Gordon B. Phillips (“Phillips”), as guardian for his depéndent sbn, James G. Phillips (“James”), brought this action in the district court under the Employee Retirement Income Security Act of 1974 (ERISA), 29 U.S.C. §§ 1001-1461, against Lincoln National Life Insurance Company (“Lincoln”) seeking recovery of benefits under an employee welfare benefit plan (“the Plan”) established by his employer, Seedboro Equipment Company (“Seedboro”), pursuant to its purchase of an insurance policy from Lincoln. Both parties agree that the Plan is an employee welfare benefit plan governed by ERISA. Phillips argued in the district court that Lincoln erroneously applied the Plan’s mental illness benefit limitation to Phillips’ benefits claim for expenses incurred in the treatment of James’ illness. The district court entered summary judgment in favor of Phillips, and Lincoln, appeals. We affirm.
I. BACKGROUND
On November 1, 1984, Seedboro established a benefit plan for its employees through its purchase of a group health insurance policy from Lincoln. The Plan defined the health insurance benefits available for eligible employees of Seedboro and their dependents. Phillips, who is Seed-boro’s president, and his dependent son James were eligible participants in the Plan. The Plan provisions relevant to the instant dispute are as follows:
MAJOR MEDICAL BENEFITS
Maximum Benefit (Lifetime
Aggregate)............... $1,000,000
Except that, the Maximum Benefit (Lifetime Aggregate) for charges for mental illness(es) is........... $ 25,000
s|< * * *
MAJOR MEDICAL LIMITATIONS
The maximum payment for care of mental illness or care of nervous conditions of any type or cause by a doctor will not be more than:
1. $20.00 for each visit; and
2. one visit on any one day; and
3. 50 visits during any calendar year.
(A “visit” occurs each time the doctor provides care to the patient).
* # * *
DEFINITIONS
Illness-means:
1. A disorder or disease of the body or mind...............................
* * * *
The Plan provided no definition of the term “mental illness.”
In December, 1981, when he was 16 years old, James was diagnosed as suffering from what doctors call “congenital encephalopathy.” Congenital encephalopathy is a brain malfunction, manifested by neurological defects, whose exact location in the brain cannot be pinpointed. James’ condition was diagnosed through extensive neuropsychological testing measuring language ability, motor and perceptual skills, visual and verbal memory, and intelligence. In addition, a brain scan performed, in 1989 revealed that a portion of James’ brain was “not putting out the electrical impulses or processing electrical stimuli and impulses the way it would normally.” In James’ case, his neurological defect has given rise to a cluster of behavioral disorders. Those afflicted with this condition are referred to as suffering from “organic brain syndrome.” Dr. Erin Bigler, a neuropsycholo*305gist who has examined James, explained in a deposition that organic brain syndrome is
“a disorder of brain function that in turn affects the behavior of the patient in some way. Some organic brain syndromes only affect intellectual and reasoning abilities. Other organic brain syndromes may affect only certain aspects of emotional control or functioning. In a mixed organic brain syndrome it means that you basically have some diffuse brain dysfunction that’s affecting pretty much all aspects of behavior.”
According to Dr. Bigler, James suffers from mixed organic brain syndrome. James’ disorder manifests itself in numerous and varied behavioral episodes. James has what his doctors describe as “horrible” social skills and is “often very loud and obnoxious.” James frequently inappropriately touches others and engages in “repetitive mannerisms” such as rocking back and forth. James can be hyperactive, hy-perexcitable, “extremely self abusive” and “physically abusive to [others].” At times he is psychotic, suffers from periods of compulsive eating, and has rapid mood swings. James also has significant difficulties processing visual information. Although he is not mentally retarded, he does have a learning disability.
At the time of the filing of this appeal, James was being treated at Healthcare Rehabilitation Center in Austin, Texas (“Healthcare”). Healthcare specializes in the treatment and rehabilitation of persons suffering from neurological disorders. Healthcare does not provide general psychiatric treatment and refers patients who are diagnosed as requiring psychiatric care to other hospitals. Healthcare is licensed as a mental health facility in the state of Texas, but, according to Dr. John Batemen, the neuropsychiatrist who oversees James’ treatment at Healthcare, the facility is so licensed solely for the purpose of confining its patients for their own safety.
According to the record, medical science has been unable to implement a treatment program to successfully attack and contain the mental affliction from which James suffers. Therefore, doctors are forced to confine their efforts to helping James control and function with the behavioral manifestations of his illness. Under Dr. Bateman’s care, James is treated with medications to control his neurological problems and lives in a monitored environment where he is hopefully learning how to cope with his condition. James is also given low dosages of medications to ameliorate some of the behavioral manifestations of his illness, such as irritability and extreme anxiety.
