baird:

`Hearing conservation programs are now a necessity in almost all industry. Even more for the railroads, since their compensation problems do not have the monetary limits imposed in Workmen's Compensation in the various states.

`Both you and Dr. Hanson certainly have a knowledge of the problems. There remains no doubt that noise induced hearing loss is the number one, non-fatal, hazard in industry today. Experience in the railroad industry shows that large numbers of your personnel, particularly in the shops and operating groups, are in the various exposure patterns. The risk proabilities [sic] of hearing impairment are significant.

`In order to reduce future liability and the human problems related to this hearing impairment, we must prevent the hearing loss from occurring. A complete controlled, continuing, valid hearing conservation program provided in the proposal presented to Mr. McMillan is designed to meet these requirements.'

"Santa Fe admits that quotation is accurate but argues that the letter was a marketing letter to Santa Fe and should not be used to infer that Santa Fe had knowledge that MTS Associates, Inc.'s statements were true and accurate.

"4. July 14, 1971, one of Santa Fe's attorneys wrote to Dr. Hanson, Santa Fe's Medical Director, the following:

`Since our recent telephone conversation, I contacted General Claim Agent Harrington to [illegible] claims we have had involving hearing loss by employees. He checked with Messrs. Eschenburg and Ketring, and [illegible] forwarded to me a summary of claims involving hearing loss for the three Grand Divisions over the past five years. Attached are copies of their reports.

`As you can see, relatively few of the claims are based on an overall high noise level [illegible]. I suspect you are correct, however, that [illegible] the type of claim we can expect to see more of in the future, especially if and when federal standards are set.'

"5. On September 7, 1972, the Kansas Department of Labor conducted a sound level survey at one Santa Fe location and found that the noise levels exceeded the permissible standards for the following area in one tie gang: Rail Lifter and Tie Pusher, Tie Handler, Tie Inserter Operator and Helper, Dual Hydraspiker, Air Compressor and Jack Hammer Operators and Helpers. Since the tie gang was soon to be disbanded no further action was taken on the complaint.

"6. In response to the September 7, 1972 citation, Santa Fe conducted its own noise survey and confirmed that the noise levels did in fact exceed the permissible 90dba [decibels] for eight hours.

"7. There were several noise-related complaints at various Santa Fe sites that resulted in some type of citations by an outside agency that were resolved by Santa Fe or by agreement between the agency and Santa Fe. Some of the noise-related complaints were due to violations of city noise ordinances and were not based on impermissible employee noise exposure levels.

"8. November 1, 1974, Mr. Shaver (In 1973, Mr. Shaver was assigned by Mr. Cena, Santa Fe's Vice President of Operations, to oversee and coordinate all noise related matters) sent Mr. Cena a 4 page letter in which he discussed the proposed OSHA regulations for noise levels and further advised:

`OSHA printed its proposed new noise standard in the October 24, 1974 Federal Register (Volume 39, No. 207). While most of the requirements mentioned in my February letter now appear in the OSHA proposal, there are some very important changes that may greatly affect the railroads. All of this, of course, depends on the long-pending decision of who will ultimately wind up with jurisdiction over the railroads, the DOT or the DOL. Should FRA end up with jurisdiction, they might possibly interpret the regulations differently, but I am convinced that the ground rules will remain essentially the same.

`While 90dba is still the time weighted, eight-hour exposure, OSHA is now concerned about the 85-90 dba range. This range is important to us because noise levels in most locomotive cabs fall between those values. Previously, we felt safe if we kept locomotive cab noise levels below 90 dba. However, should the new regulation be adopted in its present form, we could be subjected to a whole host of requirements that would be almost impossible to comply with. . . .

`We have three major areas that would be affected by these regulations: trainmen and enginemen, Maintenance of Way and shops. Identifying trainmen and enginemen who are subjected to an eight-hour time-weighted exposure of 85dba would be extremely difficult. Most freight and passenger runs today are not a duration that would expose the crew to over 85dba for an eight-hour period. However, there are many runs that do and the problem of determining which employees fall into which category would be next to impossible. The only way we could be certain that we comply would be to identify and test ALL enginemen and trainmen. In order to avoid getting into this in the first place, I suggest that we see what can be done to reduce the noise level in locomotive cabs below 85dba. If this can be done, we would eliminate a large part of our employee population who would otherwise be subject to these noise regulations.

`The next area, Maintenance of Way, will be our greatest concern. Most of our roadway machines do not meet the 90dba limit much less 85. Here, too, we must identify those employees who are subjected to an eight-hour time-weighted exposure of 85dba and arrange to give them annual audiometric tests as outlined in the quoted passages preceding. I can foresee the day when we will have to equip an old passenger car with audiometric equipment and send it around to various locations to perform these tests. The testing of all those connected with roadway machines would be an administrative and physical nightmare and I suggest we continue to devote much effort toward quieting our machines. The CWE Shop has been involved in conducting noise studies of roadway machines for several months now and I suggest they do all that is necessary to reduce the number of machines that fail to comply with noise standards.

`The third area is that of our shops. These will probably be the easiest to deal with because of their stationary nature. Nevertheless, it will take considerable effort to identify those needing testing and arrange the necessary tests to see that this is accomplished annually. Here, too, we should see what can be done to quieten [sic] those machines and processes that do not meet the regulations.

`In summary, I suggest that we immediately get very involved in a hearing conservation program. The less people we have exposed to 85dba noise, the less people we will have to test. Mr. J.M. McMillan of my office is available to help in any way possible.'

"9. In connection with the hearing loss claim of a Santa Fe employee in 1975, Santa Fe's General Attorney, Robert Bateson, advised Mr. Shaver:

`This has reference to your February 25 letter to Dr. Hanson and myself regarding a complaint as described above, which arises as a result of a letter from an Amarillo doctor to Superintendent K.C. Kay. From a reporting standpoint there is little doubt in my mind but that the matter is reportable. The patient contends the hearing loss arose following his assignment to a "spike hammer" and the doctor in testing the complainant finds a 45% sensorineural hearing loss in the left ear. Notwithstanding the possibility if such loss exists it is not a result of job connected activity, the fact is from a reporting standpoint the situation is presented in such a way it should be reported. There is simply no provision in the requirements which apply to us which permits the facts to be questioned and because of the magnitude of the problem generally I would not think it practical to do so. If the complaint matures into a FELA cause of action we will of course investigate thoroughly the possibility of a defense based on some physical element other than work connected causes.

`The situation also suggests another problem which I think should be brought to your attention. In this case there is no specific incident of the type which normally would come to the attention of the Claim Department. However, with the advent of health and occupational illness regulations we are confronted with new types of cases which have possibilities for liability under the FELA. It is my suggestion that cases of this type in which a work connected disability is claimed and doctor's advice to the effect is received, the Claim Department should be advised so that it can maintain a file and have a starting point in the event that thereafter some contention about FELA liability is made. It may be we should discuss how best to determine which situations should be referred to the Claim Department, but certainly cases such as the one which is the subject of this letter must be.'

"However, Santa Fe points out that it is unclear from the cited document whether the claim discussed was for traumatic hearing loss or for NIHL. In addition, Santa Fe points out that the reporting referred to was not insurance reporting but reporting to the Santa Fe claims department.

"10. May 5, 1977, R.L. Riess, an audiologist, wrote to Dr. Don Lynch of Santa Fe Employees Hospital Association:

`It has come to my attention that a number of employees of the Santa Fe Railroad are possibly not having adequate ear protection available in their work. I am somewhat concerned about the great number of people that we are testing that have a significant hearing loss. I believe that it would be of Santa Fe's best financial interest to initiate a hearing conservation program, including use of ear protection.'

"11. May 6, 1977, Dr. Lynch wrote to Dr. Hanson, Santa Fe's Medical Director-System:

`Over the past few months I have become increasingly concerned about the hearing problem which seems to occur among Santa Fe Railroad employees. We have sent a great many of these folks to the audiology department for check and today I received this memo from Dr. R.L. Riess, the audiologist at Scott and White, who also focuses concern upon this problem.

`I am relating this to you as a matter of information and would appreciate hearing from you as to any suggestions or remedies that you might have. I do not know whether the railroad has any type of program set up in this regard nor how well enforced it is if such program is in effect.'

"12. May 11, 1977, Dr. O.L. Hanson wrote Dr. Don J. Lynch at the Santa Fe Employees Hospital Association the following:

`Thank you for your letter dated May 6, 1977, with enclosure from Dr. R.L. Riess regarding hearing loss. For several years I have encouraged the Santa Fe to conduct a hearing conservation program including audiograms, etc. The program we have now is chiefly the furnishing of hearing protectors. No well-organized employee education program has been established. I am sure that we will establish one but as you may know there has been a big hassle in Washington for several years about hearing conservation requirement[s]. At any rate, I appreciate your letter and will circulate it where it will do the most good. We have had many offers of help from audiologist[s], otologists, safety engineers, etc.'

"13. June 10, 1977, Mr. Shaver wrote Mr. Cena the following:

`Dr. Hanson contacted Dr. Don J. Lynch of the Temple Hospital concerning the apparent hearing loss of employees at Cleburne Shops. Dr. Lynch stated that there appears to be a significant loss to those employees working within the Main Shop buildings. There does not seem to be a problem with those working in other areas.'

"14. July 22, 1977, J. Fitzgerald quoted Dr. Lynch of the Santa Fe Employees Hospital Association as reporting:

`We have investigated this to some extent and have found no clear-cut pattern in the records that have been reviewed, although, we get the impression that there may be a higher incidence of the problem in the machinist and surprisingly enough in the operating trainmen. However, we are doing an on-going study of all requests for audiograms and may be able to develop some statistics at a later date.'

"15. August 23, 1977, Mr. T.R. Lenert, Santa Fe Safety Department in Los Angeles, wrote Mr. Shaver the following:

`My purpose in writing this letter is manifold. First, I believe the time for us to begin an audiometric testing program is near. More and more we can expect engineers and their trainmen to claim total or partial hearing loss at the end of their career and blame it entirely upon their occupation. With no records to come to our aid, we will be entirely at their mercy.

`With an audiometric testing program, problems can be spotted before they reach a serious stage. Without such a program, hearing loss is usually quite serious before the loss is noticed. . . .'

"However, Santa Fe contends that the document included highly speculative language.

"16. R.D. Shaver received a letter from MTS Associates, Inc., hearing consultants, dated October 6, 1978, which read:

`It was good to talk with you recently at the National Safety Council meeting. The discussions on communication and motivation Wednesday morning were impressive. You are to be congratulated on the program.

`Data over a number of years indicate that hazardous noise exposure with resultant noise induced hearing loss are indeed the number one non-fatal health problem in the railroad industry. You, I, and Dr. Hanson have discussed the problem a number of times. A proposal was presented to Santa Fe in 1971. It has been demonstrated on a number of railroads that effective Hearing Preservation Programs can be initiated and continued in railroad facilities.

`The enclosure defines an effective Hearing Preservation Program. The only way to reduce further liability, is to prevent as much noise induced hearing loss as possible. The program must be complete, coordinated, controlled, and continuing with involvement of all employees at a given location.'

. . . .

"18. In connection with the investigation into the noise levels generated by the humpyard retarders in Barstow, professional consultants recommended to Dr. Khuri in February 1981:

`1. Instituting mandatory ear protection in the form of [ear] plugs for all individuals moving to the observation tower by the retarders, into and out of the fabricated house and to the parking lot, while the humpin process is in action. I am sending plugs to Tom Lennert.

`2. Immediately initiating a hearing testing program of all those individuals in the tower, and those moving around the retarders and in the bowel. This can be set up with Steve Barwick, Vector Management, Inc.

`3. Obtain a spectrum analysis of the high frequency noise generated by the retarders and calculate the risk of that noise to hearing.

`4. Construct a reflecting and isolating, moveable wall along the 250 ft of retarders.'

"19. On June 18, 1971, Dr. O.L. Hanson of Santa Fe wrote to Mr. S.R. Brittingham of Santa Fe Legal Department:

`Because of the Walsh-Healey Act of 1969, the Occupational Health and Safety Act of 1970 impending regulations of the Federal Railway Administration, the subject of acoustic trauma has surfaced. Activity in this field is increasing.. . . Since there are many legal factors involved in conducting a hearing conservation program, I would appreciate the opportunity of consulting with one of the members of your department.'

"20. In approving the submission of the sample noise readings to the EPA, Santa Fe's Vice-President of Operations, Cena, expressly acknowledged the forthcoming noise regulations regarding community noise pollution.

"21. April 15, 1974, C.R. Kaelin, Director of Technical Research and Development responded to The Association of American Railroads regarding a DOT proposed plan for a comprehensive study of noise related to locomotives and trains.

"22. December 19, 1975, Santa Fe received correspondence from The Association of American Railroads regarding `Control of Noise Exposure, Temperature and Exhaust Emission in Locomotive Cabs and Cabooses including an attached memorandum entitled "Noise Environment in Locomotive Cabs and Cabooses"' which states at paragraph 66 `it does not appear that a train crew noise exposure problem, in the context of OSHA regulations, exists in the railroad industry.'

"23. Santa Fe's efforts to establish a system-wide Hearing Conservation Program continued into 1982.

"24. In an April 7, 1982, memorandum, Santa Fe reported:

`The Santa Fe Heering [sic] Conservation Committee has been investigating numerous locations throughout the System to determine the extent and nature of occupational noise exposure among our employees. Occupational noise is a hazard in several ways:

a) Persons exposed to excessive levels of noise can develop permanent hearing damage;

`Occupational noise is also the subject of extensive federal regulation and FELA litigation. . . .'

"25. By August 9, 1982, Santa Fe in an internal memo announced that its hearing conservation program was ready for implementation and would be tested first in the Argentine ships and then implemented system-wide tentatively scheduled for early 1983. In connection with this announcement, Santa Fe reported:

`Discussion of the OSHA regulations (Appendix A) and hearing conservation initiated investigation into the nature of existing noise measurement results. Review of the data previously collected by the Medical and T&RD Departments indicated that excessive noise levels were present and that further measurements should be done. Noise surveys were then made at numerous locations. The results (Appendix B) were reviewed by the Committee, which concluded that noise exposures in excess of 85dba were present for a wide variety of tasks and that the need for a hearing conservation program existed.

`OSHA regulations apply to areas where the FRA does not exercise its jurisdiction. In the case of hearing conservation the FRA does not have requirements for hearing testing. The Law Department has recommended that hearing testing be done for both groups. It is their opinion that the major liability involves FELA claims and not regulatory penalties.'

"26. By the end of 1970, H.W. Chastain advised Mr. Baird that he had complied with a request to look into the procurement of hearing protectors for workers in areas of hazardous noise and had tested the NOISEFOE MARK IV at Santa Fe's Argentine Shops in Kansas City, Missouri. December 1970, Mr. H.W. Chastain reported that the employees were amazed as to the amount of noise the hearing protector kept out while retaining the ability to hear the voice of another nearby. Finally, Mr. Chastain stated that the intent was to provide the hearing protectors to those employees subjected to high noise levels such as load testing of diesel engines.

