Board of Trustees of the University of Illinois v. Insurance Corp. of Ireland, Ltd.

MEMORANDUM OPINION AND ORDER SHADUR, District Judge. University of Illinois’ Board of Trustees *1376 (“U of I”) 1 seek alternative relief here — either issuance of a declaratory judgment as to the limits of an insurance policy (the “Policy”) issued to U of I by Insurance Corporation of Ireland, Ltd. (“ICI”) or reformation of the Policy to reflect the parties’ original intent. U of I now moves under Fed.R.Civ.P. (“Rule”) 56 for summary judgment in one of the following forms: 1. a declaration that the Policy already provides for total coverage of $10 million ($5 million in the aggregate for the period March 1 through June 30, 1984 and a separate aggregate of $5 million for the policy period July 1, 1984 through July 1, 1985); or 2. reformation of the Policy to provide for a separate $5 million aggregate limit for each of those two periods. For the reasons stated in this memorandum opinion and order, U of I’s motion for summary judgment for such a declaration of the policy limits is denied, but its motion for summary judgment to reform the Policy is granted. Facts 2 U of I is a body politic and public corporation formed and acting pursuant to Illinois law (Ill.Rev.Stat. ch. 144, 11 22). All members of its Board of Trustees are Illinois citizens. ICI is a foreign corporation 3 with its principal place of business in Ireland. Those facts establish diversity-of-citizenship jurisdiction under 28 U.S.C. § 1332(a)(2). From approximately 1983 through April 1987 ICI was licensed to and did transact business in Illinois, maintaining an office in Chicago. In January 1984 U of I issued bid specifications seeking quotes on excess public, hospital and medical professional liability insurance covering a one-year period from March 1, 1984, with an option to renew for two additional years. U of I’s insurance broker Marsh & McLennan (“M & M”) issued a quote in accordance with the specifications, and ICI then bid for the insurance coverage in a somewhat different but legally equivalent form (three years, but cancellable at the end of any policy year). As a result of the ensuing negotiations, ICI and U of I agreed to the terms of insurance policy # SC8-16019-0 (already designated the “Policy” in this opinion) for the period from March 1, 1984 to March 1, 1987. 4 In relevant part the Policy reads: SUPPLEMENTAL DECLARATIONS Item 1. LIMIT OF LIABILITY $5,000,-000 Combined Single Limit each and every occurrence and in the aggregate. Excess of a Self Insured Retention of $100,000 each and every occurrence $1,000,000 overall annual aggregate *1377 Item 2. PREMIUM $1,485,000 (triennial) payable Vk annually. Endorsement # 1 initially specified that the premium installments (Item 2) were payable on March 1, 1984, March 1, 1985 and March 1, 1986 and that the “premium will be subject to review as of the anniversary dates.” During the first quarter of 1984, however, U of I asked that the premium payment dates should be changed to coincide with U of I’s fiscal year, which begins on July 1. On February 8, 1984 Paul Burston (“Burston”), a broker at M & M, reported to U of I that he was therefore asking ICI to issue an insurance policy to cover the periods from March 1 to July 1, 1984, July 1, 1984 to July 1, 1985, July 1, 1985 to July 1, 1986 and July 1, 1986 to July 1, 1987. ICI agreed, and M & M then received $165,000 5 from U of I on March 1, 1984 as the premium for the four-month period beginning on that date and ending July 1, 1984. To reflect its earlier agreement to U of I’s request, on May 23, 1984 ICI issued Endorsement # 3, amending the policy period to read “From March 1, 1984 to July 1, 1987.” Endorsement # 3 also modified Item 2 of the Supplemental Declarations to read: $1,650,000 (triennial plus three months [sic — obviously should have said “four months”]) payable as per Premium Payment Schedule as follows: Billing Date Amount 3/1/84 $165,000 7/1/84 495,000 7/1/85 495,000 7/1/86 495,000 It went on to say that “All other terms and conditions [of the original Policy] remain unchanged.” There was an occurrence during the Policy’s initial four months (the stub period from March 1 to July 1, 1984) that triggered a potential U of I liability, which ICI later settled for $4.5 million over and above U of I’s retention obligation. ICI has can-celled the Policy effective July 1, 1985, leaving only a 16-month effective coverage period (March 1, 1984 to July 1,1985). And because the Policy was an occurrence policy rather than a claims-made policy covering that period, the issue of the amount of coverage remains a live controversy. At this point the parties vigorously dispute the amount of coverage that U of I had during the bobtailed effective period of the Policy. Much of their disagreement centers on their differing readings of Endorsement #3, which changed both the coverage period and the premium payment dates. U of I asserts that the purpose of the endorsement was to change the Policy to cover an initial four-month period with a separate aggregate limit of $5 million, followed by three annual periods — each having its own separate annual aggregate of $5 million in coverage. 