Dr. Bateman stated in his deposition that James does not have a mental illness, but rather has an “organic mental disorder as one manifestation of his underlying neurological disease.” However, in a deposition Dr. Bigler agreed with the statement made by a Lincoln attorney that organic brain syndrome is an example of “a mental disorder or a mental illness” that is organically based. Both doctors agreed that certain mental illnesses can be organically based.
From the effective date of the Plan in 1984 until October, 1987, Lincoln paid $25,-000 in Plan benefits for the expenses Phillips incurred in the treatment of James’ illness. Having reached the Plan’s mental illness limitation cap, Lincoln refused to reimburse Phillips for any of the subsequent expenses arising from his son’s treatment. In early 1990, Phillips appealed Lincoln’s determination to the Lincoln National Appeals Committee, arguing that James’ illness was a physical condition which gave rise to certain mental deficiencies and thus fell Outside the scope of the Plan’s mental illness limitation. The Appeals Committee affirmed the original claim determination.
Following the decision of the Lincoln Appeals Committee, Phillips filed a complaint in the district court. In Count I of his complaint, Phillips sought a declaratory judgment that the Plan’s $25,000 mental illness limitation did not apply to his claim for benefits and further sought recovery of costs and attorneys’ fees pursuant to 29 U.S.C. § 1132(g)(1). In Count II of his complaint, Phillips sought recovery of monetary damages in the amount of denied benefits for expenses incurred in the treatment of James’ condition through Novem*306ber 1, 1990.1 Both parties filed motions for summary judgment.
In a published opinion, the district court explained that Phillips and Lincoln advanced “competing reasonable interpretations” of the Plan term “mental illness”. Phillips v. Lincoln National Life Insurance Company, 774 F.Supp. 495, 499 (N.D.Ill.1991). The court noted that Phillips argued in his summary judgment motion that the term “mental illness” referred only to those illnesses with non-physical causes, “such as illnesses traceable to abuse suffered in one’s childhood or to other types of traumatic experiences. According to [Phillips], behavioral problems traceable to an organic or physical cause are other than ‘mental illnesses.’ ” Id. Lincoln’s summary judgment motion took issue with Phillips’ definition of mental illness, arguing that the cause of a behavioral illness does not determine whether it is a mental illness. Id. at 500. In Lincoln’s view, an illness is a mental illness if its symptoms include extremely abnormal behavior. Id.
The district court questioned Lincoln’s definition of “mental illness”, reasoning that “aberrant behavioral symptoms are a necessary but not sufficient component of any definition of mental illness” because
“[t]here is no question that some conditions are marked primarily by symptoms of dementia and aberrant behavior yet would not be considered mental illnesses. Indeed, under Lincoln’s proposed ‘unambiguous’ meaning of ‘mental illness,’ an accident victim who exhibits abnormal behavior as the result of a traumatic head injury, a person suffering from brain cancer who develops unusual behavior and'an elderly person who has contracted Alzheimer’s Disease would all bé considered mentally ill.2
“The facts that James behaves abnormally and that he has received medication, psychotherapy and training to modify his behavior do not necessarily mean that he suffers from a mental illness. Those facts do not preclude the possibility that he might be suffering from a physical illness whose symptoms are behavioral.”
Id. at 501.
The district court, however, did not entirely reject Lincoln’s proposed definition of mental illness. Instead, the district court found that both definitions “have intuitive appeal and partially describe the qualities of a mental illness.” Id. Since the Plan offered no definition of the term “mental illness,” the district court concluded that the term as used in the Plan was ambiguous because it was susceptible to more than one reasonable interpretation. Id.
Faced with this ambiguity, the district court adopted the state law rule of contract interpretation contra proferentem in fashioning the federal common law governing the ERISA dispute between Lincoln and Phillips. Id. at 502. The rule of contra proferentem dictates that ambiguities in a contract be construed against the drafter. The district court thought this case presented an appropriate instance for the application of contra proferentem:
“Because of [Lincoln’s] failure to define mental illness [in the Plan], the parties were forced to litigate this dispute. [Lincoln], with its expertise and experience in drafting insurance contracts, could have anticipated that disputes surrounding the meaning of ‘mental illness’ were likely to arise and' could also have resolved the problems in advance by drafting a definition of ‘mental illness.’ ... [For example], it could have directly defined ‘mental illness’ with regard to the patient’s symptoms or the treatment rendered rather than saving such definitions for litigation.”
Id. Accordingly, the district court read the ambiguity in the Plan in Phillips’ favor, *307ruling that, as a matter of law, the Plan’s mental illness limitation did not apply to James’ condition. The district court denied Phillips’ request for an award of attorneys’ fees under ERISA. Id. at 503. The district court’s partial grant of summary judgment in Phillips’ favor did not include a determination of the amount Lincoln owed Phillips under the Plan. Lincoln appealed the summary judgment order; Phillips did not appeal the court’s denial of his request for attorneys’ fees.