"27. December 8, 1980, Mr. Chastain advised Mr. Townley:

`THE N.S.A. NOISEFOE MARK IV hearing protector as examined by Mr. Baird, Dr. Hanson and representatives of this office and approved for use. . . . Please issue the necessary detailed instructions to Mechanical Managers, Mechanical Superintendents, Superintendents of Shops and all others concerned as to the approval of the hearing protector and how it is to be issued and used. . . .'

"28. August 17, 1979, Santa Fe requested the Safety Department to conduct some safety tests for the use of earplugs by its train, engine and yard men of the Operating Department under certain conditions.

"29. Prior to 1979, Santa Fe's policy was not to allow employees in Train, Engine and Yard Service to wear ear plugs.

"30. October 1981, the Santa Fe Hearing Conservation Committee was formed to review the August 21, 1981 Workplace Noise Exposure Regulations issued by the Occupational Safety and Health Administration.

"31. October 14, 1987, in the underlying lawsuit of John R. Smith v. The Atchison, Topeka and Santa Fe Railway Company, Santa Fe responded to interrogatory No. 24 as follows:

`INTERROGATORY NO. 24: Have any claims been made against this Railroad in the past for damages resulting from hearing loss? If so, for each claim made state:

a. Name of claimant and address;

b. Date of claim;

c. Job classification of claimant;

d. If suit was filed, name and address of attorney representing claimant;

e. How claim was resolved.'"

This was followed by answers identifying nineteen employee claims made against Santa Fe. Of the nineteen claims, four were settled, eight were pending in various courts, and the remainder were either open or time barred.

Findings 32 through 44 identify Santa Fe employees who settled or filed lawsuits against Santa Fe alleging hearing impairment, loss, or injury from December 10, 1976 to November 16, 1989.

"45. In the underlying lawsuit of David L. Wallace v. The Atchison, Topeka and Santa Fe Railway Company, Santa Fe responded to the following interrogatory:

`11. Please state the following:

a. How many employees of the Atchison, Topeka and Santa Fe Railway Company have been found, to the knowledge of Defendant, to have suffered a hearing loss within the past ten years? State their names, the nature of their employment (i.e., switchman, brakeman, engineer, etc.). [answer] There is no data available prior to 1982. Hearing Conservation Program tested 1,418 employees in 1982 of whom 73.3% were found to have hearing deficiencies in excess of 25 decibels. In 1983, 11,340 employees were tested; and . . . of them 75.8% were found to have deficiencies in excess of 25 decibels. In 1984 14,711 employees were tested; and of them 73.5% were found to have deficiencies in excess of 25 decibels.'

"However, Santa Fe points out that the data does not show that the hearing loss occurred while the employees worked at Santa Fe. Santa Fe further contends that a large percentage of the audiogram results were baseline results and that the results don't specify whether the hearing loss was NIHL or traumatic or for that matter [either one].

. . . .

"47. On April 17, 1978, OSHA issued a noise citation to Santa Fe in connection with the excessive noise generated by the diesel injector calibration machine at the Argentine locomotive shops. However, Santa Fe furnished protection within 2 weeks that would lower the level of noise exposure for the machine operator to levels acceptable to OSHA.

"48. A representative insuring agreement from one of the accident based insurance policies at issue provides:

`3. To indemnify the Assured for any and all sums which the Assured shall become liable to pay, and shall pay to any person or persons as compensation for injury or damage to person (whether such injury or damage be fatal or nonfatal) and injury or damage to property (excluding property of the Assured or in its care, custody or control) arising out of any accident or accidents caused by or growing out of the Assured's Railroad operations in the United States of America and all operations incidental thereto during the period commencing 15th January, 1971 Noon and ending 15th January, 1972 Noon, local standard time, at the various places where the Assured's operations are conducted, including also all loss and legal expenses (excluding salaries of employees and office expenses) incurred in investigation, adjustment or litigation of claims made against the Assureds as hereinbefore provided.'

"49. Santa Fe did not receive a single NIHL claim before or during the 1950's and only one in the 1960's, despite employing thousands of operational employees. Of the ten hearing loss claims between 1965 and 1971, nine were for traumatic hearing loss injuries — not NIHL, and the only claim that was arguably for NIHL settled in 1970 for $1,300.

"50. The U.S. Department of Transportation, prepared a report in 1976 regarding noise exposure among locomotive train crews which concluded that:

`From the limited data available, it does not appear that a train crew noise exposure problem, in the context of OSHA regulations, exists in the railroad industry. Equipment duty cycles and distance restrictions, combined with generally low speeds and long periods of idling preclude any general problem. However, noise levels do exist that are above those considered as a threshold for these regulations and therefore indicate that a thorough survey of crew environments is needed if for no other reason than to make clear the facts.'

"51. However, Certain Defendants also point out that under the conclusions and recommendations the study states that `if the noise exposure criteria were reduced to 85db at 8 hours as proposed by NIOSH, a serious problem could arise.'

"52. A scientific study of train service employees' hearing, based upon hearing loss data from 1982, found that the overall hearing loss in railroad employees is statistically no greater than that found in the general population, and stated that `trainmen are not typically exposed to hazardous occupational noise.' Clark and Popelka Study, Laryngoscope (November, 1989) Santa Fe's Exh. 4, at SF1002, SF1006-08.

"53. The Clark and Popelka Study tested trainmen with more than 20 years of experience and concluded:

`In this sample, 18.2% of the trainmen experienced a material impairment in hearing. Because this percentage was less than the percentage from non-occupational factors alone (20%), it can be concluded safely that there is no risk of developing an occupational NIHL by working as a trainman, and that the exposure of trainmen does not typically exceed 8-hour TWA of 80 dba.' Id. at SF1007.

"54. The Clark and Popelka Study reviewed previous published results from the 1971-1980 time period and stated:

`Taking into consideration typical locomotive crew workshifts and the relative durations of exposure caused by horn, brake and radio operations, Aurelius [in 1971] estimated typical crew noise exposures at 73% of the permissible exposure level [an 8-hour time weighted average (TWA) of 90 dba]. The data reported [in 1976 and in 1979] from the Remington studies of cab noise and the maximum levels reported by Urman [in 1978] are consistent with the measures reported by Aurelius.'

. . . .

`Taken together, the studies of locomotive cab noise published by Aurelius [in 1971], Remington [in 1979], Rudd [in 1976], Urman [in 1978] and the study of cab noise and crew noise exposure by Kilmer [in 1980] all suggest that typical noise exposure levels for locomotive crews is less [than] 8-hour TWA of 80dba, and therefore would not be expected to cause significant hearing loss.' Id. at SF1006-07.

"55. After 25 to 30 years of employment, over 80% of a normal population will experience high frequency hearing loss and approximately 20% of the population as a whole will develop a `material impairment of hearing' from non-occupational sources. Id. at SF1003-07.

"56. The Clark and Popelka Study relied on a population study contained in annex B of the ISO Standard R1999, which was comprised of data from surveys conducted over a 20-year span. Id. at SF1004-05. The ISO control group data, which is graphed in the article, shows that almost 90% of persons aged 55 to 64 have a 25 dB or more hearing deficiency at the higher frequencies. Id. at SF1004-06. In the article, the medical term `[m]aterial impairment in hearing' is defined as involving a 25 dB or more deficiency in both ears at three low speech-frequency ranges, i.e., the 500 Hz, 1000 Hz and 2000 Hz frequencies. Id. at SF1007. The study explains that, `for individuals 55 to 70 years, the risk of [material] impairment due to nonoccupational factors was 20%.' Id.

"57. As part of Santa Fe efforts to explore whether it had a noise problem in the 1970's, W.L. Barnow reported in November 1973 to R.D. Shaver that he had:

`used the sound-level instruments to measure noise levels in the mechanical facilities at Arkansas City, as well as on several Maintenance of Way machines. I have also made the equipment available to all safety supervisors on the Eastern Lines. As of the present time, I have not found any area which does not comply with the OSHA standards.'

November 26, 1973 letter from W.L. Barnow to R.D. Shaver, Santa Fe's Exh. 8.

"58. In explaining why Santa Fe considered hearing protection to be unsafe for railroad work, M.H. Haverty stated:

`[I]t is my concern that if we begin issuing ear plugs to engineers and head brakemen that we may create a condition that is actually adverse to our Safety program. In other words, our employees may not be able to hear emergency radio conversations or for that matter, warnings that another employee or employees in the cab of the unit may try to issue audibly. We could also set up a condition whereby our engineer could say that he did not hear the radio conversation of a trainman trying to do work in the field with a packset. This most assuredly could create an unsafe condition.'

July 3, 1978 letter from M.R. Haverty to J.R. Fitzgerald, Santa Fe's Exh. 9, at SFC64112.

"59. In response to an inquiry regarding locomotive noise tests, R.D. Shaver stated in an August 1974 letter to L. Cena that `[a]ll three tests have been completed and . . . dosimeter readings are all relatively low and indicate no noise problem whatsoever.' August 14, 1974 letter from R. D. Shaver to L. Cena, Exh. 14. However, Certain Defendants note that the three tests were conducted at certain points and the dosimeter only registered those sounds above 90dba.

"60. In 1977, Santa Fe's safety director stated that the `Safety Department feels that it is doing a pretty fair job identifying those areas where there is excessive noise and providing protection for employees working there.' June 3, 1977 letter from R.D. Shaver to L. Cena, Santa Fe's Exh. 15.

"In the same letter Shaver also stated that `there needs to be more work done with those operating our maintenance of way machines.' He also indicated that `(t)here is a difference, however, between a hearing protection program such as ours and a true hearing conservation program. Hearing conservation involves the audiometric testing of employees on an annual basis where they are exposed to the higher noise levels. We have been waiting for three years for OSHA to promulgate its new house standards. These standards spell out in detail what will be required of us in the way of conducting a hearing conservation program. We feel that such a program is necessary, but we would like to wait and see what will be required of us.' Id.

"61. In August 1977, C.R. Kaelin, Santa Fe's director of the Technical Research & Development Department, explained that his

`department has been taking sound level measurements in and around diesel locomotives for the past twenty years, and conclusions to date have been that locomotive operating personnel (engineers, brakeman, fireman) do not require ear protective devices for the noise levels normally prevailing inside locomotive cabs.'

August 11, 1977 letter from C.R. Kaelin to L. Cena, Santa Fe's Exh. 16.

"62. Sound level tests conducted by Santa Fe in early 1981 at one of its hump yard facilities showed that `the noise level for the average eight hour exposure was extremely favorable and that the peak levels attained were also for below the maximum allowable peak levels.' February 27, 1981 letter from H.D. Fish to D.G. McInnes, Santa Fe's Exh. 17.

"63. On August 21, 1981, OSHA issued amended workplace noise exposure regulations that specified for the first time the implementation of an Hearing Conservation Program (`HCP') for employees exposed to noise exceeding 85 dB(A). See 46 Fed. Reg. 162 (Aug. 21, 1981), Santa Fe's Exh. 18. These regulations lowered the noise standard from 90 dB(A) to 85 dB(A) and spelled out the requisite elements of an HCP. Id. The HCP amendments required initial baseline audiograms to be performed by 1984. Id.

"However, certain Defendants point out that the existing standard required a continuing effective hearing conservation program be implemented when employee exposure levels exceeded 90dba without regard to the use of hearing protectors.

"64. Santa Fe formed its Hearing Conservation Committee in October 1981. See October 26, 1981 letter from Dr. R.K. Khuri to Messrs. Briscoe, Fish, Fitzgerald, Santa Fe's Exh. 19; August 9, 1982 letter from Dr. R.K. Khuri to Messrs. Briscoe, Fish, Fitzgerald, Autrey, Mason with attached memo, Santa Fe's Exh. 20, at SFC62327.

"65. Santa Fe's Hearing Conservation Committee's task was to adopt and then to implement a system-wide HCP in compliance with the new OSHA regulations. Id.; see also Sept. 30, 1981 letter from C.R. Kaelin to Dr. R.K. Khuri, Santa Fe's Exh. 21.

"66. Santa Fe's Hearing Conservation Committee's first step was `gathering information, and performing extensive noise measurements in the different areas of the railroad.' October 26, 1981 letter from Dr. R.K. Khuri to Messrs. Briscoe, Fish, Fitzgerald, Santa Fe's Exh. 19.

"67. In an August 1982 letter summarizing the Santa Fe Hearing Conservation Committee's initial efforts towards the implementation of a system-wide HCP, Dr. Khuri explained:

`Discussion of the OSHA regulations (Appendix A) and hearing conservation initiated investigation into the nature of existing noise measurement results. Review of the data previously collected by the Medical and T&RD Departments indicated that excessive noise levels were present and that further measurements should be done. Noise surveys were then made at numerous locations. The results (Appendix B) were reviewed by the Committee, which concluded that noise exposures in excess of 85 dba were present for a wide variety of tasks and that the need for a hearing conservation existed.'

August 9, 1982 letter from Dr. R.K. Khuri to Messrs. Briscoe, Fish, Fitzgerald, Autrey, Mason with attached memo, Santa Fe's Exh. 20, at SFC62327.

"68. The Hearing Conservation Committee directed that `all employees who are exposed to an average noise levels above 85 dba, in an 8 hour average period for 30 or more days in a calendar year are subject to periodic testing.' Id. at SFC62343.

"69. Before launching the system-wide examination of workers, Santa Fe conducted a pilot program at a couple facilities in October 1982, which provided `valuable insight for determining the optimum protocol' under which to conduct the system-wide program. August 9, 1982 letter from Dr. R.K. Khuri to Messrs. Briscoe, Fish, Fitzgerald, Autrey, Mason with attached memo, Santa Fe's Exh. 20, at SFC62328; March 3, 1983 letter from Lew Malter to Dr. R.K. Khuri, Exh. 23.

"70. The trial run at the Kansas City Terminal and Argentine Shops included hearing examinations conducted on 769 mechanical personnel and 317 operating personnel. November 30, 1982 letter from Dr. R.K. Khuri to D.G. Ruegg, Santa Fe's Exh. 24.

"71. In a February 21, 1984 letter discussing the HCP, D.P. Valentine explained:

`Audiometric testing is required annually for all mechanical, train, engine, yard, track, B&B and terminal services employees (excluding clerical). Two mobile testing units are provided and staffed by our contractor, Health Evaluation Programs, Inc. (HEP). A large (two booth) test trailer will visit 38 locations in 1984. It will be followed six to eight weeks later by a small trailer (one booth). The former will visit major stations and the shops. The latter will provide coverage for small stations and follow-up for employees who are unavailable at major locations to be on the large trailer.'

