POSNER, Circuit Judge.

Terry Manny, a 58-year-old truck driver who is a participant in an ERISA welfare plan, asked the trustees who administer the plan to cover the expense of a proposed gastric-bypass operation. Man-ny, who is 6 foot 1 inch tall and weighs 470 pounds, suffers from a variety of serious health conditions' undoubtedly caused or exacerbated by his obesity, including type 2 diabetes (his pancreas produces insulin, but not enough to eliminate excess sugar from his blood), high blood pressure, joint and respiratory problems, swelling of his legs and feet, lower back pain, and depression. The trustees ruled that the plan does not cover such an operation, precipitating this suit, which the district court dismissed on the ground that the trustees’ decision was not an unreasonable interpretation of the plan.

Because the plan confers on the trustees “discretionary and final authority in making ... decisions interpreting plan documents,” judicial review of their interpretations is deferential. Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989). And conflict of interest is not a concern with respect to the present plan, as it sometimes is when an insurance company is both plan administrator and insurer of benefits, or when the employer is the administrator of a self-funded single-employer plan. Pinto v. Reliance Standard Life Ins. Co., 214 F.3d 377 (3d Cir.2000); Kathryn J. Kennedy, “Judicial Standard of Review in ERISA Benefit Claim Cases,” 50 Am. U.L.Rev. 1083, 1146-53 (2001). The deferential standard that courts use in reviewing determinations made by trustees to whom the plan gives discretion to interpret — -the “arbitrary and capricious” standard, Firestone Tire & Rubber Co. v. Bruch, supra, 489 U.S. at 115, 109 S.Ct. 948; Dabertin v. HCR Manor Care, Inc., 373 F.3d 822, 827-28 (7th Cir.2004); Daill v. Sheet Metal Workers’ Local 73 Pension Fund, 100 F.3d 62, 67-68 (7th Cir.1996)— is “a range, not a point.” Van Boxel v. Journal Co. Employees’ Pension Trust, *243 836 F.2d 1048, 1052-53 (7th Cir.1988). It is “a sliding scale” that requires that judicial review be “more penetrating the greater is the suspicion of partiality, less penetrating the smaller that suspicion is.” Id. A conflict of interest on the part of the plan’s trustees may start a slide. Chojnacki v. Georgia-Pacific Corp., 108 F.3d 810, 815 (7th Cir.1997).

But not in this case. The teamsters plan is a multi-employer welfare plan the trustees of which are required to consist of an equal number of union and employer representatives. 29 U.S.C. § 186(c)(5)(B); Connolly v. Pension Benefit Guaranty Corp., 475 U.S. 211, 232, 106 S.Ct. 1018, 89 L.Ed.2d 166 (1986). The union trustees, at least, have no discernible incentive to rule against an applicant. And the trustees were unanimous in turning down Manny’s application. Given the language of the plan conferring interpretive authority on the trustees and the absence of a conflict of interest, the only question for us is whether the trustees’ interpretation of the plan was completely unreasonable.

The plan defines “cosmetic” as “care, treatment, services or supplies the primary effect of which is to improve the physical appearance.... The fact that there may be an incidental medical benefit does not prevent a determination that the care, treatment, services or supplies are cosmetic.” With regard to such care, etc., the plan goes on to provide as follows:

4.08 EXCLUSION FOR PAYMENT FOR TREATMENT CONNECTED WITH SURGERY FOR COSMETIC PURPOSES

A Covered Individual shall not be entitled to payment on a claim for benefits for any charge incurred for treatment or service connected with a cosmetic procedure, even if performed for psychological reasons, unless the treatment or service is medically required as a result of an Accidental Bodily Injury incurred while a Covered Individual.

This exclusion includes, but is not limited to:

(a) Any surgery primarily for obesity, including gastric bypass, gastric stapling, intestinal bypass, lipectomy, suction lipectomy, panniculectomy, and any other surgical procedure, a purpose and result of which is primarily to remove adipose tissue;

(b) Augmentation mammoplasty, unless part of reconstructive surgery for the treatment of malignancy of the breast necessitating removal of a portion or all of the breast, tissue;

(c) Rhinoplasty, unless the patient has sustained a traumatic fracture of the nasal septum, or unless the patient has chronic nasal obstruction and the procedure is undertaken to relieve this obstruction;

(d) Otoplasty for irregular deformity or macrotia. This is sometimes referred to as plastic surgery for lop ears or cauliflower ears;

(e) Blepharoplasty, or repair of drooping eyelids, unless the droop of the eyelids is such as to restrict the field of vision and the visual field restriction is documented by the ophthalmological consultant;

(f) Radical Keratectomy or Keratotomy, unless the patient has myopia of such a severe degree that it cannot be corrected by lenses;

(g) Rhytideetomy (face lift);

(h) Dyschromia (tattoo removal); and

(i) Genioplasty (chin augmentation).