II. STANDARD OF REVIEW
Phillips brought his suit against Lincoln under 29 U.S.C. § 1132(a)(1)(B) of ERISA which authorizes a plan participant to bring a civil action “to recover benefits due to him under the terms of his plan____” The Supreme Court has made clear that “a denial of benefits challenged under § 1132(a)(1)(B) is to be reviewed under a de novo standard unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” Firestone Tire & Rubber Company v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 956, 103 L.Ed.2d 80 (1989). The district court action in reviewing Lincoln’s denial of Phillips’ benefits claim de novo was proper as neither of these exceptions apply here.
We, in turn, “review de novo a district court’s grant of summary judgment.” United States v. First National Bank of Cicero, 957 F.2d 1362, 1366 (7th Cir.1992) (citation omitted). In conducting this review, we must keep in mind that
“[sjummary judgment should be granted only when no genuine issues of material fact exist and when the moving party is entitled to judgment as a matter of law____ This standard closely resembles that used for entry of a directed verdict, ... where the district court must direct a verdict if there can be only one reasonable decision made under the governing law.”
Id. (quoting Central States, Southeast, and Southwest Areas Pension Fund v. Jordan, 873 F.2d 149, 152 (7th Cir.1989)).
III. ANALYSIS
In Hammond v. Fidelity and Guaranty Life Insurance Company, 965 F.2d 428 (7th Cir.1992), we considered several of the issues raised by this appeal. Hammond centered around the interpretation of the term “totally disabled” in an employee benefit life insurance plan governed by ERISA. Id. at 428. In determining the law applicable to the dispute, we noted that
“ERISA explicitly states that it ‘shall supersede any and all State laws insofar as they may now or hereafter relate to any employee benefit plan.’ 29 U.S.C. § 1144(a). However, Congress also enacted an ‘insurance savings clause’ in § 1144(b)(2) which provides that a state law is not preempted if it regulates insurance; that is, if it ‘has the effect of transferring or spreading a policyholder’s risk, ... is an integral part of the policy relationship between the insurer and the insured, and ... the practice is limited to entities within the insurance industry.’ Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 48-49, 107 S.Ct. 1549, 1553-54 [95 L.Ed.2d 39] (1987)....”
Id. at 430. The question we faced in Hammond reappears in the instant case: whether “state laws governing insurance policy interpretation [are] preserved under ERISA.” Id. We decided that they were not so preserved because “state rules of contract interpretation” did not “regulate insurance” within the meaning of ERISA § 1144(b)(2) and because Congress intended “uniformity of decisions under ERISA.” Id. “We therefore concludefd] that ERISA preempts state decisional rules [of contract interpretation], and that any ambiguities in ERISA plans and insurance policies, should be resolved by referring to the federal common law rules of contract interpretation.” Id. See also Senkier v. Hartford Life & Accident Insurance Company, 948 F.2d 1050, 1051 (7th Cir.1991) (applying federal common law to the interpretation of an ERISA-governed group accident policy). In Hammond, we summarized the relevant rules of federal common law contract interpretation as follows:
*308“[W]e interpret the terms of the [ERISAgoverned] policy ‘in an ordinary and popular sense as would a [person] of average intelligence and experience.’ Evans v. Safeco Life Ins. Co., 916 F.2d 1437, 1441 (9th Cir.1990) (quoting Allstate Insurance Co. v. Ellison, 757 F.2d 1042, 1044 (9th Cir.1985)). Ambiguous terms in an insurance contract will be strictly construed in favor of the insured. Northbrook Excess & Surplus Ins. Co. v. Proctor & Gamble, 924 F.2d 633, 638 (7th Cir.1991). However, we will ‘not artificially create ambiguity where none exists.’ Evans, 916 F.2d at 1441.”
Hammond, 965 F.2d at 430.
With these interpretive principles in mind, we turn to the questions raised in this appeal. We agree with the district court that the term “mental illness” in the disputed ERISA plan is ambiguous. Lincoln acknowledges that the Plan provides no definition of the term mental illness, but nevertheless insists that the term plainly encompasses illnesses like the one afflicting James because “the average layperson, by merely observing James’ broad range of psychiatric and behavioral symptoms and the nature of his treatment, would conclude that he was suffering from and being treated for a mental illness or disorder”. Appellant’s Brief at 14. In Lincoln’s view, the cause of an illness is irrelevant in determining whether .an illness is physical or mental. This is in direct conflict with Phillips’ view that because James’ condition flows from an organic defect, he is suffering from a physical illness, with behavioral and emotional manifestations.