February 21, 1984 letter from D.P. Valentine to Messrs. Spann, Merritt, Caldwell, Dixon, Moser, Santa Fe's Exh. 25, at SFC65870.

"72. In a 1978 letter to C.H. Harrington, C.R. Kaelin reported:

`Federal Railroad Administration Inspector Percy Carney on August 14 and 15, 1974, measured noise levels in locomotives no. 3502, 3518 and 6612 while in regular train operation between Chicago and Chillicothe, Ill. These locomotives, times and locations for the inspection were selected by the inspector from information contained in complaints received from his organization. At the time of the inspection, no personal hearing protective devices of any kind were being used by the locomotive's personnel. Following the inspections, no notification or violation or citation was issued, acknowledging in effect that no Federal noise regulation was being exceeded.'

Id.

"73. A report sent to Santa Fe in April 1974 by Wayne Kitchen, industrial hygienist for the Kansas Department of Labor, regarding results from sound level tests taken at the Argentine Yard and while onboard a train engine, stated:

`It is the finding of this testing survey that only under certain operations did the sound level exceed the permissible standards set forth by OSHA (Table G-16). These operations included the blowing of the train whistle, the release of air pressure used for braking and the radio loud speaker. These readings, however, are so intermittent that they would fall within the acceptable limitation for a time weighted average.'

April 26, 1974 letter from Wayne Kitchen to H.L. Rogers with attached report, Santa Fe's Exh. 27, at SFC64448.

"74. In an 1977 letter to L. Cena, C.R. Kaelin in response to a request for information on noise levels in locomotives and use of personal noise protective equipment by locomotive engineers, stated:

`We have had State and Federal Inspectors ride locomotives of their choosing in regular service for compliance with OSHA noise standards. . . . In the two cases mentioned below, the locomotives were selected by the inspectors for being of the same type and class as that mentioned in complaints received by them regarding noise levels. In all cases of regulatory authorities inspection for noise levels in locomotive cabs, as far as we know we have received no objections or complaints.'

August 11, 1977 letter from C.R. Kaelin to L. Cena, Santa Fe's Exh. 16.

. . . .

"77. The only Santa Fe NIHL case that went to trial was Morris D. Derting, Leslie E. Holly and Richard C. Owens v. The Atchison, Topeka & Santa Fe Railway Company, No. 249-188-89, Johnson County, Texas (the `Cleburne Case'). See Cleburne Voir Dire Transcript, Santa Fe's Exh. 32.

"78. Mr. Dana Kirk, an attorney representing the plaintiffs in the Cleburne Case, stated to the jury:

`Ladies and gentlemen, the Santa Fe Railroad, of course, is headquarterd [sic] in Chicago, Illinois and had its facility down here. . . . And we believe the evidence will indicate that there was an enormous lack of any conduct on the part of the management and medical people at the head office of the Santa Fe Railroad in Chicago that largely caused the hearing loss that these three men and other employees at the railroad here in Cleburne have suffered. And, I guess, what I'm asking you is: Does anyone have a problem evaluating the conduct of the people up in Chicago where these decisions were being made about not instituting the hearing conservation program in a timely way and about not providing proper hearing protection in a timely way and about not providing proper training to these men timely and about not providing them a safe place to work in terms of the noise that they were exposed to? Does anyone have a problem holding the Santa Fe Railroad responsible for the decisions that were made by the hierarchy of the Santa Fe Railroad in Chicago, Illinois? . . . [D]oes anyone have a problem with holding the Santa Fe Railroad responsible for that conduct of the people in Chicago even though the local people really didn't do anything particularly wrong? They were simply following orders.'

Id. at SF168-69.

"79. David Pels, a Santa Fe defense attorney, testified regarding the NIHL cases against Santa Fe:

`Q: Did you make a separate assessment in each case as to the potential that the plaintiff could prove that the railroad was liable?

`A: Fairly on in the litigation, particularly with the Helm Pletcher firm, it became apparent as to what their theory of liability was, and it was basically the same in all of their cases.

`So I didn't have to go back and reinvent the wheel, if you will, in terms of a liability assessment. The evaluations from that point forward were—pertained to primarily damages.

`Q: And that theory of liability that you are referring to is the Helm Pletcher's allegation that the Santa Fe Railroad provided a null and safe workplace?

`A: Well, failure to timely implement a hearing conservation program or an effective hearing conservation program.'

Id. at 40.

"80. David Pels testified that all the NIHL claims had the same theory of liability. Id. at 46, 60.

"81. David Pels testified that in every case the Claimants alleged that the railroad's untimely HCP caused their NIHL. Id. at 63.

"82. Robert Attridge, Santa Fe's Director of Occupational Claims, testified:

`every time I would meet with the plaintiffs' attorneys, they would continue to tell me over and over again to the point of ad nauseam that, you know, it was Santa Fe's lack of timely implementation to provide a Hearing Conservation Program that caused these people's loss dating back to Dr. Ar[a]m Glorig's famous speech in San Francisco in 1966.' Attridge Dep., Santa Fe's Exh. 34, at 102.

"83. Charles Bruce, Santa Fe's Regional Claims Manager, also testified that the alleged cause of the plaintiffs' NIHL claims was Santa Fe's untimely HCP. Bruce Dep., Santa Fe's Exh. 36, 15 190, 211.

"84. Santa Fe settled its NIHL claims due to potential liability based on the untimely HCP theory of liability and allegation by the plaintiffs. Attridge Dep., Exh. 34, at 105-106, 266-268; Bruce Dep., Exh. 36, at 199; see also Pels Dep., Santa Fe's Exh. 33, at 40, 54-55.

"85. One of the plaintiffs' lawyer who filed hundreds of claims stated: `[I]t was our hope to be able to prove that [the lack of HCPs among railroads] was a concerted decision. That was not proof that we were ever able to find, at least with regard to this railroad.' Hanks Dep., Santa Fe's Exh. 30, at 86.

"86. When asked, `Did you uncover any evidence that Santa Fe conspired with other railroads?,' the plaintiffs' attorney testified: `We did not.' Id.

"87. In connection with allegations of intent in the form of misrepresentation or conspiracy, Santa Fe's lead outside defense counsel testified: `[i]t may appear in the pleadings, but it was something that from my experience was never pursued by any of the plaintiffs.' Pels Dep., Santa Fe's Exh. 33, at 82.

"88. Santa Fe's policies do not contain any definition of the term "accident." See, e.g., Lloyd's Policy No. 614/53930, 12/15/69-12/15/70, Exh. 37; Lloyds Policy No. 614/43600, 2/15/74-2/15/75, Exh. 38; Lloyds Policy No. 614/50439, 10/16/67-10/16/70, Santa Fe's Exh. 39.

"89. Santa Fe's policies omit the phrase contained in Comprehensive General Liability (CGL) policies which requires that an `accident' be `neither expected nor intended from the standpoint of the insured.' See Id.

"90. Santa Fe's policies provide coverage for many injuries which do not result from accidental means, including injury caused by a variety of intentional torts:

`false arrest, false imprisonment, false eviction, detention, discrimination, malicious prosecution, humiliation, invasion of right of privacy, libel, slander, or defamation of character; and as respects claims arising out of the assured's advertising activities, infringement of copyright, of title, or slogan, piracy or unfair competition, idea, misappropriation under an implied contract, committed or alleged to have been committed.'

See, e.g., Lloyds Policy No. 614/53930, 12/15/69-12/15/70, Santa Fe's Exh. 37, at SF12.

"91. Under Santa Fe's policies, both the Santa Fe railroad and all of its `employees' are Assureds for purposes of coverage. See, e.g., Lloyds Policy No. 614/43600, 2/15/74-2/15/75, Santa Fe's Exh. 38, Section V.B. at SF29.

"92. The documents relied on by the insurers confirm that the noise on Santa Fe's railroad was generally believed to be within safe levels. See Ins. Mem. at 12, 19. The documents cited by the insurers state: (i) `[i]n nearly all cases the noise levels fall around 85dba and does not exceed the 90dba allowable for an 8 hour work day;' (ii) `the inspectors did not consider the OSHA standards to be violated;' (iii) `the maximum unprotected daily limits specified by OSHA were not being exceeded and that Santa Fe was not in violation;' and, (iv) `[m]ost freight and passenger runs today are not a duration that would expose the crew to over 85dba for an eight hour period.' Id.

"93. A 1961 letter from Robert S. Judson, Secretary of Lloyd's Underwriters' Fire & Non-Marine Association, to the London underwriters and deputies quotes an excerpt from the October 13, 1961 issue of Smart's Confidential Insurance Bulletin that there were a `flood of loss of hearing claims' being presented at that time, and states that `some 3,000 of these claims are expected to have been filed within the next year [1962].' See November 3, 1961 letter from Robert Judson, Santa Fe's Exh. 50, at SF713-SF714. See Declaration of Michael T. Novak, Santa Fe's Exh. 51, regarding use of this letter.

"94. One lead counsel for Certain Underwriters at Lloyd's, London and London market insurers stated in August of 1989 that `[a]s recently as five years ago [NIHL] hearing claims were virtually unknown except as incidental injuries related to other traumatic injuries.' Walton N. Smith of Lord, Bissell & Brook, `Hearing Loss Litigation Under FELA,' presentation to the American Bar Association Section of Tort and Insurance Practice Rail and Motor Carriers Committee at 1 (Aug. 8, 1989), Santa Fe's Exh. 52, at SF235.

"95. A May 17, 1955 letter to London underwriters identified two documents—the Association of Casualty and Surety Companies' Special Hazards Bulletin on the Subject of Industrial Deafness (March 1955) and Theodore C. Waters, Occupational Impairment of Hearing (January 1955). These reports discuss hearing impairment resulting from noise in the workplace. See Santa Fe's Exh. 53, at SF649-SF703. See Novak Decl., Santa Fe's Exh. 51, regarding use of the May 17, 1955 letter and its attachments.

"96. Employers Insurance of Wausau published a booklet in 1967 entitled `Industrial Noise and Hearing Protection' which states it was

`intended to help you . . . solve one of the most perplexing problems facing American industry and its employees[,] . . . the slow destruction of hearing that takes place in men and women who work every day in locations with high levels of noise.'

`Industrial Noise and Hearing Protection,' Employers Insurance of Wausau (1967), Santa Fe's Exh. 54, at SF632.

"97. The booklet `Industrial Noise and Hearing Protection,' published in 1967 by Employers Insurance of Wausau, stated that `[e]xcessive noise has become a threat to the hearing of millions of American men and women. The growing use of loud machinery in industry . . . is slowly wearing away the hearing of many employees.' Id. at SF633.

"98. Roger Quigley, corporate representative of ISLIC, testified as follows:

`Q: It was during this time, I believe you testified, from `73 to `76, that hearing loss issues became a factor or consideration in at least some types of workers' compensation policies, correct?

`A: Some types of workers compensation submissions I would call them.

`Q: And is it also true that during this time period, that the industry in general, the insurance industry in general was becoming more familiar with hearing loss issues in at least in the context of workers' compensation policies.

. . . .

`A: I've probably already stated that the question of compensability of hearing loss was becoming more of an issue.

Quigley Dep., Santa Fe's Exh. 57, at 64.

`Q: One of the risks that one has to consider in writing a comp policy is whether claims for noise-induced hearing loss may be covered under such a policy?

. . . .

`A: Well, again, you're entering into the word `may,' and if memory serves me correctly, in the early `70's, that was an unresolved issue that was not a cause of great concern other than the recognition that it is a possibility. So there may be certain kind of businesses that one would not want to write a workers' compensation on. But others, either the rating structure or the rating approach might have made that risk writeable because hearing loss cases historically have never involved a great deal of money.

Id. at 145-146.

`Q: As an insurance professional, you were aware that industrial hearing loss was an emerging issue, correct?

. . . .

`A: The concern over it was an emerging issue.

Id. at 151-152.

`Q: Regardless of the type of business you were writing at the time, regardless of that, in the mid to late `70's, did you know that excessive regular noise in the workplace could cause hearing loss?

. . . .

`A: As I best recall, that was a developing issue on a lot of workplace exposures, not the least of which was the creation of OSHA in that kind of a time frame and what OSHA was going to do in a number of areas. But because we were not writing the kind of business such as a shipbuilding firm, it was not a focal point for us. The construction business did not have that same kind of concern as an example.'

Id. at 175-176.

"99. Roger Prickett, another corporate representative from ISLIC, testified:

`Q: When was the first time that you had occasion to handle hearing loss claims?

`A: I don't recall.

`Q: Would it have been the time that you were involved with Liberty Mutual?

`A: Yes, it would.

`Q: Would it have been earlier than 1975?

`A: Yes.

`Q: Would it have been earlier than 1970?

`A: Yes.

Prickett Dep., Santa Fe's Exh. 58, at 36-37.

`Q: Were they citing any one specific event like a shotgun blast, or were they alleging that this was a condition that developed over a period of time?

`A: I believe they were alleging that their injury occurred as the result of repetitive occurrence over time in the course of their employment.

Id. at 39.

`Q: By the early 1970's, did Roger Prickett know that hearing loss in the workplace was a potential risk for insurers?

. . . .

`A: In the worker's compensation environment, yes.

`Q: By the early 1970's, did Roger Prickett know that long term exposure to industrial noise could create a risk of hearing loss?

. . . .

`A: Yes.'

Id. at 63-64.

"100. James Laughlin, corporate representative for Lexington Insurance, testified that he believes some railroads submitted reports to Lexington that reflected that they received more than five NIHL claims in some calendar years between 1960 and 1983. Laughlin Dep., Santa Fe's Exh. 59, at 169-71.

"101. Robert Attridge testified that the first NIHL claims he could recall was a group of 35 that were brought in 1989. Attridge Dep., Santa Fe's Exh. 34, at 62-63.

"102. Plaintiffs' attorney Steve Hanks testified that the first long-term NIHL claims he was involved in were in 1982 or 1983. Hanks Dep., Santa Fe's Exh. 30, at 26-27.

"103. Dr. Raja Khuri testified that the first NIHL claims he became aware of were three claims in 1980 or 1981 at the time the HCP was being planned. Khuri Dep., Santa Fe's Exh. 66, at 162, 184."

Before we address each of the issues raised by the appellants, an overview of this case would be helpful.

Substantive issues. In this declaratory judgment action, Santa Fe sought various rulings from the district court on the questions of whether there is coverage to indemnify Santa Fe for its NIHL payments and, if so, how much. The district court concluded that there is coverage. Given that there is coverage, the question is which policies must respond to the claims. The district court concluded that indemnification is available to Santa Fe from all liability policies at issue.