6 ICFs totally different view is that the “policy was extended four months so as to expire on July 1, 1987, which coincided with the beginning of U of I’s fiscal year,” and that the $5 million aggregate coverage is all that U of I would have had for the entire Policy period of 40 months (ICI’s Response to U of I’s Statement of Undisputed Material Facts 3). 7 To support their respective positions, U of I submitted affidavits to prove and ICI submitted depositions to rebut U of I’s assertion that the mutual intent of the parties was to provide annual aggregate limits *1378 of coverage of $5 million and that the initial block of four months was intended to serve as a separate annual period. 8 Each affidavit and deposition gave an interpretation of the coverage limits for the premium payment periods of March 1 to July 1,1984, July 1, 1984 to July 1, 1985, July 1, 1985 to July 1, 1986 and July 1, 1986 to July 1, 1987. U of I’s Risk Manager James R. Gallivan (“Gallivan”) stated in bottom-line terms (Gallivan Aff. ¶ 4) that the Policy was intended to provide a $5 million aggregate coverage for each of the four separate payment periods. At his deposition, however, Gallivan could not recall any specific conversations with I Cl’s Fleming and another ICI representative establishing an agreement with ICI for the Policy limits, though he was sure that such conversations had taken place (Gallivan Dep. 34, 38). Relatedly he testified {id. 43) that no separate bidding procedure was used to establish the separate coverage for an extension of the original policy period because: We were anxious to secure a commitment for a reasonable length of time, and the ability to purchase more than one year of coverage was not known at the time we went out to solicit bids to the insurance markets at that time. U of I’s broker Burston, who “was involved in the placement of” the policy (Burston Aff. ¶ 3), also stated that the Policy was to provide $5 million aggregate coverage for each premium period {id. ¶ 5). But it is not clear whether Burston’s assertion was based on his recollection of any negotiations between the parties, or his recollection of the intention on U of Fs part when the Policy was being negotiated, or simply upon his interpretation of the Policy at the time of his affidavit. What is significant under the unusual circumstances of this case is that both Gallivan and Burston— and hence U of I — did in fact intend that the $5 million in coverage would apply to each of the original three annual periods. 9 As for ICI, its United States Manager from 1982 to 1985 William Fleming (“Fleming”) stated (Fleming Aff. ¶¶ 3-5) that he reviewed the Policy, that he was personally involved in its negotiation and that it “was to provide and did provide $5 million aggregate coverage” for each payment period. Later, however, Fleming testified that he could not remember negotiating the wording of the Policy (Fleming Dep. 44, 50), though he did recall that the Policy was to have a “conventional” 10 aggregate {id. 54). But most significantly Fleming was unequivocal in his recollection of the intention on his (and hence ICI’s) part for an annual rather than overall limit of $5 million in coverage, based on ICI’s knowledge of a prior policy of identical type {id. 85-86): Q. Under the previous ICI policy, Beacon ICI policy, it was your opinion that that provided a separate annual aggregate limit, correct? A. It was, yes. That was our intent at all times, yes. Q. And that intent was carried out with this policy as well? A. This was just a renewal of another one, yeah. Thus Fleming’s affidavit was primarily based on his specific intention in that respect (which for this purpose means Id’s intention) as well as his analysis of the Policy, although not from a recollection of the actual negotiations {id. 55). ICI’s other key player was Frederick Radichel (“Radichel”), its Assistant United States Manager from May 1983 to December 1986. Radichel Aff. 114 stated: My understanding and interpretation of the policy was that it was to provide *1379 separate aggregate limits of coverage for each billing period reflected in Endorsement No. 3 to the policy (Exhibit F). Radichel’s later testimony, however, reflected that he was not involved with the actual underwriting of the Policy — he said that was Fleming’s role (Radichel Dep. 29). But he too testified that he specifically understood from the outset that the aggregate coverage was $5 million for each annual period of the Policy [id. 46, 69), although a dispute arose sometime after July 17, 1984 as to the applicable aggregate for the initial four-month period {id. at 48, 50). As to that last-mentioned dispute, Radi-chel could not confirm and in fact denied that the parties had agreed to have a $5 million aggregate for the initial four months of the policy {id. 55-56). What he said was that ICI “didn’t have a four-month policy” and that he had no knowledge of any understanding between the parties regarding a separate aggregate for that period {id. 67). Radichel sought to reconcile any apparent inconsistencies between his earlier-tendered affidavit and