Lincoln presents several arguments in support of its view of the meaning of the Plan term “mental illness”, but we find none of them persuasive. Two of these arguments can be summarily dismissed. First, Lincoln appeals to a limited group of “expert” opinion which allegedly supports its view that the. Plan limitation applies to James’ condition. Lincoln points out that both Dr. Bateman and Dr. Bigler agreed that James suffers from a mental disorder. However, Dr. Bateman expressly stated that James suffers from an “organic mental disorder as one manifestation of his underlying neurological disease,” and not from a mental illness. He defined a mental illness as a “state of being psychotic or out of contact with reality” when there is no accepted organic basis for the condition. It is true that Dr. Bigler appeared to agree that organic brain syndrome is an example of a “mental disorder or mental illness” that is organically based, but, at best, that simply presents us with a semantic disagreement among experts as to the categorization and definition of organic brain syndrome. Similarly, the standard psychiatric reference texts cited by Lincoln which classify organic brain syndrome as a mental disorder are of no assistance in our effort to interpret the language and meaning of the Plan’s mental illness limitation.
Lincoln’s second argument is based on the language of the Plan limitation. Lincoln maintains that because the Plan limits benefits for mental illnesses “of any type or cause” the Plan expressly excludes the cause of an illness as a relevant factor in determining the applicability of the limitation to a given mental condition. Thus, it argues that the fact that James’ mental disorder has an organic cause cannot be a reason for excluding his condition from the mental illness limitation’s coverage. As Phillips points out, this textual interpretation begs the question, because Phillips’ position is that James does not suffer from a mental illness and therefore the limitation, no matter how it is modified by the “any type or cause” language, is inapplicable to his condition.
Having rejected as unconvincing Lincoln’s “expert” and “textual” arguments, we turn now to the crux of Lincoln’s position: its contention that a mental illness ought to be defined as a condition whose primary observable symptoms. are behavioral. This view was adopted by a panel of the Eighth Circuit in Brewer v. Lincoln National Life Insurance Company, 921 F.2d 150 (8th Cir.1990), cert. denied, — U.S. —, 111 S.Ct. 2872, 115 L.Ed.2d 1038 (1991). Brewer presented facts similar to the instant case: a father, as guardian for his son, sued an insurance company for its refusal to reimburse the full costs of his *309son’s treatment for affective mood disorder, a condition which had been successfully treated with a combination of drugs and psychotherapy. Id. at 152. The Brewer court, citing ERISA’s directive, contained in 29 U.S.C. § 1022(a)(1), that summary plan descriptions “be written in a manner calculated to be understood by the average plan participant,” stated that terms in insurance policies governed by ERISA “should be accorded their ordinary, and not specialized, meanings.” Id. at 154. The court then applied that interpretive rule to the case at hand:
“[t]he cause of a disease is a judgment for experts, while laymen know and understand symptoms. Laymen undoubtedly are aware that some mental illnesses are organically caused while others are not; however, they do not classify illnesses based on their origins. Instead, laypersons are inclined to focus on the symptoms of an illness; illnesses whose primary symptoms are depression, mood swings and unusual behavior are commonly characterized as mental illnesses regardless of their cause.
“Neither policy in this case limited the definition of ‘mental illness’ to only those illnesses that have a non-organic origin, and the district court should not have adopted a definition that effectively imposed such a limitation. By focusing upon the disease’s etiology, the district court considered factors that are important to experts but not to laypersons. The court thus failed to examine the term ‘mental illness’ as a layperson would have, which is the examination we concluded ERISA and federal common law require.”
Id.
Lincoln also cites in support of its proposed definition of mental illness Blake v. Unionmutual v. Stock Life Insurance Company of America, 906 F.2d 1525 (11th Cir.1990). We do not believe that Blake helps Lincoln’s cause. In Blake, a plaintiff who was hospitalized for depression shortly after she gave birth to a baby girl claimed she was entitled to compensation under a group health insurance plan governed by ERISA. An Eleventh Circuit panel adopted the district court’s order finding that the mental illness limitation in the insurance plan applied to the plaintiff’s benefit claim. Id. at 1527. The plaintiff in Blake had argued that she was “suffering from physical illness with psychiatric manifestations” and thus the mental illness limitation did not apply to her. Id. at 1530. However, the district court concluded that because of the plaintiffs failure “to prove an organic causation for [her] illness, we find that the treatment [plaintiff] received is only more convincing proof that she suffered a mental illness within the terms of the policy.” Id. The court then went on to describe how the “plaintiff was treated primarily by psychiatrists [and received] well recognized psychiatric treatment____” Id. We do not believe that Blake supports the position taken by Lincoln. The crucial distinction between Blake and the instant case is that here both parties acknowledge that James’ illness has an organic cause. Moreover, our reading of Blake leads us to conclude that had the plaintiff demonstrated an organic basis for her illness, the Eleventh Circuit may well have held that the policy’s mental illness limitation did not apply. We base this conclusion on the district court opinion in Blake, adopted by the Eleventh Circuit panel, which stated that “[t]he factual issue [presented by the case] is whether the post-partum depression that Pam Blake suffered after giving birth was a mental illness or a physically based illness with psychiatric manifestations.” Id. at 1527. By framing the issue this way, the district court implied that if it had found that the plaintiff’s illness had a physical cause, the plan’s mental illness limitation would have been inapplicable to her condition.