In their shorthand appellations, the substantive issues on which the district court ruled in favor of Santa Fe were nonfortuity, known loss, late notice, number of occurrences, trigger of coverage, and horizontal exhaustion and allocation. Nonfortuity would bar coverage for losses not arising from an accident, within the meaning of the policies. Known loss would bar coverage for losses that already have occurred when the insurance is purchased, based on the theory that insurance policies are contracts for some contingency or act to occur in the future. Late notice would bar coverage where insurers were not timely advised of occurrences as required by the policies. The number of occurrences involves a determination whether a single date can be assigned to a series of events, which would allow Santa Fe to aggregate NIHL claims for the purpose of satisfying its self-insured retention (SIR) limits. The trigger of liability coverage is the event that implicates a policy or policies. The question of horizontal exhaustion is whether Santa Fe must pay the amount of its self-insured retention in each policy period before looking to coverage under the insurance policies. Allocation is the method of determining which polices must respond, the alternatives considered in this case being joint and several liability among all defendant insurers or particular claims against particular insurers or a pro-rata, time-on-the-risk approach.

On these issues, the district court decided:

—Santa Fe's loss was not expected;

—Santa Fe had not suffered a known loss;

—The timing of notice did not bar coverage;

—There was a single occurrence;

—There was a continuous trigger of coverage;

—Santa Fe must pay only one self-insured retention; and

—Liability is joint and several among insurers.

The outcome is entirely favorable to the insured, Santa Fe: There is coverage, every policy issued by defendant Insurers to Santa Fe is implicated, and Insurers are jointly and severally liable for all amounts above Santa Fe's self-insured retention for one policy period.

Law of the forum. On each of these issues, the district court applied the law of the forum. Before turning to the substantive issues, the parties briefed the question of choice of law and the district court considered whether Kansas or Illinois law should govern. The district court concluded that Kansas law would govern.

The district court's decision on choice of law was the latest in a series of choice of law decisions that have been made in this case. In his 1995 Memorandum Decision and Order, District Judge Terry Bullock applied Kansas law, citing Shutts v. Phillips Petroleum Co., 235 Kan. 195, 221, 679 P.2d 1159 (1984) aff'd in part, rev'd in part 472 U.S. 797, 86 L. Ed. 2d 628, 105 S. Ct. 2965 (1985), for the principle that the law of the forum would be applied absent some compelling reason to do otherwise. Noting that the United States Supreme Court reversed that portion of Shutts, this court discussed factors for the district court's consideration upon remand. This court noted an unpublished Kansas Supreme Court opinion, No. 75,227, filed January 31, 1997, in which the rule of lex loci contractus was applied to an insurance agreement and noted that "[t]he lesson to be drawn from Shutts is that a state court's application of its own law on the ground that it is not in conflict with the law of other jurisdictions is sound only when the lack of conflict has been satisfactorily studied and established." Atchison, Topeka & Santa Fe Railway Co. v. Stonewall Insurance Co., 1997 WL 1048134, p. 6. The majority of this court's discussion, though, was devoted to consideration of waiver in that Insurers' conduct was inconsistent with advocacy of application of another state's law as well as with reservation of the issue of choice of law. This court stated:

"These circumstances, coupled with Insurers' intermittent express reliance on Kansas law, are inconsistent with a genuine or valid challenge to the district court's applying Kansas law. No reason has been presented why the choice of law issue cannot be waived, and authority for enforcing a contract clause which specifies applicable law certainly indicates that choice of law is a matter within the hands of the parties. On remand, the matter needs to be resolved by the district court." Atchison, Topeka & Santa Fe Railway Co., v. Stonewall Insurance Co., 1997 WL 1048134, p. 7.

Upon remand, District Judge Fred Jackson determined that Kansas law would govern. In his order of December 1998, the district judge concluded that the "service of suit" provision in the policies did not constitute a choice of law provision, that Insurers had not waived the choice of law issue before appeal of the 1995 decision to this court, but, for the following reasons, that Insurers waived the issue on remand:

"[I]n keeping with the spirit of the Kansas Supreme Court opinion in this case and the evident intent for the district court to find a conflict in the law before applying choice of law rules, this Court finds that since remand of this case and the clear direction by the Supreme Court that Insurers demonstrate a conflict, that Insurers have failed to establish and analyze a conflict and thus, have waived the issue of whether a conflict exists. Without a demonstrated conflict, this Court will not analyze its choice of law rules. Therefore, Kansas law will apply to this case."

In March 1999, District Judge Nancy Parrish ordered the December 1998 decision to "be altered to allow the Defendant Insurers time to demonstrate a conflict in the laws between Kansas and Illinois and for the Plaintiff to respond, if necessary, to demonstrate a lack of material conflict before Kansas Law is applied." Later, in November 1999, the district court filed a Memorandum Decision and Order determining that Kansas law would apply to all issues. In reaching the decision, the district court compared Illinois and Kansas law on each of the substantive issues, finding a conflict only with regard to the effect of late notice. The district court concluded that the clear public policy of this state, "not to allow insurance companies to escape liability under a technicality when insureds provide notice late," allowed application of Kansas law to the late notice issue despite the conflict.

As previously noted, all issues were decided by the district court on motions for summary judgment or partial summary judgment. Summary judgment is appropriate if there are no genuine issues of material fact and the movant is entitled to judgment as a matter of law. K.S.A. 60-256(c). Our standard of review is de novo. Associated Wholesale Grocers, Inc. v. Americold Corp., 261 Kan. 806, 820, 934 P.2d 65 (1997). Interpretation of an insurance policy is a question of law. We construe insurance policies in a way that will give effect to the intention of the parties. Absent ambiguity, we enforce the policy as made. All pertinent provisions of an insurance policy must be considered together, rather than in isolation, and given effect. Brumley v. Lee, 265 Kan. 810, 812-13, 963 P.2d 1224 (1998).

A few additional comments are in order. At the onset, we acknowledge the difficulty of our task, which is somewhat akin to fitting a square peg in a round hole. This case not only raises issues of first impression for this court, but counsels' arguments, in part, appear to conflict with the court's findings of uncontroverted facts. For example, the trial court identifies the policies of insurance as "programs of excessive insurance coverage," "excess policies," "successive layers of coverage," "indemnify in excess" of certain amounts. Despite these findings and notwithstanding the obvious structuring of the policies as excess coverage policies, Santa Fe argues they are not excess policies but rather general liability policies covering "all losses." The argument is based on the "All Sums" and "Other Insurance" language in the policies. It seems the argument is made that the policies are illusory, in that they are not what they appear to be. Further, that the SIRs are not other insurance, but are deductibles. Black's Law Dictionary 584 (7th ed. 1999) defines an excess clause as "[a]n insurance-policy provision—usu. contained in the `other insurance' section of the policy—that limits the insurer's liability to the amount exceeding other available coverage. This clause essentially requires other insurers to pay first"; and excess limits as "[i]nsurance coverage against losses in excess of a specified limit." Black's Law Dictionary 584 (7th ed. 1999). In Americold, this court held that "an excess insurance policy is one that provides that the insurer is liable for the excess above and beyond that which may be collected on primary insurance." 261 Kan. 806, Syl. ¶ 3. With that in mind we turn to the issues raised by the insurers.

Appellants first challenge the trial court's application of Kansas law. There are constitutional limitations on choice of law. "[F]or a State's substantive law to be selected in a constitutionally permissible manner, that State must have a significant contact or significant aggregation of contacts, creating state interests, such that choice of its law is neither arbitrary nor fundamentally unfair." Allstate Ins. Co. v. Hague, 449 U.S. 302, 312-13, 66 L. Ed. 2d 521, 101 S. Ct. 633 (1981). Where clearly established and material conflicting law has been brought to the attention of the trial court, the Due Process Clause of the Fourteenth Amendment and the Full Faith and Credit Clause of Art. IV, § 1, require the forum to respect the laws of other significantly interested states. The trial court's duty is to settle on a choice of law that is neither arbitrary nor fundamentally unfair, and an important element of fairness is the expectation of the parties. Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 822, 86 L. Ed. 2d 628, 105 S. Ct. 2965 (1985).

District Judge Bullock decided the issue on the basis of a principle from this court's decision in Shutts that had been overruled by the United States Supreme Court. As a result, when District Judge Bullock's decision was appealed, this court discussed Shutts in its opinion, along with Kansas courts' adherence to the lex loci contractus choice of laws rule for insurance policies, the service of suit clause, and Insurers' conduct relative to waiver. Upon remand, District Judge Jackson ruled that the service of suit clause was not a choice of law provision and that Insurers's conduct amounted to waiver. Upon reconsideration, District Judge Parrish concluded that on the subject matter of this litigation, no conflicts exist between Kansas and Illinois law that would preclude application of Kansas law.

With the exception of the issue of late notice, which the district court decided on a public policy basis, the no conflicts decision was driven almost entirely by the lack of Kansas law. The reasoning was that, where there is no Kansas law, Kansas law does not conflict with Illinois law. For example, the district court determined that Kansas courts have not recognized a "known loss" doctrine, and there is no Kansas law on the issues of trigger of coverage, allocation, horizontal exhaustion, and number of occurrences. In other words, Kansas courts have not had occasion to address a majority of the dispositive issues in this case. Because there were no conflicts in existing law, the district court determined that the law of the forum would govern without applying Kansas' choice of law rule.

The district court's decision that forum law should apply in the absence of conflicts was, in the abstract, sound. However, that decision must be made after a satisfactory review of Illinois law. "There can be no injury in applying Kansas law if it is not in conflict with that of any other jurisdiction connected to this suit." Shutts, 472 U.S. 816. No choice of law question is presented when the laws of the involved states do not differ on substantive issues. In the particular circumstances of this case, however, it was foreseeable that the vacuum of Kansas law might be filled with law that conflicted with Illinois law on substantive issues. And that is what happened. The district court's decisions on several of the substantive issues created conflicts with Illinois law.

Thus, we must resolve the substantive issues before we can resolve the choice of law question. If we establish that Kansas law is in accord with Illinois law, there would be no choice of law question. As a practical matter, even if the Kansas law that is established does not comport completely with Illinois, if the result of the application of Kansas law is the same as it would be were Illinois law to be applied, there is no choice of law question as a practical matter. For this reason, we turn first to the substantive issues raised in this appeal.

We first consider the trial court's conclusion that the 3,800 hearing loss claims constitute a single occurrence.

The insurance policies, subject to their terms and conditions, require the Insurers to indemnify Santa Fe for sums Santa Fe became liable to pay for injuries arising out of an accident or accidents or arising out of any occurrence or occurrences. The policies define occurrence as "one or more accidents or series of accidents arising out of or resulting from one event" or "[o]ne or more accidents or disasters and/or series or accidents or disasters arising out of or resulting from one event."

In Santa Fe's view, the event is its negligent failure to timely implement an effective hearing conservation program that would have protected its workers from the excessive noise inherent in railroad operations. In Santa Fe's view, there is one event and one occurrence.

Insurers contend that the event is the exposure to the injurious noise. Because each worker has a unique work history, according to Insurers, the many exposures to different noises at different times and places constitute multiple occurrences. Insurers contend that they are not liable for the sums Santa Fe paid to NIHL claimants because none of the individual multiple occurrences led to a claim for an amount larger than Santa Fe's self-insured retention level.

The district court agreed with Santa Fe that there is a single occurrence. First, the district court examined a number of cases and a treatise supporting the proposition that a corporate failure to act may be considered an event within the meaning of the language of the policies, and it rejected the Insurers' contention that, even if Santa Fe's failure is an event, it is not a single occurrence because there were multiple failures. Second, the district court considered how broadly to analyze the cause of the injuries in determining whether there was one continuing cause that resulted in all of the injuries. The district court found support for a broad causal analysis in a number of Kansas cases in which the phrase "arising out of" was construed to allow liberal causation. Third, the district court discussed this court's 1997 decision and rejected Insurers' contention that this court adopted the multiple occurrence ruling from CSX v. Continental Insurance, 343 Md. 216, 680 A.2d 1082 (1996).

As we noted in our 1997 opinion in the present case, Maryland's highest court affirmed a trial court's submitting the question of the number of occurrences for NIHL claims to a jury. There, as here, the railroad and insurers agreed that occurrence should be defined according to the cause of the injury, but disagreed on what the cause was. The railroad contended that the cause of NIHL was the company's failure to require hearing protection, but the jury identified more than 20,000 occurrences. The district court correctly concluded that CSX was quoted by this court in 1997 to illustrate the need for discovery when an argument can be made that an issue is one of fact or mixed fact and law. The district court also noted that the Maryland court defined cause more narrowly than it has been defined in Kansas cases. Fourth, the district court concluded that NIHL claimants' numerous and varied allegations all would have been addressed by a single hearing conservation program so that multiple allegations are not contrary to the single occurrence theory. Finally, based on the evidence before it, the district court accepted Santa Fe's suggestion that "the single theory of liability upon which it settled all the NIHL claims was the failure to timely implement" a hearing conservation program.

On appeal, Insurers' position is that the claims did not arise out of one occurrence as a matter of law and the number of occurrences is a question of fact. They generally rely on CSX, 343 Md. 216, and Norfolk & W. Ry. v. Acc. & Cas. Ins. of Winterthur, 796 F. Supp. 929 (W.D. Va. 1992) (applying Virginia law), aff'd in part, dismissed in part 41 F.3d 928 (4th Cir. 1994).

In our 1997 opinion in the present case, we stated the following with regard to CSX, the Maryland case:

"CSX . . . was an action brought by the railroad's excess liability insurance carriers, seeking a determination as to coverage for indemnification of CSX's payments for employees' NIHL claims. As in the present case, the railroad and insurers agreed that occurrence should be defined by focusing on the cause of the injury, but they disagreed on what the cause was. The railroad argued that NIHL `resulted not only from the unprotected exposure to hazardous noise, but also from the failure of each policyholder management group to mandate hearing protection.' 343 Md. at 223-24 . . . . It attempted to prove to the jury's satisfaction that the cause of NIHL was its failing `to mandate protection from hazardous noise.' 343 Md. at 225 . . . .

"The question was submitted to a jury. The jury sided with the insurers in identifying the cause, determined that there were at least 20,235 occurrences, and concluded that the railroad had not proved that noise induced hearing losses are accidents within the meaning of the policies. In addition, the district court concluded that NIHL is an occupational disease and entered summary judgment on that issue. Thus, the final judgment declared `that the insurers were under no obligation to indemnify [the railroad] for the settlement of the NIHL claims.' 343 Md. at 230 . . . .

"On appeal, the Maryland Court of Appeals commented: `This case makes clear . . . that, in many cases, the cause of an injury or damage is neither obvious nor easily discernable.' 343 Md. at 235 . . . . The court noted that both sides cited cases applying the cause test, but that the railroad emphasized cases `in which the event that has been determined to be the proximate cause and the resulting damage or injury do not have a direct nexus.' 343 Md. at 235 . . . . The insurers in CSX relied on Norfolk & W. Ry. v. Acc. & Cas. Ins. of Wintershur . . . , just as Insurers in the present case do. 343 Md. at 236 . . . .