his later deposition testimony by explaining that he intended his affidavit to state that “[t]here is a separate aggregate policy limit for each billing period” but that it was not his position that the same aggregate that would apply for an annual term would apply to a four-month term {id. 69). Issues for Decision At issue in this case is the coverage to which U of I is entitled under the Policy. As for the document itself, its two relevant portions are the Supplemental Declarations of the original Policy and Endorsement # 3. Although the original Policy as amended would potentially have covered a period of three years and four months, the effective Policy period was cut short by an early cancellation of the Policy after only one year and four months. Thus U of I obtained coverage only for the period from March 1, 1984 to July 1, 1985. This Court is called upon to determine whether there is any genuine issue of fact to prevent a decision on U of I’s motion and, if not, what the Policy limit or limits is or are for that entire effective period. Because U of I initially seeks a declaration of the meaning of the contract language, this opinion will first examine that Policy language to determine whether it is ambiguous. As the discussion of that issue will reflect, the language is clear and does not warrant judicial construction. That however simply shifts the focus of inquiry to whether that unambiguous language does or does not reflect the mutual intention of the parties. And because that question gets a negative answer and because the parties shared an identical intention, the necessary result is to call for reforming the contract to correspond to the mutual intent of the parties. Ambiguity in the Policy Under Illinois law, 11 a court’s primary purpose in construing insurance contracts is to give effect to the expressed intention of the parties (Allstate Insurance Co. v. Elkins, 77 Ill.2d 384, 390-91, 33 Ill.Dec. 139, 142, 396 N.E.2d 528, 531 (1979)). United States Fire Insurance Co. v. Schnackenberg, 88 Ill.2d 1, 5, 57 Ill.Dec. 840, 842, 429 N.E.2d 1203, 1205 (1981) (citations omitted) explains: *1380 [I]f the provisions of the insurance policy are clear and unambiguous there is no need for construction and the provisions will be applied as written. All the provisions of the insurance contract, rather than an isolated part, should be read together to interpret it and to determine whether an ambiguity exists. * * * * * * In applying the rules of interpretation, the words in the policy should be given their plain and ordinary meaning, and the court should not search for an ambiguity where there is none. 12 Hence the first line of inquiry is a look at the Policy to determine whether there is any ambiguity in its language. And because both parties have focused their searchlights on the meaning of the word “aggregate” in the paragraph that sets forth the limits of ICI’s liability, this opinion will examine the meaning of that word in the context of the entire Policy. U of I Mem. 9 argues that the plain and ordinary meaning of “aggregate” could denote either a “total” or an “annual” aggregate. 13 But that contention does not reflect the common understanding of the word. ICI points to the dictionary: Webster’s Third New International Dictionary 41 (1976) defines “aggregate” as “the whole sum or amount: SUM TOTAL.” In the absence of a modifying adjective or phrase, the conventional meaning of “aggregate” would therefore signify the “sum total” over the entire length of the contract. Indeed, “total aggregate” is as redundant as “total sum total,” and it is apparent from the face of the Policy that the word “aggregate” as it addresses the Policy limits would normally denote the limits applied over the entire duration of the policy. In contrast with the Policy’s characterization of the coverage limits, the Self Insured Retention (“SIR”) limits of $1 million were clearly designated on the same page of the Policy as an “overall annual aggregate.” In terms of Policy language alone, the fact that it thus specified that the SIR limits were to accumulate within each “annual” period, while the word “aggregate” was left unmodified by “annual” in the other paragraph in which the Policy limits are described, buttresses ICI’s *1381 assertion that in literal terms the latter provision refers to the aggregate for the original three-year Policy term. Thus both the plain-meaning approach and the use of the word in the Policy identifies “aggregate” as an unambiguous term that need not be evaluated further. U of I’s argument is a bit different from one that would find an intrinsic meaning of “annual” within the word “aggregate.” Instead U of I's position seems to be that “aggregate” can be limited or expanded by an adjective that is absent from the Policy language. On that premise the issue would not be the meaning of the word “aggregate” in isolation, but rather the lack of a modifier that would change the focus of the word from the overall policy term to separate annual periods. In those terms, however, the predicate for a shift in meaning would be a language omission, not an ambiguity. And on that score