How all this bears on Manny’s claim is unclear. Section 4.08 begins by excluding benefits for “treatment or service connected with a cosmetic procedure,” and if that were all the plan said, Manny might be *244 home free. While some people may undergo the dangerous and painful procedure of a gastric-bypass or gastric-stapling operation merely to look better, Manny is not one of them. He wants the operation for health reasons. The “medical benefit” that he seeks is not “incidental.” And the “primary effect” of the operation would not be “to improve [his] physical appearance”; that would be an incidental effect. Of course, should the operation cause Manny to lose weight yet not yield any medical benefit, then its primary and indeed only effect would be to improve his appearance. But probably “primary effect” means primary intended effect; this is implied, as we’ll see, by the exceptions to the exclusion discussed next.

The second paragraph of section 4.08 explains that the exclusion of cosmetic procedures from coverage extends to “any surgery primarily for obesity, including gastric bypass, ... and any other surgical procedure, a purpose and result of which is primarily to remove adipose [i.e., fatty] tissue.” A gastric-bypass operation is “surgery primarily for obesity.” In fact, the word “primarily” could be deleted without loss of meaning. The sole purpose and only effect of the operation are to shrink the patient’s fatty tissue by reducing the ability of his gastrointestinal system to convert food to fatty tissue. The shrinkage is valued not in itself, however, but only as a means to a further end. As is true of all surgery. The purpose of an operation to remove a tattoo is to remove the tattoo, but the usual reason for wanting to remove it — the end to which removal is the means — is not love of surgery but that the tattoo is unsightly or embarrassing.

So while the first paragraph of section 4.08 leans in favor of coverage of gastric-bypass operations motivated solely by concern with health, the second paragraph appears to contain a categorical exclusion of such operations. When different clauses of a contract clash, creating an ambiguity, we have an interpretive task confided in this instance to the teamster plan’s trustees. We cannot say that their interpretation is unreasonable, especially in light of the treatment in the second paragraph of section 4.08 of other procedures that are excluded from coverage as being “cosmetic.” Importantly, some of these exclusions are qualified and others not. Breast augmentation is excluded “unless” it is motivated by loss of breast tissue as a result of breast cancer. Operations to reshape the eyeball are excluded “unless” they are necessary to correct myopia that is so severe that it can’t be corrected by glasses. (There are similar “unless” exceptions for “nose jobs” and repairs of drooping eyelids.) Most breast-augmentation operations and most operations to reshape the eyeballs are driven by the desire to improve one’s normal appearance. But some are necessary to avoid disfigurement or to enable a person to see, and these are excepted from the exclusion even if, to return to an earlier point, the operation fails to achieve its medical purpose and thus yields merely a cosmetic benefit. Several other procedures listed in the exclusion, however, are excluded altogether, without exception. Tattoo removal is not covered even if the removal is necessary to prevent infection. Similarly, surgery to reduce obesity is excluded whether driven by vanity or health concerns.

Granted, there is a difference between breast augmentation and eyeball-reshaping, on the one hand, and gastric bypass or gastric stapling on the other hand. Most of the operations of the former two types (and likewise most “nose jobs” and eyelid operations) are motivated by cosmetic considerations as defined in the plan. Indeed, all are so motivated in the case of breast augmentation. (Even when the augmenta *245 tion is designed to reconstruct a breast amputated in a mastectomy, the objective is to alter the patient’s physical appearance rather than to ward off a further illness — but it is to restore rather than to improve normal appearance and this limits potential abuses of coverage.) In contrast, most obesity surgery is motivated exclusively by health concerns. These operations are dangerous and are rarely performed other than on people like Manny who are morbidly obese, American Gas-troenterological Association, “Medical Position Statement on Obesity,” 123 Gastroenterology 879, 881 (2002), “morbid” here denoting a degree of obesity that is highly dangerous to the individual’s health.

Probably the operations are nonetheless excluded from the plan’s coverage not only because they are expensive (and welfare funds have limited assets) but also because, while they have their strong defenders, see, e.g., Henry Buchenwald et al., “Bariatric Surgery: A Systematic Review and Meta-Analysis,” 292 JAMA 1724 (2004); Robert E. Brolin, “Bariatric Surgery and Long-Term Control of Morbid Obesity,” 288 JAMA 2793 (2002); Alan C. Wittgrove & G. Wesley Clark, “Laparo-scopic Gastric Bypass, Roux en-Y — 500 Patients: Technique and Results, with 3-60 Month Follow-up,” 10 Obesity Surg. 233 (2000), they are controversial. Many medical experts are of two minds on whether surgery to reduce obesity produces health benefits commensurate with the danger of such surgery. Mike Mitka, “Surgery for Obesity: Demand Soars Amid Scientific, Ethical Questions,” 289 JAMA 1761, 1762 (2003); C. David Sjostrom et al, “Differentiated Long-Term Effects of Intentional Weight Loss on Diabetes and Hypertension,” 36 Hypertension 20 (2000); Gina Kolata, “Health and Money Issues Arise Over Who Pays for Weight Loss,” N.Y. Times (late ed.), Sept. 30, 2004, p. Al. A recent study found that the surgery kills 2 percent of extremely obese patients, such as Manny, on whom it is performed. David R. Flum & E. Patchen Dellinger, “Impact of Gastric Bypass Operation on Survival: A Population-Based Analysis,” 199 J. Am. College of Surgeons 543, 547 (2004).