"The Maryland appellate court narrowly defined its task, ruling out determining what caused the NIHL claims, and fixed on determining the adequacy of the jury instructions. In approving the instructions, the court concluded that the instructions did not foreclose the railroad's arguing that the `common cause' of all the injuries, even where the proximate cause of each injury was a particular noise or combination of noises, was its failing to institute a prevention program. 343 Md. at 243 . . . . It went on, though, to express serious doubts about the soundness of the railroad's theory. . . .

. . . .

`CSX is not, as we have previously noted, the first insured to attempt to generalize the cause test for the purpose of maximizing insurance coverage. As noted earlier, the court in Norfolk & Western Railway Co., supra, 796 F.Supp. 929, rejected, as nonsensical, an argument similar to that advanced here. In that case, the railroad argued that it was negligent with respect to the handling of excessive noise in the work place, which, it claimed, was the proximate cause of the NIHL claims brought against it. . . .

. . . .

`The "one occurrence clause" was not drafted to protect railroads from NIHL claims. Its focus is much more general. Rather than a matter of interpretation, the factual assertions made by CSX were matters for the jury to determine when considering whether to aggregate the NIHL claims. Moreover, and, in any event, the evidence upon which the factual assertions made by CSX were based was presented to the jury. It is apparent from its verdict that the jury rejected those assertions.' 343 Md. at 247-51 . . . ." 1997 WL 1048134 at 5.

In Norfolk, the district court held that, under Virginia law, noise-induced hearing loss was an "occupational disease" for purposes of employer's excess liability insurance policies. 796 F. Supp. 929. The federal Court of Appeals agreed. 41 F.3d 928. With regard to the issue whether the railroad's negligence was a single "occurrence" within the meaning of the policies' liability limits, the district court denied the railroad's motion for partial summary judgment. The district court expressed the opinion that the railroad's position, that the single occurrence was the systematic, unprotected exposure of workers to high levels of noise, was illogical. Recognizing the significance of the issue, the federal Court of Appeals stated:

"That the distinction between a single and more than one occurrence might be critical under some policies is evident because of the wording of Paragraph 4 of Policy No. 509/68 D.D.1040 which attaches liability to the insurance company following the payment by N & W of the self-insured requirement `in respect of each occurrence.' Using N & W's definition of `occurrence,' if the numerous noise-induced hearing loss claims against N & W, even if each is for less than the self-insured requirement, are all caused by the same occurrence, the total amount paid by N & W on account of the aggregate of those claims would be the figure used in ascertaining the self-insured requirement." 41 F.3d at 932.

But, because that policy provided for the aggregation of "occupational disease" claims, the Court of Appeals concluded that the issue with respect to "occurrence" was immaterial so that the issue was moot. 41 F.3d at 932.

In this case, Insurers take issue with the district court's view of Kansas case law. They cite Wilson v. Ramirez, 269 Kan. 371, 2 P.3d 778 (2000), for their position that Kansas courts apply a narrow cause test rather than a broad causal analysis. On five separate occasions over a period of several years, Dr. Ramirez failed to diagnose a malignant lesion on Wilson's lower lip. In suing Ramirez, Wilson alleged five counts of malpractice, as if he had five separate claims. The court concluded that the doctor's failure to diagnose was the one and only "cause of Wilson's injuries, and thus, [there was] one claim or occurrence for purposes of coverage." 269 Kan. at 381. Of its analysis, the court stated that it was "consistent with the `cause theory' adopted by a majority of the cases we have discussed. The cause theory asks if there was but one proximate, uninterrupted, and continuing cause that resulted in all of the injuries and damages. [Citations omitted.]" 269 Kan. at 380-81. For several reasons, we are not convinced that Wilson adds support to Insurers' case. First, the Wilson decision was based on health care provider insurance statutes, K.S.A. 40-3401 et seq., not relevant in the present case. Second, and more important, the outcome of Wilson does not support a multiple occurrence argument. Faced with minimal policy limits, Wilson sought to maximize his recovery by claiming multiple occurrences. The court rejected his theory: "Under the facts of this case, where the cause of plaintiff's injury was failure to diagnose and the defendant physician continually failed to diagnose, there was but one proximate, uninterrupted, and continuing cause which resulted in all of the plaintiff's injuries and damages." 269 Kan. 371, Syl. ¶ 3. The court added: "When there is only one cause of plaintiff's injuries, there is only one claim or occurrence for purposes of medical malpractice insurance coverage." 269 Kan. 371, Syl. ¶ 4. Thus, only one policy limit of coverage applied to plaintiff's loss of chance of survival claim.

In broad outlines, there are similarities between Wilson and the present case. The cause of Wilson's injury, the lesion, was cancer. Wilson did not sue his doctor because Wilson had a malignant lesion. He sued to recover from the doctor for his failure to diagnose and treat the cancer. In the present case, the cause of NIHL claimants' injury, hearing loss, is high-level noise. The district court found that high levels of noise are inherent and unavoidable in the railroad's operation. In this circumstance, Santa Fe is not faulted for the noise but rather for failing to protect its employees from the noise. That failure is the single occurrence which may trigger coverage under the excess policies.

Santa Fe directs the court's attention to a line of Kansas cases that pre-date Wilson, which stand for the principle that coverage issues are controlled by the legal theory of liability rather than the immediate cause of injury. These cases involve exclusions of coverage rather than a determination of the number of occurrences. In Upland Mutual Insurance, Inc. v. Noel, 214 Kan. 145, 519 P.2d 737 (1974), the insurance company sought a declaratory judgment whether there was coverage and a duty to defend under a homeowner's insurance policy issued to Raymond and Viola Noel. Their sons, drivers of separate cars, collided. Both sons and a passenger died. Survivors of the passenger sued the Noels seeking to recover damages on the theory that they negligently entrusted the automobile that one of their sons was driving. The policy contained an exclusion clause for use of automobiles away from the premises. The court concluded that coverage was not excluded by the clause because it did not extend to a cause of action based on negligent entrustment of the automobile to a known reckless driver. In so deciding, the court looked to the underlying theory of liability rather than to the collision, the immediate cause of the death. In Catholic Diocese of Dodge City v. Raymer, 251 Kan. 689, 840 P.2d 456 (1992), the court again looked to the theory of liability rather than to the immediate cause of the injury in determining coverage. Judgment was entered in favor of the Catholic Diocese and against Allan and Brenda Hammeke for property damage intentionally caused by their minor son. The Catholic Diocese filed for an order of garnishment against the Hammekes' homeowners insurance carrier, which disclaimed coverage for excluded intentional property damage. This court concluded that there was coverage for the parents' negligent supervision, which was the theory on which the parents were held liable. In Marquis v. State Farm Fire & Cas. Co., 265 Kan. 317, 961 P.2d 1213 (1998), an employee acting in the scope of his employment and driving a vehicle owned by his employer ran a red light and severely injured Marquis, the driver of another vehicle. The court found that there was coverage under the employer's contractor's policy, which had not expressly excluded Marquis' claims of negligent hiring, retention, or supervision.

Santa Fe asserts that two Illinois trial courts have entered summary judgment in favor of the policyholders, finding one occurrence to aggregate many NIHL claims. The rulings of one of those trial courts was reviewed in part by the Illinois Court of Appeals in Missouri Pacific R.R. v. Int'l Insur. Co. (MoPac), 288 Ill. App.3d 69, 679 N.E.2d 801, app. denied 174 Ill. 2d 567, 686 N.E.2d 1164 (1997). The case reached the appellate court through the trial court's certification of two questions, which assume rather than put at issue the trial court's determination that the NIHL claims arose out of a single occurrence. The Illinois trial court concluded as a matter of law that NIHL claims arose from a single occurrence and that Missouri Pacific must satisfy only one self-insured retention for each occurrence. The trial court also concluded that the self-insured retentions were not the equivalent of primary insurance or "other insurance" so that horizontal exhaustion principles did not apply. 288 Ill. App. 3d at 74. The questions certified by the trial court were (1) whether the "all sums" rule or the pro-rata by time-on-the-risk theory governs allocation of coverage "where the Circuit Court has found that noise-induced hearing loss claims constitute a single occurrence" and (2) whether the insured must exhaust all self-insured retention amounts for each period of insurance coverage before looking to the insurer for coverage where the NIHL claims span several coverage periods with different self-insured retention amounts. 288 Ill. App. 3d at 74-75. In addressing the first question, the Court of Appeals stated that it was accepting the trial court's finding that the NIHL claims constitute a single occurrence. 288 Ill. App. 3d at 75.

We conclude that support for the district court's single-occurrence decision is found in the policy language and can be inferred from the case law of either Kansas or Illinois. Although the Illinois Court of Appeals in MoPac merely accepted the district court's single-occurrence decision for purposes of the interlocutory appeal, it is persuasive, in part, because the outcome of the first certified question depended on it.

We next consider the trial court's entry of summary judgment on the trigger of coverage.

We note that appellants Fireman's Fund Insurance Company and National Surety Corporation do not join in this argument.

Insurance policies do not refer to a trigger or a trigger of coverage. Those terms are labels for the event or events that under the terms of the insurance policy determines whether a policy must respond to a claim in a given set of circumstances. Owens-Illinois, Inc. v. United Ins. Co., 138 N.J. 437, 650 A.2d 974 (1994). A trigger of coverage enters the discussion when the policyholder's claim implicates more than one policy period, and particularly when the claim implicates policies purchased over a number of years, as in the present case. Five theories for the trigger of coverage in cases involving slowly evolving injuries are described in Owens-Illinois. They are (1) the exposure theory, (2) the manifestation theory, (3) the continuous-trigger theory, (4) injury-in-fact approach, and (5) the double-trigger theory. 138 N.J. at 449-51. (1) The exposure theory places the occurrence at the time the injury-producing agent first contacts the claimant's body. (2) The manifestation theory holds that there is no occurrence until the injury resulting from exposure manifests itself. (3) The continuous trigger theory includes the continuous period from first exposure to manifestation of injury in the occurrence. (4) The injury-in-fact approach holds that coverage is triggered by an inchoate injury which may be inferred by calculating backward from discovery of the injury to the time when that harm actually began. (5) "[T]he `double-trigger' theory holds that injury occurs at the time of exposure and the time of manifestation, but not necessarily during the intervening period. [Citation omitted.]" 138 N.J. at 451.

In the present case, the district court, citing Keene Corp. v. Ins. Co. of North America, 667 F.2d 1034 (D.C. Cir. 1981), concluded that the coverage under Santa Fe's policies was continuously triggered over the years in question, 1945-86. In Keene, the court noted that the first step in analyzing the insurers' duty to indemnify the insured is to determine what event triggers the coverage under the policies. With asbestos, as in Keene, inhalation of asbestos and disease manifestation may be separated by many years, with no outward symptoms during the interim while the disease process develops internally. Thus, the injury continuum stretched over a number of years from asbestos inhalation through asbestos residing in and altering bodily tissue to disease manifestation. Once asbestos fibers have been inhaled, the injury may progress to manifest disease in the absence of continued exposure to asbestos. NIHL is like injury due to asbestos exposure in that it develops gradually over a period of time. It differs from asbestos injury in that it does not progress in the absence of excessive noise.

"In the language of the policies, the question is when did `injury' occur?" 667 F.2d at 1042. The court in Keene concluded, "that inhalation exposure, exposure in residence, and manifestation all trigger coverage under the policies. We interpret `bodily injury' to mean any part of the single injurious process that asbestos-related diseases entail." 667 F.2d at 1047. Here, the district court, Santa Fe, and the Insurers agreed that the policies were triggered if there was exposure, hence bodily injury, occurring during the policy period. Insurers argued that Santa Fe was obligated to show exposure for each NIHL claimant in each policy period. The district court rejected Insurers' argument:

"In this case, Santa Fe is not trying to prove that the NIHL claimants were actually exposed to excessive noise. Instead, Santa Fe is only seeking to show that the claimants had believable evidence that they could present if the underlying cases had been tried before a court of law. As a result of this believable evidence, Santa Fe settled with the underlying claimants.

"It is not now appropriate to relitigate each underlying claim to determine the extent of injury. Santa Fe had knowledge of all pertinent evidence when it negotiated the underlying settlements (dates of employment, potential NIHL injuries prior to employment with Santa Fe, alleged hearing loss due to age, etc.). One has to assume that Santa Fe used this evidence to negotiate favorable settlements. Therefore, it is inappropriate to relitigate the claims and second guess the settlements made by Santa Fe.

". . . [C]ourts applying Kansas law have held that when liability issues have been tried or settled, coverage issues are `controlled' by the settlement in the underlying litigation. The general rule is that `the duty to indemnify is determined by the facts as they are established at trial or as they are finally determined by some other means.' United Wats, Inc. v. Cincinnati Insurers. Co., 971 F. Supp. 1375, 1385 (D. Kan. 1997) [applying Kansas law].

"After a settlement, the underlying claimants' liability contentions are accepted as true for purposes of determining whether there is coverage. See Rozenfeld v. Medical Protective Co., 73 F.3d 154, 158 (7th Cir. 1996) [applying Illinois law]."

The unprotected employees have been subjected to excessive noise levels for a continuous period of time. It is part of the single injurious process which resulted in hearing impairment. Santa Fe's failure to protect the claimants is the occurrence which gives rise to the injury. It may be noted that in this case, the district court relied on counsel's summations of the underlying claimants' liability contentions. Insurers' objections to the district court's use of the attorney affidavits were overruled. Insurers reiterate their objection in this court. As previously noted, Insurers' arguments are directed to the application of the law to the findings of fact. We find no merit to their objections.

On appeal, Insurers contend that the district court improperly determined that the history of each NIHL claimant's exposure to high levels of noise in the workplace was not material to the issue of trigger of coverage. The only cases cited by Insurers are Cloud v. Trinity Companies, 5 Kan. App. 2d 437, 617 P.2d 1277 (1980), and Utah Farm Bur. Ins. Co. v. Dairyland Ins., 634 F.2d 1326 (10th Cir. 1980) (applying Utah law). The questions in both cases were whether there was coverage under a policy and who bore the burden of proof. Cloud involved coverage under a named perils policy that excluded coverage for damage to the interior of the structure caused by rain, snow, sand or dust, unless the elements entered through a roof damaged by direct force of wind or hail. The policyholder unsuccessfully sought to recover the cost of repairs. The Court of Appeals concluded that the policyholder "failed to meet her initial burden of proving a prima facie loss within the coverage of the policy, and it would be mere conjecture to say the opening in the roof was caused by wind." 5 Kan. App. 2d at 439. The principal issue in the Utah case was which party had the burden of proof as to whether there was coverage under the omnibus clause of the automobile insurance policy. The general rule is that the insured or someone claiming through the insured has the burden. Observing that Utah courts had not ruled on the issue and that there was a split of authority elsewhere, the federal court predicted that Utah courts would not shift the burden of proof in a declaratory judgment action brought by the carrier. 634 F.2d at 1328. Neither case addresses the district court's determination that Kansas and Illinois courts accept the underlying claimants' liability contentions as true for purposes of determining whether there is coverage after a settlement.