Consolidated Bearings Co. v. Ehret-Krohn Corp., 913 F.2d 1224, 1233 (7th Cir.1990) (citing Illinois cases) says: Silence creates ambiguity, however, only when the silence involves a matter naturally within the scope of the contract as written. A contract is not ambiguous merely because it fails to address some contingency; the general presumption is that “the rights of the parties are limited to the terms expressed” in the contract. Here the claimed “silence” in the Policy does not leave the meaning of the word “aggregate” ambiguous. That word requires no modifier to have a natural meaning, which in the literal sense would call for the $5 million in coverage to apply over the sum total of the Policy term. Despite such apparent clarity, U of I also urges recourse to extrinsic evidence for proof of an ambiguity. But this is not a situation in which (FDIC v. W.R. Grace & Co., 877 F.2d 614, 620 (7th Cir.1989)): [Although the agreement itself is a perfectly lucid and apparently complete specimen of English prose, anyone familiar with the real-world context of the agreement would wonder what it meant with reference to the particular question that has arisen. In fact, the extrinsic evidence that U of I would call into play does not show that the word “aggregate” is capable of more than one meaning. At most it shows that the Policy unintentionally omitted a word that would have redirected the effect of “aggregate” by giving it a separate application to each one-year period. Such nonpurposeful silence is not a ground upon which a court may construe a contract that is clear on its face (as contrasted with the contract’s possible reformation, which will be looked at next). 14 LaSalle National Bank v. Service Merchandise Co., 827 F.2d 74, 78 (7th Cir.1987), applying Illinois law, states the operative principle: The starting point must be the contract itself. If the language of the contract unambiguously provides an answer to the question at hand, the inquiry is over. 15 Here the meaning of “aggregate” is indeed unambiguous. Hence U of I’s motion for summary judgment on the issue of whether the Policy expressly provides for separate aggregate limits of liability is denied. *1382 Reformation of the Policy As stated at the outset of this opinion, U of I alternatively asks for reformation of the Policy to reflect a claimed intention of the parties to provide a $5 million aggregate for each of two periods: one from May 1 to June 1, 1984 and the other from June 1, 1984 to June 1, 1985. Under Illinois law parties have a right to contract reformation when their written documentation does not reflect the true intent of the parties due to mutual mistake. Magnus v. Barrett, 197 Ill.App.3d 931, 934-35, 145 Ill.Dec. 482, 485, 557 N.E.2d 252, 255 (1st Dist.1990) (citations omitted) explains: When the contracting parties to a policy of insurance have made a mistake and the policy fails to express the written contract between them, and provisions other than those intended are inserted or admitted [sic — this seems to be a garbled reference to the other alternative of intended provisions being “omitted”], equity has the right to grant relief by reformation of the contract. Even though the contractual language may appear to be plain and unambiguous and even though at law no extrinsic evidence may be used to show otherwise, the equitable remedy of reformation allows such evidence to establish the true intent of the parties. That is scarcely a novel proposition — it was stated as far back as Justice Story’s 1 Commentaries on Equity Jurisprudence as Administered in England and America § 240, at 239 (14th ed. 1918 16 ) (footnotes omitted): For however just in general the rule may be, “Quoties in verbis nulla est ambigui-tas, ibi nulla expositio contra verba expressa fienda est”; yet that rule shall not prevail to defeat the manifest intent and object of the parties, where it is clearly discernible on the face of the instrument, and the ignorance or blunder or mistake of the parties has prevented them from expressing it in the appropriate language. What is perhaps novel in this case is the manner in which the parties’ true intent is evidenced. This opinion has already spoken briefly of the objective principle of contract law, with its focus on communicated intent (see n. 12). Just as a like communicated-intent policy underpins such sources of contract doctrine as the “twenty bishops” concept and the statute of frauds, such insistence on communicated intent seeks to avoid any easy effort by a party to squirm out of its agreement by hoping that a trier of fact will buy the argument “Gee, I really didn’t mean that even though I said it.” “Meeting of the minds” is ordinarily a metaphysical concept of little use in fashioning contract principles. In this instance, however, it has a real conceptual predicate and special value. There is no potential for such squirming out here, for ICI’s own controlling agents (and therefore ICI itself) have confirmed the identical mindset as have U of I’s agents (and therefore U of I itself): Both sides intended that the initial version of the Policy would have an annual rather than an overall