The Brewer court, however, did embrace Lincoln’s preferred definition of mental illness, and the appellant’s main argument is that we, like the Brewer court, ought to conclude that the term “mental illness” can only be understood as a description of the symptoms of a condition, and that a condition’s cause is irrelevant to its classification. See also Equitable Life Assurance Society v. Berry, 212 Cal.App.3d 832, 260 *310Cal.Rptr. 819, 824 (1989) (manifestation (i.e., symptoms) of a condition, not the cause of the condition, is the appropriate “yardstick” to use in determining whether a condition is a mental illness).
Countering this argument, Phillips correctly points out that no Plan provision provides that the symptoms of an illness controls whether it is subject to the mental illness limitation. In fact, Phillips points to the provision of the Plan which provides that “[a]ll illnesses due to the same cause, or to a related cause, will be deemed one illness,” as evidence that, under the Plan, an illness’ cause is the determinative factor in its classification. Phillips employs the following hypothetical to illustrate its point: a car accident victim who suffers physical injuries, including brain damage which produces behavioral disorders, must be considered, under the “one illness” provision, to be suffering from one illness because the car accident was the single cause of all his ailments. However, Phillips argues that under Lincoln’s approach to defining mental illness, the mental illness limitation would apply to the treatment of the victim’s brain injury because the behavioral manifestations of the brain injury would appear to Lincoln’s hypothetical layperson to be evidence of a “mental illness”. Although all the injuries had a common cause (the car accident), the brain injury would be separated out as a “mental illness” and treated differently than the other injuries. Phillips argues that this result, required under Lincoln’s approach to defining mental illness, would contravene the Plan’s express mandate that “[a]ll illnesses due to the same cause ... be deemed one illness.” Thus, Phillips concludes, the Plan itself implies that a condition’s cause is an important element in the classification of mental illnesses.
Phillips also contends that the Plan’s specific policy limitations applicable to mental illnesses — a $20.00 cap on each visit, only one visit per day and only 50 visits during a calendar year — suggest a situation in which a person is seeking periodic counseling from a psychologist, psychiatrist or other professional on an out-patient basis for “problems caused by environmental [social or human] factors, such as death, divorce or any other traumatic experience.” Ap-pellee’s brief at 20. Phillips maintains that the description of the treatment contemplated by the Plan for mental illness is inapplicable to a condition, like James’, in which an organic defect creates behavioral problems so severe that the victim is unable to function in society. Because the treatment contemplated by the Plan for mental illness does not correspond to a condition such as James’, Phillips concludes that the language of the Plan itself suggests that James’ condition is not covered by the mental illness limitation.
Moreover, like Lincoln, Phillips is able to point to a federal court of appeals decision approving its preferred definition of “mental illness.” A panel of the Ninth Circuit, in Kunin v. Benefit Trust Life Insurance Company, 910 F.2d 534 (9th Cir.), cert. denied, — U.S. —, 111 S.Ct. 581, 112 L.Ed.2d 587 (1990), upheld a finding of the district court that autism was not a mental illness under an ERISA employee welfare benefit plan because “mental illness refers to a behavioral disturbance with no demonstrable organic or physical basis____ [l]t stems from reaction to environmental conditions as distinguished from organic causes.” Id. at 538 (citation to district court opinion omitted). Since autism has an organic cause, it did not qualify as a mental illness. Id. See also Malerbi v. Central Reserve Life Insurance Company, 225 Neb. 543, 407 N.W.2d 157 (1987) (affirming trial court’s finding that a mental illness limitation did not apply to a person whose organic brain defect caused behavioral abnormalities); Arkansas Blue Cross & Blue Shield, Inc. v. Doe, 22 Ark. App. 89, 733 S.W.2d 429 (1987) (because the cause of manic-depressive disorder was organic, the disorder was held not to be subject to an insurance policy’s mental illness limitation).