Insurers complain of the following findings of fact of the district court:

"2. The underlying claimaints and their attorneys alleged that they sustained NIHL from the continuous exposure to excessive noise during the course of their employment.

. . . .

"9. The NIHL claimants alleged that their work environment involved a network of multiple excessive noise sources that they were exposed to on a daily basis in different ways to noises from a variety of sources. The claimants, whether engineers, carmen, or maintenance-of-way workers, were mobile employees, changing locations and job assignments in the course of their work."

In addition, Insurers complain of the implied finding that Santa Fe's implementation of a hearing conservation program is the end of the continuous trigger and complain that the district court ignored Insurers' evidence that some of the NIHL claimants sustained hearing loss before they were employed by Santa Fe. Insurers' point with each of these complaints is that the proof must be individualized.

Their insistence on individualized proof for each of the thousands of NIHL claimants sets the stage for their ultimate argument, which is that the district court's memorandum decision and order on trigger of coverage is in error to the extent that it suggests that Santa Fe may aggregate all claimants' claims together so as to trigger each policy jointly and severally. As the district court noted, Insurers are making an allocation argument, and they are advocating pro rata allocation among insurers and policies by time-on-the-risk. In a separate memorandum decision and order, the district court rejected that argument and concluded that the appropriate method of allocation in this case was joint and several liability.

A determination that the trigger of coverage was continuous is consistent with a determination that the policies from the triggered policy periods bear joint and several liability for indemnifying Santa Fe for its payments to NIHL claimants. Indeed, a determination that the trigger of coverage was continuous is necessary to joint and several allocation. If it is shown that injury producing exposure occurred during the term of a policy, coverage would be triggered so that the policy would be called upon to respond. That same showing would have to be made for allocation other than joint and several liability, in order to implicate policy coverage. We conclude the trial court correctly concluded that coverage under the policies was continuously triggered during the years in question. However, we disagree with and reverse the trial court's conclusion that Santa Fe must exhaust only one self-insured retention before looking to the insurers for indemnification and that insurers are jointly and severally liable.

Self-insured retentions. The district court decided that self-insured retentions are not insurance and that self-insured retentions are deductibles. Applying these decisions to provisions in Insurers' policies, the district court concluded that Santa Fe must exhaust only one self-insured retention before looking to the excess insurers for coverage. One of the provisions is the "other insurance" provision, which pre-1974 typically states: "This insurance does not cover any loss or damage which at the time of the happening of such loss or damage is insured by any other existing policy or policies of valid and collectible insurance except in respect of any excess beyond the amount which would have been payable under such other existing policy or policies had this insurance not been effected." Post-1974 policies typically have the following "other insurance" clause:

"Notwithstanding anything to the contrary contained herein, it is further understood and agreed that where there is any other valid and collectible insurance providing coverages, as described in Section I-A and I-B hereof, this insurance shall be considered as excess insurance over and above such other insurance in effect at the time of the loss or damage except that permission is hereby granted for excess insurance over the limits of liability expressed in this Policy without prejudice to this insurance, and the existence of such insurance, if any, shall not reduce any liability under this Policy."

Another provision appears in both pre- and post-1974 policies:

"Underwriters agree that in the event of a claim for damages for bodily injury or property damage being made by any Assureds hereunder against any other Assureds hereunder, this Policy shall cover such other Assureds against whom claim is made in the same manner as though separate policies shall have been issued to each Assured, but in no event as shall there be imposed more than one deductible set forth in Condition 4 in respect to one occurrence." (Emphasis added.)

"Condition 4" sets the amount of Santa Fe's self-insured retention for the policy period.

Self-insured retentions as "other insurance." The district court concluded that the policy language is unambiguous and hence is to be given its plain meaning. Without reference to authority, the district court concluded that Santa Fe's self-insured retention is the absence of insurance rather than "other insurance" within the meaning of the policy language. The district court further concluded that the self-insured retention is like a deductible in that it "is merely a threshold to be met before an insurer must provide indemnification."

On appeal, Insurers rely primarily on MoPac and the cases cited therein. The second question certified to the Illinois appellate court was "whether Missouri Pacific must exhaust all its [self-insured retentions] SIRs for each period of insurance coverage before looking to the insurers for coverage, where the NIHL claims . . . span several different periods of coverage with different SIR amounts." 288 Ill. App. 3d at 80. The Illinois trial court had agreed with Missouri Pacific that "SIRs operate as deductibles and are not `other insurance' within the meaning of the policies." 288 Ill. App. 3d at 83. The Illinois Court of Appeals disagreed:

"In United States Gypsum, the court held that horizontal exhaustion was required because of the `other insurance' provision contained in the excess policies. See United States Gypsum [Co. v. Admiral Insurance Co.], 268 Ill. App. 3d [598, 654, 643 N.E. 2d 1226 (1994)]. The court noted that the provision clearly set forth the policy's status as an excess policy `to all triggered primary policies, regardless of whether they extend over multiple policy periods or only one.' . . . 268 Ill. App. 3d at 653. The court then concluded that liability under an excess policy would not attach until all underlying coverage was exhausted.. . . 268 Ill. App. 3d at 653-54. This underlying coverage included Gypsum's fronting insurance, which the court noted was the equivalent of self-insurance.. . . 268 Ill. App. 3d at 652-53.

"In light of United States Gypsum, we conclude that Missouri Pacific is required to exhaust one SIR per occurrence per policy period. For all practical purposes, Missouri Pacific's SIRs are the equivalent of the fronting insurance involved in Gypsum. As such, the `other insurance' provision in the policies requires Missouri Pacific to exhaust all underlying coverage, including its SIRs, before it can seek coverage under the policies. See, . . . 268 Ill. App. 3d at 652-54." 288 Ill. App. 3d at 83-84.

Insurers also cite Associated Wholesale Grocers, Inc. v. Americold Corp., 261 Kan. 806, 934 P.2d 65 (1997), for its discussion of the relative positions of primary and excess insurers, which includes this quote from Union Indem. Ins. Co. v. Certain Underwriters, 614 F.Supp. 1015, 1017 (S. D. Tex. 1985):

"`Primary insurance coverage is insurance coverage whereby, under the terms of the policy, liability attaches immediately upon the happening of the occurrence that gives rise to liability. [Citation omitted.] An excess policy is one that provides that the insurer is liable for the excess above and beyond that which may be collected on primary insurance. [Citation omitted.] In a situation in which there are primary and excess insurance coverages, the limits of the primary insurance must be exhausted before the primary carrier has a right to require the excess carrier to contribute to a settlement. [Citations omitted.] In such a situation, the various insurance companies are not covering the same risk; rather, they are covering separate and clearly defined layers of risk. The remote position of an excess insurer thus greatly reduces its chance of exposure to a loss. This reduced risk is generally reflected in the cost of the excess policy.'" 261 Kan. at 829.

Santa Fe would distinguish MoPac on the ground that the policies at issue in the Illinois case contained language not found in the policies issued to Santa Fe. According to Santa Fe, Missouri Pacific's policies expressly provided for retentions to be treated as primary insurance. Noting that MoPac relied on two other Illinois cases, the district court rejected the proposition that the policy language should distinguish MoPac from the present case. Instead, the district court discounted MoPac on the ground that there is a split of authority in Illinois and it found the other authority more persuasive. The district court based its view that there is an Illinois split on an unreported trial court decision.

The district court's consideration of the question whether Santa Fe's self-insured retentions constitute insurance for legal purposes was rather narrowly confined to a definition of insurance as a contract whereby one party promises for a consideration to indemnify the other against certain risks. See Londerholm v. Anderson, 195 Kan. 649, 662, 408 P.2d 864 (1965). Because there is no contract or second party involved in Santa Fe's self-insurance scheme, the district court concluded that the self-insured retentions are neither insurance nor "other insurance." The district court's decisionmaking was bolstered by counsel for one of the Insurers "conced[ing] at the Oral Argument on the Summary Judgment Motions that self-insurance is not insurance." Judging from the arguments of Insurers in this case, any such concession amounted only to an agreement that Santa Fe's self-insured retentions do not fit within the Londerholm definition.

Santa Fe would have the court dismiss Americold as immaterial because Santa Fe's self-insured retentions are not insurance. For that proposition Santa Fe cites Stratford Sch. Dist. v. Employers Reins. Corp., 162 F.3d 718 (1st Cir. 1998), in which the analysis is restricted to the definition of insurance as an agreement involving more than one party. Santa Fe also cites Beech Aircraft Corp. v. United States, 797 F.2d 920, 922 (10th Cir. 1986), and USX Corp. v. Liberty Mutual Insurance Co., 269 Ill. App. 3d 233, 645 N.E. 2d 396 (1994). In Beech, the subject was tax deductions. The court held that "funds set aside as reserves against contingent losses, as a plan of `self-insurance,' where there is no real transfer of risk to a separate entity, are not deemed to be insurance premiums and are therefore not deductible as an ordinary business expense." (Emphasis added.) 797 F.2d at 922. Reliance on USX by the Illinois trial court on the issue of self-insured retentions was disapproved by the Illinois Court of Appeals in MoPac. 288 Ill. App. 3d 69, 80, 679 N.E.2d 801 (1997).

Examination of Self-Insurance against Liability as Other Insurance within Meaning of Liability Insurance Policy, 46 A.L.R.4th 707, shows that courts have applied many different analyses. Several courts relied strictly on a definition of insurance, as Santa Fe suggested in citing Londerholm. Other courts applied fact-based analyses on the particular forms of self-insurance at issue. And some courts make public policy and fairness the touchstones for decision. In Hillegas v. Landwehr, 176 Wis. 2d 76, 83-84, 499 N.W.2d 652 (1993), for example, the court concluded that self-insurance was "other insurance" on the ground that self-insurers gain the dual benefit of avoiding premiums and of avoiding primary liability if they can force another insurance company to pay on the ground that self-insurance is not "other insurance." Kansas courts are not recorded as having considered the issue. A Tenth Circuit Court of Appeals case predicting Oklahoma law included in the annotation is Air Liquide America Corp. v. Continental Cas. Co., 217 F.3d 1272 (10th Cir. 2000), in which an approach very like that of the Wisconsin court in Hillegas was taken. The federal Court of Appeals held that under predicted Oklahoma law, a truck owner's "fronting" policy that was a form of self-insurance constituted "other collectible insurance," so that agency's policy provided only excess coverage. Here is the court's discussion of the issue:

"Although Oklahoma apparently has not decided whether self-insurance is `other collectible insurance,' we believe our decision in Industrial Indemnity Co. v. Continental Casualty Co., 375 F.2d 183 (10th Cir. 1967), is controlling on this issue. Industrial Indemnity concerned an incident arising out of two subcontractors' work on an Oklahoma oil lease. The first subcontractor, Halliburton Oil Well Cementing Company (`Halliburton'), was insured by Continental Casualty Company (`Continental Casualty'); the second, Johnson Construction Company (`Johnson'), was insured by Industrial Indemnity Company (`Industrial'). See Id. at 184. While en route to the job site, a Halliburton employee was injured when attempting to disconnect a Halliburton truck from a Johnson tractor that had towed the truck through the mud. See Id. The Halliburton employee sued Johnson, and Johnson claimed that, in addition to being an insured under its policy with Industrial, it was an insured under Halliburton's policy with Continental Casualty. The court agreed that Johnson was also an insured under the Continental Casualty policy, and then addressed the question of how to allocate the loss between the two policies.

"It is apparent from the court's reasoning that both policies contained substantially identical `other insurance' clauses purporting to render that policy excess if there was other insurance. [FN6] See Id. at 185. It is further apparent that Halliburton's policy with Continental Casualty provided that Halliburton would reimburse Continental Casualty for any losses. See Id. Continental Casualty argued to this court that `Halliburton was a self-insuror by reason of its agreement to reimburse the insurance carrier for the losses and hence there was no "other insurance."' Id. We rejected this argument and affirmed the district court's proration of the loss between the policies, stating that `[t]here is no authority in Oklahoma which would dictate that a distinction should be made between the nature of the coverage.' Id. at 186. Thus, we held that the fact that one policy was, in effect, self-insurance did not diminish the applicability of an `other insurance' clause.

"FN6. We note that the terms `other insurance' and `other collectible insurance' are, for present purposes, substantially identical. See Self-Insurance Against Liability as Other Insurance Within Meaning of Liability Insurance Policy, 46 A.L.R.4th 707, 709 (1986) (noting that `[c]ourts do not generally distinguish among these phrases.').

"We follow the reasoning of Industrial Indemnity in the present case and conclude that Air Liquide's self-insurance is `other collectible insurance' within the meaning of the Continental policy. We have found no authority in Oklahoma contradicting our decision in Industrial Indemnity, and we remain of the opinion that it reaches the correct result. Air Liquide's decision to self-insure does not relieve it from primary liability simply because the underlying accident was also covered by another insurance policy. Were we to hold otherwise, Air Liquide would receive the double windfall of avoiding significant premium payments under a standard insurance policy and avoiding primary liability for an accident caused by one of its vehicles. We predict that Oklahoma would not reach such an inequitable result.

"We also note that at least two of the cases from other jurisdictions relied upon by the district court are plainly distinguishable from the present case. See St. John's Reg'l Health Ctr. v. American Cas. Co., 980 F.2d 1222, 1225 (8th Cir. 1992) (addressing a pooled liability fund); Physicians Ins. Co. of Ohio v. Grandview Hosp. & Med. Ctr., 44 Ohio App.3d 157, 542 N.E.2d 706 (1988) (addressing a hospital's contractual obligation to indemnify an employee). Neither of these cases involved a fronting policy like the CIGNA policy in the present case, in which a company that is in the business of insuring against loss acts as a surety for the self-insured's ability to satisfy a judgment. We also note that other jurisdictions have followed the approach of Industrial Indemnity and the case at bar. See Self-Insurance Against Liability as Other Insurance Within Meaning of Liability Insurance Policy, 46 A.L.R.4th 707 §§ 3[a], 4[a] (collecting cases); 1 Couch on Insurance 3d § 10:6 (same). In any event, the cases relied upon by the district court are from other jurisdictions and, hence, are not controlling, whereas Industrial Indemnity represents controlling Tenth Circuit authority interpreting Oklahoma law.

"Thus, we conclude that Air Liquide's CIGNA policy is `other collectible insurance' within the meaning of the Continental policy, notwithstanding Air Liquide's obligation to fully reimburse CIGNA for any losses. [FN8] As a result, the terms of the Continental policy make it excess to the CIGNA policy. [Footnote omitted.] Thus, we reverse the district court's holding that the Continental policy was primary.

"FN8. Cf. 15 Couch on Insurance 3d S 217:9 (`Any applicable deductible is relevant between the insurer and the insured only, and does not apply to proration.')" 217 F.3d at 1278-79.