aggregate limit of $5 million in coverage. All of Fleming, Radichel, Gallivan and Burston said so. It therefore makes no difference that one or more of them could not recall the facts as to the communication of that intention — such uncontradicted evidence that there was an identical shared intent is enough to cause the documentation, which did not mirror that shared intent, to be conformed to what both sides really wanted their contract to be. And if that were not enough (as it certainly should be), it is fully confirmed by the only objective evidence that is available to assist in ascertaining the parties’ intention. It will be remembered that U of I had requested, in its original bid, coverage for one year with a coverage limit of $5 million. Nothing in the evidence even suggests that the resulting Policy, though it expanded the total term from one year to three, provided U of I coverage for each of *1383 those years in an amount that dramatically reduced the amount of that intended coverage — to a figure that would provide (in average terms, though not necessarily in any particular year) only one-third of the requested coverage per year. 17 At the first step, then, U of I is plainly entitled to relief under this Court’s equitable power of reformation, having met the test set forth in Sheldon v. Colonial Carbon Co., 116 Ill.App.3d 797, 800, 72 Ill.Dec. 289, 291-92, 452 N.E.2d 542, 544-45 (1st Dist.1983) (citations omitted): There is a presumption that a written instrument conforms to the intention of the parties thereto. Therefore, in an action for reformation, plaintiff has the burden of proving by very strong, clear and convincing evidence that there has been a meeting of the minds resulting in an actual agreement between the parties; but, at the time the agreement was reduced to writing and executed, some agreed-upon provision was omitted or one not agreed upon was inserted either through mutual mistake or through mistake by one party and fraud by the other. There is no question but that the Policy at its inception must be reformed, in accordance with the uncontroverted mutual intent of the parties, to one that would have provided U of I with $5 million in coverage for each of its one-year periods at an annual premium of $495,000. That then leads to the second-level question: What was the effect of the addition of four months to the Policy term in consideration of an additional premium of $165,000? For that purpose the Policy must be viewed as though it had been written in the manner just described in the preceding paragraph as its reformed version. As ICI would have it, U of I paid that extra $165,000 to receive no greater coverage than what ICI claims was the originally-bargained-for $5 million in the aggregate. Conversely U of I urges that it paid the $165,000 for $5 million in coverage for four months, obviously on the premise that the insurer’s risk of encountering $5 million in claims in four months is one-third of the risk of doing so in 12 months. Framing the issue in those terms demonstrates that U of I is right and ICI is wrong. To see why that is so, we may examine the kind of premium that an insurer would be expected to charge under alternative scenarios, assuming in each instance a liability insurance policy for which the risks were such that $5 million in total (“aggregate”) coverage for a year would call for a $495,000 premium: 1. What premium would the insurer charge for a policy that carried the same $5 million of total risk, but over a two-year rather than a one-year coverage period? Although the answer to that question cannot be quantified precisely without knowing more about the various factors that enter into the setting of premiums, it is morally certain that the total premium would be more than $495,000 (because the insurer’s period during which it is at risk has just been increased) but something less than $990,-000 (because the amount of the insurer’s risk has not precisely doubled — even though the period of coverage has doubled, the nature of the policy as one that provides diminishing coverage, plus the fact that the total dollar exposure of the insurer has remained constant, negates an outright doubling of the risk). 2. What premium would the insurer charge for a policy that carried $5 million of risk for each of two yearly periods, looked at separately? Here the answer is easy — $990,000 (for each year carries the same $495,000 premium for the same $5 million of total risk for a like period). And why is that second proposition true? Purely and simply, it is because the statistical likelihood of encountering any particu *1384 lar claim 18 is a direct function of time: It is twice as likely that a claim will be made against an insured in either one of two years as it is that the same claim will be encountered in one year. To move in the other direction, it is one-third as likely that a claim will be made against an insured in four months as it is that the same claim will be encountered in one year. Thus when U of I paid another $165,000 for the tacking on of a four-month period at the beginning of the Policy, that had to be for the obtaining of $5 million in