Obviously we’re not going to try to referee the debate. The point is only that it would be no surprise if section 4.08 had indeed been intended to exclude all obesity surgery from coverage, as the trustees found. But there is one thing to give us pause, and that is their submission of Man-ny’s medical records to the plan’s medical consultant, Dr. William B. Buckingham. In Exbom v. Central States, Southeast & Southwest Areas Health & Welfare Fund, 900 F.2d 1138 (7th Cir.1990), a case that involved another morbidly obese teamster wanting obesity surgery, and the same plan, Buckingham assumed that gastric-stapling operations were covered by the plan if medically necessary; he just believed that Exbom did not have a medical need for the operation. Id. at 1140. There are a number of similar cases. See Roberts v. Central States, Southeast & Southwest Areas Health & Welfare Fund Plan 503, No. 97-1454-CV-W-5 (W.D.Mo. Sept. 15, 1998); Livingston v. Central States, Southeast & Southwest Areas Health & Welfare Fund, 900 F.Supp. 108 (E.D.Mich.1995); Strader v. Central States, Southeast & Southwest Areas Health & Welfare Fund, No. 86-3340-CV-S-4 (W.D.Mo. May 4, 1987); Berry v. Central States Southeast & Southwest Areas Health & Welfare Fund, No. Civ-1-86-004 (E.D.Tenn. Jul. 22,1986). Yet in the present case Buckingham changed his tune and testified that gastric-bypass operations, and presumably gastric-stapling operations as well (for that was the operation sought by Exbom, and it is treated the same in the plan), are ahuays within the plan’s *246 exclusion for surgery to correct obesity. Buckingham “states that gastric bypass ... is specifically excluded from coverage in the Plan document. He explains that our experience is that this procedure does not constitute treatment of comobilities [sic — ‘comorbidities’] such as sleep apnea, diabetes, hypertension and cardiac disease. In addition, the initial weight loss usually is regained after this procedure. He feels that it is NOT physically possible to gain or maintain weight without excessive eating and this procedure is not medically necessary.”

There has been, however, a change in the plan language between Exborn (and the other cases we cited) and the present case. Under the earlier language a procedure was “cosmetic” only if it did not “correct or materially relieve a medical condition.” This definition has been broadened; as we know, the plan now defines “cosmetic” as “care, treatment, services or supplies the primary effect of which is to improve physical appearance .... The fact that there may be an incidental medical benefit does not prevent a determination that the care, treatment, services or supplies are cosmetic.” The tension between the definition of “cosmetic” and the specific exclusion of obesity surgery was greater under the old language, and this may explain the different interpretations by Buckingham, endorsed by the trustees. Moreover, Exbom was decided a decade and a half ago, and what may have seemed a promising procedure then may not seem so now in the light of the longer experience with it.

Admittedly it is odd for a doctor to be testifying about the meaning of a medical plan. With rare exceptions the interpretation of a document is a task that a judge or other resolver of legal disputes performs without the aid of testimony, expert or otherwise. In some state courts, however, legislators testify to the legislative history of a statute they voted for, Friends of Mammoth v. Board of Supervisors, 8 Cal.3d 247, 104 Cal.Rptr. 761, 502 P.2d 1049, 1055-56 (1972); Stewart v. Board of Medical Quality Assurance, 80 Cal.App.3d 172, 143 Cal.Rptr. 641, 646-47 (1978), and perhaps Buckingham’s testimony should be viewed in that light. See also Indiana Aeronautics Comm’n v. Ambassadair, Inc., 267 Ind. 137, 368 N.E.2d 1340, 1344 (1977). Trustees are not bound by the rules of evidence. We pointed out, moreover, that one thing that would make it plausible to think that the draftsmen of the teamsters plan had excluded surgery for obesity without exception would be skepticism in the medical profession about the medical value of such surgery, a skepticism that might have moved Buckingham to suggest the exclusion to the drafters of the plan. Or perhaps the purpose of Buckingham’s testimony was to provide a fallback position for the plan: that even if a gastric-bypass operation is sometimes covered'by the plan, it shouldn’t be in the present case.

Manny’s case is an appealing one, but there is no doubt that the trustees were acting reasonably in interpreting the plan to exclude coverage for gastric-bypass operations. (There is also no doubt that the trustees based their decision on an interpretation of the plan as excluding coverage for all such operations, rather than on a judgment that Manny’s motivation was cosmetic rather than health.) Indeed, for all we know, they were acting in Manny’s own best interests. The surgery that he wants is dangerous, and of uncertain value. His medical problems, though undoubtedly aggravated by his obesity, are, most of them anyway, treatable without surgery. But we are wandering from the issue, which is simply the reasonableness of the *247 trustees’ interpretation of the plan; it was reasonable, so our hands are tied.

AFFIRMED.