Faced with these competing definitions of “mental illness” which have divided not only the litigants but also federal and state courts, we have no trouble agreeing with the district court’s finding that the term “mental illness” as used in the Plan is *311ambiguous. As in Kunin, where the Ninth Circuit found the term “mental illness” in a welfare benefit plan ambiguous, here the Plan “contains no definition or explanation of the term ‘mental illness,’ and offers no illustration of the conditions that are included or excluded. Nor does the policy contain any language suggesting whether the cause or the manifestation [of an illness] determines whether an illness is cov-ered____” 910 F.2d at 541. We thus hold that the Plan term “mental illness” is ambiguous as applied to individuals like James who have mental disorders caused by organic illnesses.
Following Hammond, which states that federal common law rules of contract interpretation require us to construe “[a]mbiguous terms in an insurance contract ... strictly ... in favor of the insured,” 965 F.2d at 430, we hold that the district court correctly ruled that the Plan’s mental illness limitation does not apply to James’ condition. The adoption by the district court of the state law rule of contract interpretation contra proferentem as part of the federal common law of ERISA was proper. In Fox Valley & Vicinity Construction Workers Pension Fund v. Brown, 897 F.2d 275 (7th Cir.) (en banc), cert. denied, — U.S. —, 111 S.Ct. 67, 112 L.Ed.2d 41 (1990), we held that when “ERISA is silent on an issue, a federal court must fashion federal common law rules to govern ERISA suits____ In making such rules, we must of course look to the statute itself for guidance, ... and it is also proper to turn to state law when creating such rules, ... as long as such state law is consistent with the policies underlying the federal statute at issue____” Id. at 281 (citations omitted). Following the analysis laid out by Fox, and anticipating our holding in Hammond, the district court decided to borrow the state law contract doctrine of contra proferentem in fashioning the federal law governing this dispute.
Lincoln argues that the application of the rule of contra proferentem in the ERISA context was foreclosed by the Supreme Court in Firestone. We disagree. In Firestone, the Court considered the question of the appropriate standard of review for ERISA actions brought under 29 U.S.C. § 1132(a)(1)(B), challenging denials of benefits by ERISA plan administrators. 489 U.S. at 108, 109 S.Ct. at 952. The Court concluded that de novo review was required also making clear that principles of trust law are useful guides in the development of federal common law in ERISA cases. Id. at 108, 111, 109 S.Ct. at 952, 954. In its discussion of relevant trust law principles, the Court stated that as “they do with contractual provisions, courts construe terms in a trust agreement without deferring to either party’s interpretation.” Id. at 112, 109 S.Ct. at 955. The Court also noted that in actions challenging an employer’s denial of benefits before the enactment of ERISA “[i]f the plan did not give the employer or administrator discretionary or final authority to construe uncertain terms, the court reviewed the employee’s claim as it would have any other contract claim — by looking to the terms of the plan and other manifestations of the parties’ intent.” Id. at 112-13, 109 S.Ct. at 955 (citations omitted).
Lincoln argues that this language in Firestone precludes application of the rule of contra proferentem in ERISA actions because that rule is inconsistent with the Supreme Court’s instruction that the intent of the parties to an ERISA plan should be the focus of the inquiry. We disagree for the reasons articulated by the Kunin court in rejecting this reading of Firestone’s dicta:
“In [Firestone], ... the Supreme Court was discussing the standard according to which courts should review a plan administrator’s interpretation of plan provisions. The Court held that reviewing courts should interpret disputed provisions de novo, not defer to the administrator’s interpretation. The Court said nothing whatsoever about ordinary principles of construction according to which courts and administrators alike should arrive at their interpretations____ [A distinction exists] between (1) using certain presumptions, including presumptions in favor of a party, in order to arrive at the correct interpretation, and *312(2) deferring to a particular party’s interpretation. Rejecting (2) obviously does not undermine the soundness of (1). We might say, for example,- that a jury should not defer to a defendant’s plea of “not guilty” — we want the jury to determine his guilt or innocence de novo. However, this in no way implies that the jury, in making this determination, is not required to presume innocence — and indeed, to presume it quite strongly. Again, we might say that a baseball umpire should not defer to a baserunner’s claim that he is safe; but this proposition is fully consistent with the separate rule that ties go to the runner. Likewise, while we may not defer to an insured’s construction, we must not fail to indulge in a presumption that ambiguous language favors the insured.
“This common-sense view of the matter is supported by Firestone itself. The Court there noted, ‘The trust law de novo standard of review is consistent with the judicial interpretation of employee benefit plans prior to the enactment of ERISA[, when] [a]ctions challenging an employer’s denial of benefits [were] governed by principles of contract law.’ 489 U.S. at [112, 109 S.Ct. at 955.] ... It is, then quite plain that the Firestone Court intended no wholesale rejection of prevailing principles of plan interpretation when it looked to trust law on the subject of the appropriate standard of judicial review.”