coverage for that stub period. In this instance it makes no difference that the parties did not talk about the coverage issue— the result flows from the one item of totally objective evidence, the premium itself. It is instructive on the issue of intent to note that U of I’s conduct was entirely consistent with the result just reached. In addition to obtaining the Policy itself, U of I negotiated for and obtained an umbrella policy for coverage in excess of that provided by ICI. That umbrella policy was written for two periods — one from March 1 through July 1, 1984 (corresponding to the just-discussed four-month stub period under the Policy) and the other for the ensuing year from July 1, 1984 to July 1, 1985 (the identical period covered by the Policy as it was terminated by ICI). That umbrella policy (1) afforded U of I $5 million in aggregate coverage for each of those two periods, (2) referred specifically to the Policy with ICI as providing the underlying coverage and (3) for each period identified the ICI Policy coverage as $5 million — the same coverage that it has here urged upon this Court. To return to the Policy itself, nothing in what has been advanced by ICI on the current motion undercuts the conclusion— compelled both (1) by Fleming’s acknowl-edgement of ICI’s intention and (2) by the objective fact of premium payment — that a separate $5 million in aggregate coverage was applicable to the first four months of the Policy term. That conclusion wraps up the necessary analysis. Conclusion From the analysis of the Policy as a whole and the uncontroverted extrinsic evidence it is clear that the parties, through a mutual mistake, failed to have the document conform to their mutual intent as to the amount of Policy coverage. 19 Further *1385 more, the evidence is equally uncontrovert-ed — there is no “genuine issue of material fact” — as to the nature of that mutual intent in calling for an annual rather than an overall aggregate. And finally, ICI’s own representative has acknowledged that an identical separate aggregate limit of $5 million was intended to apply to the four-month stub period — an intent entirely consistent with the logic of insurance coverage and with the separate premium charged for that period. In the absence of any genuine issue of material fact, this Court therefore exercises its equitable power to reform the Policy to reflect the parties’ intention that: *1387 fied Cole at a police line-up on the day after the murder. *1385 1. For the period from March 1 to July 1, 1984, there is an aggregate limit of $5 million in coverage. 2. For the period from July 1, 1984 to July 1, 1985, there is a separate aggregate limit of $5 million in coverage. Footnotes [1] As reflected in the "Facts” section, Illinois designates U of I as a "body corporate and politic." Although Ill.Rev.Stat. ch. 144, ¶ 22 refers to the Board of Trustees in plural terms ("as they shall deem necessary"), that usage would strike the ear as awkward where (as in this opinion) “U of I” is used as a collective noun connoting the Board of Trustees. Accordingly, after the first sentence of the text (to which this footnote is appended) this opinion will follow the convenient (though imprecise) practice of using singular verb forms and pronouns when referring to U of I. [2] Familiar Rule 56 principles impose on the movant the burden of establishing the lack of a genuine issue of material fact (Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 2552-53, 91 L.Ed.2d 265 (1986)). For that purpose this Court is called on to draw all "reasonable inferences, not every conceivable inference" in the light most favorable to the nonmovant — in this case ICI (DeValk Lincoln Mercury, Inc. v. Ford Motor Co., 811 F.2d 326, 329 (7th Cir.1987)). [3] “Foreign” is used here in the literal sense — a corporation that is a citizen of a foreign state, the Republic of Ireland — and not merely to designate an entity incorporated in a state other than Illinois. Indeed the parties have advised orally (though not in a formal pleading) that Irish corporation ICI is majority-owned by the Republic of Ireland itself. If that had been properly confirmed on the current motion, it would support an independent ground of federal jurisdiction under the Foreign Sovereign Immunities Act (28 U.S.C. §§ 1603(a), 1603(b)(2), 1603(d) and 1603(e), coupled with 28 U.S.C. § 1605(a)(2)) and 28 U.S.C. § 1330(a). But because diversity of jurisdiction plainly exists, there has been no need to formalize the added facts needed to support that second and independent potential basis for federal jurisdiction. [4] Although the agreement as to the Policy’s coverage was entered into before the beginning of the Policy period (March 1, 1984) and although the critical endorsement (# 3) was signed by ICI effective in May 1984, the Policy was "issued” on July 17, 1984. [5] Simple arithmetic confirms that $165,000 is exactly one-third of $495,000, which is in turn exactly one-third of the $1,485 million premium for the original triennial period. Hence it is clear that the parties simply extrapolated the initially-agreed-upon $1,485 million premium for the three-year