In summary, the trial court in this case decided that Santa Fe's self-insured retentions are not "other insurance" within the policy language because the self-insured retentions do not fall within the two-party definition of insurance. The Illinois Court of Appeals, based on Illinois precedent in which the insured's fronting insurance was treated as the equivalent of self-insurance, decided that Missouri Pacific railroad's self-insured retentions are "other insurance" for legal purposes. Among the courts that have decided whether self-insurance is insurance it appears that a slight majority have decided it is not. Sound policy and fairness reasons have been articulated for deciding that self-insurance is insurance. In addition, excess insurance coverage, as Insurers provided to Santa Fe, generally assumes that there is primary insurance coverage. As noted in MoPac, to hold otherwise allows the insured to "manipulate the source of its recovery and avoid the consequences of its decision to become self-insured, conduct we found unacceptable in United States Gypsum and Outboard Marine. As such, Missouri Pacific must exhaust the SIRs before looking to the insurers for coverage. See United States Gypsum, 268 Ill. App. 3d at 653-54." 288 Ill. App. 3d at 82.

We must give effect to the purpose of indemnification as expressed by the terms of the policy, construed in the context of the policy as a whole. In that regard we find the reasoning in Air Liquide America Corp. and MoPac persuasive. We conclude that the SIRs are "other insurance" within the meaning of the policies in the present case.

Having decided that Santa Fe's self-insured retentions are other insurance, we need not address the district court's further conclusion that the self-insured retentions are the equivalent of deductibles.

However, the significance of the trial court's equating self-insured retention with deductible is that the insurance policies expressly limit imposition of a deductible to one per occurrence. Because there is but one occurrence, according to the trial court, Santa Fe need pay only one self-insured retention before Insurers become obligated to indemnify the remaining amount paid by Santa Fe to NIHL claimants. Insurers argued that all self-insured retentions for all affected policy must be exhausted before any of their excess coverage is tapped. The district court rejected the argument on the ground that this concept of horizontal exhaustion could not be found in the language of the policies.

As we have seen, the Illinois Court of Appeals held that Missouri Pacific must horizontally exhaust all self-insured retentions for implicated policy periods before looking to the insurers for coverage. MoPac, 288 Ill. App. 3d at 82. The Illinois Court of Appeals did not consider whether a self-insured retention is a deductible because it held that a self-insured retention is the equivalent of underlying insurance coverage. 288 Ill. App. 3d at 80-84. The Illinois court reasoned that exhaustion of the self-insured retentions was required by the "other insurance" policy provisions. We agree.

Allocation. If Santa Fe must exhaust only one self-insured retention before looking to Insurers for coverage, the claims must be allocated among the Insurers. The district court held that the Insurers were jointly and severally liable for the claims. It reasoned that the "all sums" language of the policies requires each policy to indemnify Santa Fe for all amounts it has paid to NIHL claimants above its self-insured retention.

The "all sums" language appears in the liability clauses and provides that the insurer will indemnify the policy holder "for any and all sums which the Assured" shall become liable to pay arising out of an accident or occurrence. In its memorandum decision and order, the district court expressly adopted the holdings in two asbestos cases, J. H. France Refractories v. Allstate, 534 Pa. 29, 626 A.2d 502 (1993), and Zurich Insurance v. Raymark Industries, 118 Ill. 2d 23, 514 N.E.2d 150 (1987), with regard to the "all sums" language. In adopting J. H. France and Zurich Ins., the district court ignored MoPac, the more recent case and one involving NIHL claims.

In MoPac, the Illinois Court of Appeals examined the "all sums" language of the policies, which is nearly identical to that in Santa Fe's policies and warrants the same construction. 288 Ill. App. 3d at 72. The first issue addressed by the Illinois Court of Appeals was whether the "all sums" allocation rule in Zurich Ins. governed the allocation of coverage for the NIHL claims given that the NIHL and asbestos claims each constitute a single occurrence. 288 Ill. App. 3d at 75. The Court of Appeals reasoned that the policies plainly provided that "the sums the insurers are obligated to pay must be on account of personal injuries that occur during the policy period. 288 Ill. App. 3d at 77. Rejecting the Zurich Ins. rule, the MoPac court turned to Outboard Marine v. Liberty Mutual Ins., 283 Ill. App. 3d 630, 642, 670 N.E.2d 740 (1996), which held that "the sums the insurer is obligated to pay must be on account of property damage arising out of an occurrence during the policy period."

The MoPac court further stated:

"Given that the policies provide that the sums the insurers are obligated to pay must be on account of personal injuries that occur during the policy period, the question becomes whether the NIHL damages can be allocated to particular policy periods. At this stage of the litigation, we believe it is premature for us to decide whether they can or cannot. We agree with the insurers that a genuine issue of material fact exists as to whether Missouri Pacific's NIHL damages can be measured and allocated to specific time periods within a reasonable, degree of scientific certainty. For instance, in its memorandum in opposition to Missouri Pacific's motion for summary judgment, International [Insurance Company] submitted the affidavit of Dr. William Clark, who stated that NIHL occurs only during active exposure to excessive noise in the workplace. In the same memorandum, International attached the expert witness disclosure statement of Dr. Robert Dobie, who stated that NIHL injuries can be measured to specific periods of time. Also, in their motion for clarification and, in the alternative, for reconsideration of the trial court's ruling granting Missouri Pacific's motion for summary judgment, the insurers repeated that NIHL damages could be apportioned among policy periods, and they directed the court's attention to the expert witnesses who would testify as such. In that same motion, the insurers also attached the American National Standard Institute's model upon which the allocation of NIHL is based. Collectively, this evidence creates a genuine issue of material fact as to whether NIHL damages can be measured and allocated to specific periods of time.

"Parenthetically, we note and dismiss as meritless Missouri Pacific's argument that evidence concerning the allocation of NIHL damages to particular policy periods was never presented to the trial court. Our review of the record, as explained in the preceding paragraph, reveals that this evidence was before the trial court when it made its rulings and is currently part of the record on appeal.

"We therefore hold that the trial court erred in granting Missouri Pacific summary judgment on the issue of allocation, and we remand the matter for proceedings consistent with the decision we reach today. On remand, the parties should be given the opportunity to present evidence that NIHL may (or may not) be measured and allocated, within a reasonable degree of medical and scientific certainty, to particular policy periods.

"In the event that NIHL cannot be measured and allocated to particular policy periods, the trial court should apply the pro rata, time-on-the-risk allocation of damages approach used in Outboard Marine. See Outboard Marine, 283 Ill. App. 3d at 645. In Outboard Marine, the insured maintained excess insurance policies which contained language nearly identical to that in the present case. The trial court found that a single continuous occurrence—the insured's contamination of a lake over a 23-year period—resulted in an unallocable loss which implicated successive policy periods. We held that in such a situation the best method of damage allocation is a pro rata, time-on-the-risk allocation. See Outboard Marine, 283 Ill. App. 3d at 642-45.

"Similarly, this case involves a single continuous occurrence which may result in an unallocable loss implicating successive policy periods. If it is determined on remand that the NIHL damages are unallocable, then the best method of damage allocation, as in Outboard Marine, is a pro rata, time-on-the-risk allocation. Outboard Marine, 283 Ill. App. 3d at 643-45.

"Moreover, we note that Missouri Pacific's reliance on Zurich, 118 Ill.2d 23, 112 Ill. Dec. 684, 514 N.E.2d 150, is misplaced. In Zurich, the court applied a "triple trigger" theory to determine which primary insurance policies were implicated by asbestos-related personal injuries. The court held that such policies were triggered at the initial exposure, when the disease manifested itself, and at any interim time when the claimant manifested some sickness. Zurich, 118 Ill.2d at 44; see Outboard Marine, 283 Ill. App. 3d at 641. The insurer then urged the court to adopt the pro rata approach in Insurance Co. of North America v. Forty-Eight Insulations, Inc., 633 F.2d 1212 (6th Cir. 1980), aff'd on reh'g 657 F.2d 814 (6th Cir. 1981), to allocate the indemnity and defense costs of each claim among the triggered policies. In Forty-Eight Insulations,

`the court held that the insurers' obligations under their respective policies were triggered only by a claimant's exposure to asbestos during a policy period. From this premise, that court concluded that the exposure theory provided a reasonable means of allocating the costs of defense and indemnification among the triggered policies based on the number of years of exposure.' Zurich, 118 Ill.2d at 57.

"However, the court in Zurich had already held that a policy was triggered if the claimant suffered bodily injury or sickness or disease. Accordingly, having rejected the premise underlying the pro rata allocation approach in Forty-Eight Insulations, Zurich concluded that the appellate court did not err `insofar as it declined to order a pro rata allocation of defense and indemnity obligations among the triggered policies.' Zurich, 118 Ill.2d at 57.

"We read nothing in Zurich as precluding the application of the pro rata, time-on-the-risk allocation method announced in Outboard Marine to the case at bar. Zurich's rejection of a pro rata approach was based on the narrow facts of the case before it, namely, that it had rejected the trigger analysis set forth in Forty-Eight Insulations. The court did not adopt a general rule forever precluding the application of a pro rata approach. As we explained in Outboard Marine, such an approach is appropriate where, as here, a single continuous occurrence results in an unallocable loss implicating successive policy periods." 288 Ill. App. 3d at 77-78.

On appeal, Insurers cite MoPac and Outboard Marine. In addition, they contend that Kansas law requires rejection of joint and several liability just as the Illinois cases do. Insurers cite Scott v. Keever, 212 Kan. 719, 512 P.2d 346 (1973), where the question concerned an injury arising outside the policy period with respect to products liability coverage. The court found no coverage under a policy in force at the time Scott purchased a ladder that proved to be defective and caused an injury after the policy expired. Scott does not seem capable of extension to govern this allocation question.

Santa Fe seems to assert that the "all sums" language in Insurers' policies is different from that in the MoPac policies in lacking limitation to the policy period. Insurers refute the assertion by quoting a "typical" provision that limits the insurer's obligation to all sums for injuries arising out of occurrences caused by Santa Fe's operations during the policy period.

Amicus curiae, the Complex Insurance Claims Litigation Association (CICLA), contends that the district court's joint and several liability scheme is contrary to the policy language, to the majority of authorities and certainly to the better reasoned authorities, and to fundamental principles of fairness.

Santa Fe responds by claiming that it has cited more cases for its position that CICLA did for its position and then reiterates its reliance on Zurich Ins., which was distinguished in MoPac, as we have seen.

We have concluded that the SIRs are other insurance under the policies, and thus primary insurance. We further recognize that primary coverage attaches upon the happening of an occurrence which we have determined to be the failure of Santa Fe to provide protection for its employees. An excess policy covers the loss over and above that provided by the primary insurance. We agree with the rationale expressed in MoPac that the insured must exhaust its SIRs per annual policy period. We further agree that the concept of joint and several liability is not consistent with the term "all sums" in the policies. It also clearly contradicts the fundamental insurance agreement to indemnify the insured for injuries during a specified policy period. We cannot ignore the stated terms of the policies, nor the reality of SIRs as primary insurance where the expectation and intent is to provide excess coverage. We cannot determine with certainty if the SIRs are sufficient to cover the damages for each year in question. Thus, this case must be remanded for the trial court to make that determination, and if the SIRs are not sufficient, to allocate the damages attributable to the excess coverage for that annual policy period.

We next consider whether the district court erred in concluding as a matter of law that neither lack of fortuity nor the known loss doctrine precludes coverage.

As similar as the two parts of this issue seem, they generally are treated as different defenses to coverage—lack of fortuity being based on the term, "accident," in a liability policy and known loss being based not on policy language but on a principle that losses existing at the beginning of a policy period are not proper subjects for insurance. "The two defenses frequently are asserted in tandem because they give the insurer two chances to prevail on essentially the same facts and rationale." Fruehauf, The Cost of Knowledge: Making Sense of "Nonfortuity" Defenses in Environmental Liability Insurance Coverage Disputes, 84 Va. L. Rev. 107, 131 (1998).

Nonfortuity. The district court concluded that Santa Fe's NIHL loss was unexpected and unintended. According to the district court, the conclusion was based on uncontroverted evidence that there were only 11 NIHL claims during the years 1966 through 1984-85, that Santa Fe's noise levels satisfied OSHA standards, that Santa Fe's own noise level testing in the 1970's showed that its workers were "neither being exposed to excessive noise levels nor experiencing hearing loss greater than in the general population."

Insurers set out facts they contend require the conclusion that Santa Fe's losses were not fortuitous. They direct the court's attention to a 1930 Santa Fe publication, Santa Fe's requiring its train workers by 1954 to undergo hearing examinations, and to a 1955 communication to Santa Fe's Medical Director. In the district court's view, the early indications of Santa Fe's concern about hearing loss among its employees was more than offset by Santa Fe's receiving only a single NIHL claim up through 1971. In addition, the district court distinguished between Santa Fe's awareness of NIHL and its awareness of traumatic hearing loss injuries. Between 1965 and 1971, 10 Santa Fe employees made claims for hearing loss, but only one was for NIHL.

Insurers cite Kansas cases involving intentional acts to support their contention that the legal standard in this state is whether Santa Fe's NIHL losses were sudden and unexpected. Insurers argue that there is undisputed evidence Santa Fe knew well before 1966 that railroad noise could be detrimental to employees' hearing and Santa Fe's Medical Director was advised in 1955 of a "very definite trend to consider hearing loss as a possible source of claims." The photocopy of the document is only partly legible. It seems to say that "the total number of patients is very small" but "there is a very definite trend to consider hearing loss as a possible source of claims. This applies to the rail-road as well as other industries, certainly."

There is a difference between the state of knowledge Insurers impute to Santa Fe and a state of knowledge that would render the NIHL claims uninsured. Insurers state that Santa Fe was aware that hearing loss can be induced by railroad noise and that Santa Fe had been advised of a trend to consider hearing loss as a possible source of claims. The facts raised by Insurers do not show that Santa Fe's losses were expected. At best, Insurers' facts show that Santa Fe was aware of a possibility NIHL claims might be made against it. In other words, Santa Fe was aware of a potential risk. A basic tenet of liability insurance is that a policyholder purchases insurance because there is a possibility, or even probability, that it will suffer losses due to claims made against it for known potential risks.

Known loss. The district court found that Santa Fe and the Insurer knew or should have known of the risk of NIHL before 1985. Knowledge of risks, however, was distinguished by the district court from knowledge of losses, in particular losses that would exceed its self-insured retention so as to be covered by any of the liability policies. Moreover, in the district court's view, because Santa Fe's knowledge was shared by Insurers, they could have, but did not, protect themselves by expressly excluding the known risk from coverage.