Kunin, 910 F.2d at 541. But see Allen v. Adage, 967 F.2d 695, 701 (1st Cir.1992) (relying on Firestone to hold that, in most ERISA cases, contra proferentem contradicts the combined principles of the law of trusts and de novo review); Brewer, 921 F.2d at 153-54 (state law rule of contra proferentem is preempted by ERISA, 29 U.S.C. § 1144(a) and was rejected in ERISA context by Firestone); Finley v. Special Agents Mutual Benefit Association, Inc., 957 F.2d 617, 619-20 (8th Cir. 1992) (same). Cf. Delk v. Durham Life Insurance Company, 959 F.2d 104, 105-06 (8th Cir.1992) (contra proferentem is applicable in the ERISA context when an ambiguity cannot be resolved by interpreting the language as would an average plan participant).
The Kunin court also set out persuasive reasons for applying the rule of contra proferentem to cases such as the instant one:
“[T]he contra proferentem rules is followed in all fifty states and the District of Columbia, and with good reason. Insurance policies are almost always drafted by specialists employed by the insurer. In light of the drafters’ expertise and experience, the insurer should be expected to set forth any limitations on its liability clearly enough for a common layperson to understand; if it fails to do this, it should not be allowed to take advantage of the very ambiguities that it could have prevented with greater diligence. Moreover, once the policy language has been drafted, it is not usually subject to amendment by the insured, even if he sees an ambiguity; an insurer’s practice of forcing the insured to guess and hope regarding the scope of . coverage requires that any doubts be resolved in favor of the party who has been placed in such a predicament.”
Kunin, 910 F.2d at 540 (emphasis added). Cf. Eley v. Boeing Inc., 945 F.2d 276, 279-80 (9th Cir.1991) (contra proferentem does not apply to insurance contracts resulting from arms-length bargaining between parties of equal power). As the Second Circuit reasoned in applying the rule of contra proferentem to an ERISA-governed employee health insurance plan,
“application of this rule of interpretation to de novo review of ERISA insurance plans is an appropriate implementation of the congressional expectation that the courts will develop a ‘federal common law of rights and obligations under ERISA-regulated plans.’ ... To ignore the effect of this rule 'would require us to ... afford less protection to employees and their beneficiaries than they . enjoyed before ERISA was enacted, ’ a result that would be at odds with the congressional purposes of promoting the interests of employees and beneficiaries and protecting contractually *313defined benefits____ Moreover, ‘ERISA abounds with the language and terminology of trust law,’ ... and application of this rule to resolve ambiguities in an ERISA insurance plan on de novo review is consistent with the basic principle of trust law that trust property is to be dealt with for the benefit of the beneficiary.”
Masella v. Blue Cross & Blue Shield of Connecticut, Inc., 936 F.2d 98, 107 (2d Cir. 1991) (citation omitted) (emphasis added). “[Insurance policy exceptions to liability must be expressed in unequivocal language so that it is reasonable to assume the insured understood and accepted these limitations.” Heller v. Equitable Life Assurance Society, 833 F.2d 1253, 1256 (7th Cir.1987) (citation omitted) (applying Illinois law).
Lincoln mistakenly asserts that ERISA’s requirement that plan descriptions be “written in a manner calculated to be understood by the average plan participant,” 29 U.S.C. § 1022(a)(1), prevented it from providing any definition of mental illness. The requirement that a plan description be accessible to the average plan participant is no excuse for failing to supply any definition at all of the term “mental illness.” Lincoln’s citation to Harnischfeger Corporation v. Harbor Insurance Company, 927 F.2d 974 (7th Cir.), cert. denied, — U.S. —, 112 S.Ct. 189, 116 L.Ed.2d 150 (1991), is similarly unavailing. In Hamischfeger, we stated that the rule that ambiguities must be resolved against insurers can only be invoked when there exists a “genuine (meaning, substantial) uncertainty, not resolvable by other means, and the insured’s proposed reading must be reasonable____ Rules of interpretation are tie-breakers, even when the words being interpreted appear in insurance policies.” 927 F.2d at 976. Those are precisely the circumstances we confront here. Lincoln is not being forced to provide benefits “upon every remotely colorable claim of plan ambiguity,” Appellant’s Brief at 44, but upon an ambiguity in a central plan limitation. Lincoln urges us to reject the application of contra proferentem and simply determine the intent of the parties as to the mental illness limitation. This is precisely what we are unable to do because the Plan term “mental illness” is ambiguous. Moreover, contrary to Lincoln’s protestations, reading such ambiguities against the insurer is not in contravention of the underlying purpose of ERISA, which is “to promote the interests of employees and their beneficiaries in employee benefit plans ... and to protect contractually defined benefits.” Firestone, 489 U.S. at 113, 109 S.Ct. at 955. We serve that purpose today by holding that a plan participant cannot be denied benefits under a limitation which is ambiguous as to its application to his benefit claim.