term on a pro-rata basis to deal with the additional four months of coverage. [6] Under that reading the later cancellation of the Policy as to the last two years would lop off the $5 million in coverage for each of those periods, leaving intact the first two blocks of $5 million each. [7] Although ICI does not explain just how or why this should be so, it considers the cancellation of the last two years of the Policy terms as still leaving the same $5 million in aggregate coverage in effect for the first 16 months of the Policy term that had not been affected by the cancellation. Even apart from the matters discussed at length in this opinion, that alone is somewhat disquieting as to the logic of ICI’s position on the intended deal between the parties. [8] Of course, in the context of responding to a summary judgment motion ICI need only present evidence to show the existence of a disputed material fact — it need not prove that its view is correct. [9] For the moment it is unnecessary to address U of I’s intention as to the added four-month stub period — more on that subject later. [10] "Conventional” was a term he used to describe the usual policy arrangement — one that (consistently with the already-referred-to U of I intention) involved aggregate coverage provided on an annual basis and not on a longer period, even though a policy might he written for more than a single year. [11] Because this case sounds in diversity, this Court must look to Illinois law — including its choice of law doctrines — for the applicable rules of decision (see Autocephalous Greek-Orthodox Church of Cyprus v. Goldberg, 917 F.2d 278 (7th Cir.1990)). Purcell & Wardrope Chtd. v. Hertz Corp., 175 Ill.App.3d 1069, 1079, 125 Ill.Dec. 585, 592, 530 N.E.2d 994, 1001 (1st Dist.1988) teaches that Illinois courts use the "most significant contacts" test for contract disputes: The factors to be considered in determining which state has the most significant relationship to a contract claim are: the place of negotiation, the place of execution or contracting, the place of performance, the location of the subject matter of the contract, and the domicile, residence, place of incorporation, and the business of the parties. Although ICI is an Irish corporation, it maintained an office in Chicago from which it issued the policy to U of I, which itself has its principal place of business in Illinois. Both the performance of the contract and its subject matter are in Illinois. All the significant signals thus point to Illinois — indeed each side has focused on Illinois substantive law in its briefing. [12] [Footnote by this Court] That statement of Illinois law is all of a piece with the general principle that the reading of contracts is normally a search for the objective meaning of the language used, not what one contracting party now says that it had in mind but did not communicate to the other party (see Learned Hand’s classic "twenty bishops” statement in Hotchkiss v. National City Bank of New York, 200 F. 287, 293 (S.D.N.Y.1911), aff’d, 201 F. 664 (2d Cir.1912), aff’d, 231 U.S. 50, 34 S.Ct. 20, 58 L.Ed. 115 (1913)). And of course it cannot be ignored that a litigant's current statement of its alleged frame of mind at an earlier date comes after a dispute has already arisen — so that a court has reason to be concerned lest the statement of purported current recollection may be the product of wishful thinking or worse. As the later discussion in the text will reflect, however, this case presents a special — an almost unique — occasion for departing from the objective-meaning-of-language approach. [13] U of I points out that North Carolina ex rel. Long v. Beacon Insurance Co., 372 S.E.2d 98 (1988) (unpublished order) ruled that the identical language in a similar policy involving the same parties was ambiguous. This Court does not agree with that legal conclusion — but what is significant for current purposes is that ICI's Fleming had the understanding that the Beacon Insurance policy (which had originally been issued by ICI but was then amended so as to be issued by Beacon and 100% reinsured by ICI) and ICI’s Policy at issue here were both intended to provide annual aggregate limits, not aggregate limits over the entire policy term (Fleming Dep. 85-86, quoted earlier). That intention bears critically on the subject later analyzed in this opinion. As an aside before leaving the subject of the North Carolina litigation, this opinion notes that it was settled by a June 20, 1988 agreement embodied in a consent decree, which provided in part for a dismissal of the then-pending appeal on the policy coverage issue. Apparently the settlement payments and the filing of a motion before the North Carolina Court of Appeals did not take place in time to head off issuance of the opinion referred to at the beginning of the footnote. But the resolution of the earlier litigation by settlement prevents ICI from being barred in issue preclusion terms from arguing that the Policy term "aggregate” is unambiguous rather than ambiguous (as the North Carolina opinion had decided it was). It remains true that ICI’s efforts to distinguish the North