The facts Insurers call to the court's attention as allegedly showing that Santa Fe knew it would be faced with hearing loss claims are Santa Fe's knowledge of hearing loss as early as the 1950's and its settlement of hearing loss claims between 1965 and 1971. Like the facts cited by Insurers relative to nonfortuity, these facts do not distinguish between traumatic hearing loss and NIHL. The hearing loss claims before 1971, for example, involved torpedo, air hose, and compressor explosions, and horns sounding too close to workers.

The district court quoted the standard for a known loss from Outboard Marine Corp. v. Liberty Mut. Ins., 154 Ill. 2d 90, 103-04, 607 N.E.2d 1204 (1992), and that standard is confirmed in the 1996 Outboard Marine decision:

"If the insured knows or has reason to know, when it purchases a [comprehensive general liability] policy, that there is a substantial probability that it will suffer or has already suffered a loss, the risk ceases to be contingent and becomes a probable or known loss. [Citation omitted.] Where the insured has evidence of a probable loss when it purchases a [comprehensive general liability] policy, the loss is uninsurable under that policy (unless the parties otherwise contract) because the `risk of liability is no longer unknown.' [Citations omitted.]" 283 Ill. App. 3d at 653.

On appeal, Insurers contend that a known loss standard was formulated by this court in Matlock v. Hollis, 153 Kan. 227, 233, 109 P.2d 119 (1941). Matlock is a case of fraudulent concealment rather than a known loss case, as that doctrine is described in Outboard Marine. In Matlock, the insured, undisclosed to the insurer, purchased an antedated policy to cover a period in which its employee had been injured. 153 Kan. at 228. Matlock is not applicable in the present case, and no other Kansas case has been suggested as recognizing a known loss doctrine.

The Outboard Marine test is whether, when it purchases liability insurance, the insured knows or has reason to know that there is a substantial probability it will suffer or has already suffered a loss. Outboard Marine used a hydraulic fluid in its outboard motor manufacturing process. In 1970, with notification from the hydraulic fluid manufacturer, Outboard Marine learned that the hydraulic fluid contained PCBs. Outboard Marine used up the hydraulic fluid it had in stock and did not finally rid its machines of residual amounts of PCB-laden fluid until approximately 1976. In February 1976, EPA notified Outboard Marine that its discharge of PCBs into water on its property and eventually into the harbor violated environmental statutes. 283 Ill. App. 3d at 636. An insurer argued that Outboard Marine knew from the fluid manufacturer's letter that its discharge was contaminating the harbor. 283 Ill. App. 3d at 653.

"[The trial court] found [Outboard Marine] did not know that there was a substantial probability that it would suffer a know loss. We find that the trial court's conclusion was not against the manifest weight of the evidence. While it is troublesome that [Outboard Marine] received a letter in 1970 informing it that the [hydraulic fluid] it was using contained PCBs and it continued to exhaust its current stock of PCB-laden [hydraulic fluid] until February 1, 1971, the record indicates that [Outboard Marine] made modifications to its manufacturing system and that it believed that the [hydraulic fluid] would not be discharged into the harbor. Moreover, the record does not clearly establish that prior to the 1976 USEPA complaint [Outboard Marine] had reason to know that it would suffer or had suffered a loss as a result of PCB contamination of the harbor. Accordingly, we find that [Outboard Marine] was not barred from obtaining coverage under the Northbrook policy on account of the `known loss' doctrine." 283 Ill. App. 3d at 653-54.

Outboard Marine's test in the facts of this case would be—did Santa Fe know when it purchased the liability policies that there was a substantial probability that it would suffer a known loss? Believing that the circumstances of the present case differed from those in Outboard Marine, the district court tailored the standard. The district court's inquiry took into account that an insured loss must exceed Santa Fe's substantial self-insured retention. The inquiry was whether Santa Fe knew when it purchased liability policies that there was a substantial probability it would suffer a known loss exceeding its self-insured retention. The district court concluded that Santa Fe did not have such knowledge for the policy years at issue. It appears that the district court considered its standard to be an application of Kansas law. Whether the district court's standard constitutes a deviation from established Illinois law or merely an adaptation to the circumstances is not addressed.

The district court also observed that, because Insurers were aware of NIHL as a potential risk, they were in a position to expressly exclude NIHL coverage. Because Insurers did not exclude coverage, they are not in a position to claim that coverage is precluded by the known loss doctrine. "An insurer assumes a duty to define any limitations on coverage in clear and explicit terms." Americold, 261 Kan. 806, Syl. ¶ 1. The same principle has been approved by an Illinois appellate court. See Mo. Pacific R.R. Co. v. Amer. Home Assur., 286 Ill. App. 3d 305, 675 N.E.2d 1378 (1997), cited by 7 Couch on Insurance 3d § 102:8 (1997), for this statement of the principle: "Given the underlying basis of the doctrine, and the right of the parties to agree to cover existing losses, it has been recognized that the known loss doctrine does not apply if the insurer also knew of the circumstances on which it bases the defense." We find no error in the trial court ruling on nonfortuity or known loss.

We next turn to the question of Santa Fe's notice to insurers.

The district court found that the policies generally required the policyholder to give the insured notice of an occurrence as soon as practicable and that the Insurers first received notice of Santa Fe's NIHL claims in the late 1980's. Eleven NIHL claims had been filed with Santa Fe up to and including 1985. Santa Fe had received notice of several hundred NIHL claims before giving notice to Insurers.

"The phrase `as soon as practicable' has been construed under Kansas law to mean that the insured must notify its insurer within a reasonable period of time in view of all the relevant facts and circumstances of a particular case." Cessna Aircraft Co. v. Hartford Acc. & Indem. Co., 900 F. Supp. 1489, 1515 (D. Kan. 1995), citing Travelers Ins. Co. v. Feld Car & Truck Leasing Corp., 517 F. Supp. 1132, 1134 (D. Kan. 1981). In this case, the district court seems to have concluded that Santa Fe's notice to the Insurers was not given as soon as practicable.

For the rule regarding late notice followed by Kansas courts, the district court quoted from Cessna Aircraft Co.:

"The parties agree that untimely notice, even if a breach of a condition precedent to coverage, is not alone sufficient to excuse performance of the insurer or relieve the insurer of its obligation to provide coverage when coverage otherwise should be afforded. Kansas also requires a showing of actual prejudice as a result of the untimely notice. Home Life Ins. Co. v. Clay, 11 Kan. App. 2d 280, 286-87, 719 P.2d 756 (1986). Prejudice is not presumed and the burden is on the insurer to show that the prejudice is substantial. Hunt v. Kling Motor Co., 841 F. Supp. 1098 (D.Kan.1993); Feld Car & Truck Leasing, 517 F.Supp. at 1135-36; Security Nat'l Bank of Kansas City v. Continental Ins. Co., 586 F.Supp. 139, 150 (D.Kan.1982). "When the loss will result from liability owed a third party, the insured is only prejudiced if its ability to defend against that imposition of liability is diminished by the delay." Clay, 11 Kan. App. 2d at 288, 719 P.2d at 763.[FN34] Generally, whether an insurer has been prejudiced from the failure to provide timely notice is a question of fact, but where the relevant facts are not in dispute it may be determined as a matter of law. F.D.I.C. v. Oldenburg, 34 F.3d 1529, 1547 (10th Cir.1994); Feld Car & Truck Leasing, 517 F.Supp. at 1135; Goff v. Aetna Life & Casualty Co., 1 Kan. App. 2d 171, 178, 563 P.2d 1073, 1079 (1977)." 900 F. Supp. at 1515-16.

The Insurers in the present case, according to the district court, "took no action to negotiate, defend, or prevent any NIHL claims against Santa Fe. Even if they had been given notice earlier many of the Insurers conceded that they would not have acted differently. These excess Insurers had no right nor obligation to control the defense or settlement of any claim. The policies provide for indemnification only." Because Insurers have shown no prejudice, the district court concluded that the timing of Santa Fe's notice to Insurers is not a bar to coverage.

On appeal, Insurers contend that the district court should have looked to Illinois law for guidance on this issue and that under Illinois law, whether they were prejudiced by untimely notice is a factor, but not a dispositive consideration. In cases more recent than those cited by Insurers, the Illinois Appellate Court, Fourth District, has held that an insurer must prove prejudice to avoid coverage on account of a policyholder's delay in reporting a lawsuit. Rice v. AAA Aerostar, Inc., 294 Ill. App. 3d 801, 690 N.E.2d 1067 (1998); Cincinnati Ins. v. Baur's Opera House, 296 Ill. App. 3d 1011, 694 N.E.2d 593 (1998). Writing in the Illinois Bar Journal, Stanley C. Nardoni welcomes the change in Illinois law as a shift to "the modern rule." Nardoni, Illinois Adopts the "Modern" Prejudice Rule for Insurers' Late-Notice Defense, 87 Ill. B.J. 86 (1999). According to the author, the "modern rule is fairer because it avoids forfeiting coverage for harmless delays." 87 Ill. B.J. at 88. He also notes that in Cincinnati Ins., the insurer proved prejudice and avoided "shows that the rule does not establish an impossible burden for insurers." 87 Ill. B.J. at 88. He advocates that

"the trend begun by Rice should be extended to the late-notice-of-occurrence defense. Insurance carriers have sought to avoid coverage where they believe their insureds failed to meet their obligation to notify of occurrences before any lawsuit was filed. States following the modern rule require prejudice for both the notice-of-occurrence and notice-of-suit defenses. The interests of clarity and fairness are served by applying the same prejudice rule for both defenses." 87 Ill. B.J. at 88 and 90.

After Rice and Cincinnati Ins., there is a question whether the "modern rule" should be applied without distinction to notice of occurrence and notice of suit cases. In Rice, the court stated:

"There are differences between the notice of an occurrence and notice that a suit has been filed. The purpose of notice of the occurrence is to enable the insurer to conduct a timely and thorough investigation of the insured's claim, while the purpose of notice of the lawsuit is to enable the insurer to locate and defend the suit. Kerr v. Illinois Central R.R. Co., 283 Ill. App. 3d 574, 585, 670 N.E.2d 759 (1996) (notice of occurrence). In the present case State Farm was able to conduct a timely and thorough investigation of the insured's claim, because it had notice of the occurrence.

"There is some uncertainty in the cases whether prejudice to the insurer must be shown before it is relieved of its duty to indemnify. See General Casualty Co. v. Juhl, 283 Ill. App. 3d 376, 382, 669 N.E.2d 1211, 1215 (1996) (damage is inherent in such a situation); University of Illinois v. Continental Casualty Co., 234 Ill. App. 3d 340, 364, 599 N.E.2d 1338 (1992) (when insured fails to comply with prompt notice requirement, insurer may deny liability regardless of whether it has been prejudiced by delay); Millers Mutual Insurance Ass'n v. Graham Oil Co., 282 Ill. App. 3d 129, 141, 668 N.E.2d 223 (1996) (absence of actual prejudice to insurer is not conclusive, but is a factor to be considered). Those cases, however, are all notice-of-occurrence cases. When notice of the lawsuit is the issue, the rule is that the insurer is required to show that it was prejudiced by the insured's omission or delay in order to escape liability on its policy. See Annot., Modern Status of Rules Requiring Liability Insurer to Show Prejudice to Escape Liability Because of Insured's Failure or Delay in Giving Notice of Accident or Claim, or in Forwarding Suit Papers, 32 A.L.R.4th 141, 145." 294 Ill. App. 3d at 807-08.

The differing approaches for notice-of-occurrence and notice-of-suit cases is reflected in the decisions subsequent to Rice and Cincinnati Ins. Insurers cite Montgomery Ward & Co. v. Home Ins. Co., 324 Ill. App. 3d 441, 753 N.E.2d 999 (2001), for not requiring insurers to show prejudice from delay in notice of occurrence. Montgomery Ward was decided in the First District of the Illinois Appellate Courts. Here is the discussion of the effect of late notice:

"Ward asserts that Imperial must show that it was prejudiced because of the delay in notification in order to avoid indemnifying Ward. In support of its position that the insurer must show prejudice, Ward cites Rice v. AAA Aerostar, Inc., 294 Ill. App. 3d 801, 690 N.E.2d 1067 (1998), and [Cincinnati Ins. v.] Baur's Opera House, 296 Ill. App. 3d 1011, 694 N.E.2d 593 [(1998)]. The cases on which plaintiff relies are distinguishable.

"In both AAA Aerostar and Baur's Opera House, the courts were faced with situations where the insurer was not notified of a lawsuit. In contrast, the case at bar concerns a notice-of-occurrence situation. The court in AAA Aerostar, in commenting on the uncertainty of whether prejudice to the insurer must be shown before it is relieved of its duty to indemnify, clearly made a distinction between notice-of-occurrence cases and notice-of-lawsuit cases. AAA Aerostar, 294 Ill. App. 3d at 807 . . . .

"However, an insurer does not have to prove that it was prejudiced by an insured's breach of the notice clause in a policy in order to be relieved of its duty to pay. Twin City Fire Insurance Co. v. Old World Trading Co., 266 Ill. App. 3d 1, 639 N.E.2d 584 (1993). Lack of prejudice to the insurer is a factor to be considered only where the insured has a good excuse for the late notice or where the delay was relatively brief. Old World Trading Co., 266 Ill. App. 3d at 8-9; Fletcher v. Palos Community Consolidated School District No. 118, 164 Ill. App. 3d 921, 518 N.E.2d 363 (1987). In the instant case, Ward did not present a good excuse for the late notice, nor was the delay relatively brief; therefore, prejudice to Imperial is not a factor to be considered." 324 Ill. App. 3d at 449.

Under Kansas law, no distinction has been brought to this court's attention between the need for a showing of prejudice in notice-of-occurrence and notice-of-suit cases. In Cessna, notice-of-occurrence seemed to be at issue, but notice-of-suit cases were included in the discussion. In a more recent case, this court extended its rule from notice to a proof of loss. The court entertained the certified question whether the principle anchoring Kansas courts' prejudice requirement for avoiding insurance coverage for untimely notice would be extended to an untimely proof of loss. National Union Fire Ins. Co. v. FDIC, 264 Kan. 733, 957 P.2d 357 (1998). After examining cases from a number of jurisdictions as well as Kansas law, the court found that the distinction between notice and proof of loss was not significant and the better-reasoned case could be made for requiring prejudice rather than not requiring it. Thus, this court has not indicated that it would make a distinction between notice of an occurrence and a suit. We conclude the trial court's decision on notice, as a matter of law, is correct.

In our discussions of the substantive issues, we have seen that the district court's decisions accord with or can be reconciled with the law of Illinois on the number of occurrences, trigger of coverage, and nonfortuity and known loss. The district court's decisions on Santa Fe's self-insured retentions and allocations do not comport with Illinois law. However, we have reversed those decisions. As to the timely notice issue only, Illinois law is not established, with districts of the Illinois appellate courts holding different views, some in accord with established Kansas law.

We conclude that the laws of Kansas and Illinois are not in conflict on the questions in the present case. With no conflict, there is no reason to consider choice of law.

Affirmed in part, reversed in part, and remanded.