Lincoln also cites Northbrook Excess & Surplus Insurance Company v. Proctor & Gamble Company, 924 F.2d 633 (7th Cir.1991), a case involving the application of Ohio law to the terms of a liability insurance policy, in support of its view that contra proferentem is inapplicable to the instant case. In Northbrook, we observed that
“[insurance contracts are construed strictly in favor of the insured because, usually, the insurer wrote the policy terms. As the leading treatise notes:
It is commonly stated that the reason for the rule of construction against the insurer is that policies of insurance are made on printed forms carefully prepared in the light of wide experience, by experts employed by the insurer, and in the preparation of which the insured has no voice ...
Additionally, these contracts are usually offered to the insured on a take-it-or-leave-it basis and as such are termed contracts of adhesion, justifying a construction against the insurer.
Couch on Insurance 2d 382-83, 387, § 15:78 (rev. ed. 1984) (emphasis supplied).”
Id. at 638. We also stated that “it is clear that the [contra proferentem] rule is grounded in the need to protect an insured from an insurer who has had exclusive control of the drafting process.” Id. We also "assume[d] that — like other jurisdictions — Ohio would not apply the contra *314proferentum principle to situations in which its underlying rationale is inapplicable.” Id. at 639. In Northbrook we held that the contra proferentem principle was inapplicable because “the record clearly establishes that [the insured] was a co-drafter of the ... policy and not simply a party given a take-it-or-leave-it option.” Id. Lincoln maintains that we should take a similar view here since “[e]xperience and common sense dictate that employers typically negotiate with insurers to obtain the types and amount of coverage desired for the benefit of their employees and select among plans offered by competing carriers.” Appellant’s Brief at 36. However, as Phillips points out, there is no record evidencé suggesting that the two parties negotiated the terms of the Plan. Lincoln had sufficient opportunity in the district court to present support for its contention that the Plan was negotiated, rather than drafted by Lincoln and purchased by Seed-boro. Lincoln presented no affidavits or other evidence along with its summary judgment motion to support such an inference. Even in its appellate brief, Lincoln merely states that “employers typically negotiate with insurers” in purchasing insurance plans without any reference to how this Plan was negotiated. This bald assertion is not sufficient to overturn the district court’s determination that the rule of contra proferentem is applicable to the instant case.3
In a final argument, Lincoln asks that if we find that the Plan term “mental illness” is ambiguous, we should remand for an evaluation of extrinsic evidence to allow for a determination of the parties’ intent as to the meaning of the term “mental illness”. Lincoln has already presented its arguments, both on appeal and in the district court, in support of its view of the meaning of the disputed term. Rehashing them in the district court would serve no purpose, and so we refuse to give Lincoln another kick at the cat. We are convinced that because the Plan provides no definition of the term “mental illness”, the term is inherently ambiguous and a remand for further development of the evidentiary record would be futile.
IV.
Insurance policies are almost always drafted by insurers, and they should be certain that limitations in their coverage are clear enough for a layperson to understand. Insurers should not be permitted to exploit policy term ambiguities, which they could have avoided, to deny coverage to an unsuspecting insured. See Kunin, 910 F.2d at 540. We hold that the Plan’s mental illness limitation does not apply to James’ condition, and therefore the district court’s entry of summary judgment in favor of Phillips is Affirmed. We Remand to the district court for a determination of the amount owed by Lincoln to Phillips pursuant to Lincoln’s obligations under the Plan. BAUER, Chief Judge, dissenting.
I respectfully dissent. The majority opinion does an outstanding job of covering the issues involved. What it comes down to, in my opinion, is the acceptance or rejection of the rationalization and imperatives set out in Brewer v. National Life Insurance Company, 921 F.2d 150 (8th Cir. 1990), cert. denied, — U.S. —, 111 S.Ct. 2872, 115 L.Ed.2d 1038 (1991). That case is well discussed in the majority opinion and I shall not waste any more forest land with further development; suffice to say that I agree that insurance coverage and exclusions must be “written in a manner calcu*315lated to be understood by the average plan participant.” And this policy clearly limits the coverage “for care of mental illness or care of nervous conditions of any type or cause____”
I believe that any fair reading of the problems facing James inescapably lead to the conclusion that he suffers from “a mental illness” or “nervous disorder.” The cause, as the policy clearly states, is not controlling. “Mixed organic brain syndrome” with the manifestations exhibited by James that require the present hospitalization and treatment is a “mental illness.”
I would reverse the judgment entered for the plaintiff.