Carolina litigation and the policy at issue there (Answer ¶¶ 10-15 and Mem. 6 n. 3) from the current litigation and the Policy are specious at best — and truly unpersuasive. [14] W.R. Grace, id. at 620-22 presents an enlightening exegesis of the Illinois case law and explanation of the conceptual basis for not adhering to the wooden and hidebound “four corners" approach to contract reading. Those Illinois cases sometimes talk in terms of allowing extrinsic evidence to create an ambiguity in what is otherwise seemingly unambiguous language. And it might perhaps be possible, with considerable straining of language, to manufacture an ambiguity in this instance and in that way to enable the parties’ unexpressed intent to carry the day. But that strikes this Court as a result-oriented route, one to be avoided precisely because it is result-oriented — -and doubly so where as here there is a perfectly sound path that leads to the same destination. [15] [Footnote by this Court] It is, however, important to observe (especially as a lead-in to the next section of this opinion) that the quoted language came immediately on the heels of a statement that under Illinois law "[t]he primary object in construing a contract is to give effect to the intention of the parties.” This opinion proceeds next to serve that primary object. [16] This Court does not have access to an earlier edition of Justice Story's work, but the style of the quoted excerpt and the use of the Latin quotation (and, importantly, the explanatory preface by the editor of the 14th edition) all suggest strongly that the quoted principle is traceable to Justice Story himself and to his original publication of the work in 1835. [17] That evidence is indirect, of course. But such is allowed in determining the original intent of parties to a contract. As Justice Story explained in 1 Equity Jurisprudence § 234, at 234: Courts of Equity will grant relief in cases of mistake in written contracts, not only when the fact of the mistake is expressly established, but also when it is fairly implied from the nature of the transaction. [18] For this purpose the analysis may be simplified by assuming a single claim that would blot up the entire aggregate coverage for any Policy period. When the likely incidence of claims is considered, the same analysis would extend to the probability of encountering (for example) two claims aggregating the Policy limit within any given time frame. [19] It might seem counterintuitive to find such a blunder in an important commercial transaction with millions of dollars at stake. Not so. Nothing presented by the parties has suggested the kind of care that we like to think would be extended to the preparation of documents if outside lawyers had been brought into the matter (and one need go no farther for evidence of sloppiness even on small points than to note the careless labeling of the March 1-July 1 stub period as three months rather than four). Nor is this to suggest that lawyers' intervention would have insured (to make a bad pun) a better result by avoiding the problem altogether. Any regular reader of the advance sheets — or worse, anyone regularly forced to write opinions that find their way into the advance sheets — has to be struck by the extraordinarily bad drafting that marks the insurance trade and its scriveners. By chance Judge Posner of our Court of Appeals has just written a keenly insightful opinion on the subject of why such confusing and litigation-breeding language finds its way into — and then persists in — insurance contracts (Continental Casualty Co. v. Pittsburgh Corning Corp., 917 F.2d 297, 299 (7th Cir.1990)). And a quarter century ago David Mellinkoff in his The Language of the Law 374-76 (1963) suggested the same phenomenon as an almost inevitable consequence of risk aversion in the legal profession. That same lesson recurs elsewhere in Mellinkoff’s book — but the following version is that of the writer here (if flawed, no responsibility is to be ascribed to Mellinkoff or anyone else): 1. Most frequently the poorest drafted documents that leave plenty of room for argument as to their meaning, rather than those whose meaning is indisputable, find their way into litigation and into the reported appellate opinions. 2. At the next level of activity, the lawyer who is later called on to draft a new contract exercises lawyerlike caution by repeating the language that now has an "established meaning” (even though it took an appellate court to tell us that the first set of lawyers, who had used the word “gray,” really meant "black” or *1385 "white," that now magically converts into an equation in which "gray” is "black” or “gray” is "white”). 3. As with barnacles on a hull or patches on an inner tube, the contracts tend to grow by accretion — combining as well as repeating the ill-chosen language that has been derived from earlier documents in the manner just described. That practice creates the maximum potential for inconsistency and more ambiguity (a subject well addressed in Continental Casualty Co., 917 F.2d at 299. And so it goes, forcing the issuance of still more opinions such as this one.