A coal company appeals from an order directing it to pay disability benefits to a miner, James Franklin, under the Black Lung Benefits Act, 30 U.S.C. §§ 901 et seq. Franklin's claim had at first been turned down by the agency and he had not pursued his administrative remedies; but within a year he requested that the proceeding be reopened, and eventually his request was granted and benefits awarded. The coal company challenges both the propriety of reopening the claim and the miner's substantive entitlement to benefits. As so often in black lung cases, the processing of the claim has been protracted scandalously. The original claim was filed in 1978 and the request for reopening in 1980--more than a decade ago. A hearing was not held until 1987, and appellate review within the Department of Labor consumed another three years. Such delay is not easy to understand. These are not big or complex cases (this one certainly is not). The monetary stakes are modest though not trivial, the medical reports and other documentary evidence (mainly depositions of physicians) are usually not voluminous, and the typical hearing lasts, we are told, no more than an hour. And there is not that extra layer of judicial review between us and the Benefits Review Board that there is in social security disability cases, where the appeal from the Appeals Council (the counterpart to the Benefits Review Board) goes to the district court first. The delay in processing these claims is especially regrettable because most black lung claimants are middle-aged or elderly and in poor health, and therefore quite likely to die before receiving benefits if their cases are spun out for years. We hope that Congress will consider streamlining the adjudication of disability benefits cases (not limited to black lung) along the lines suggested by the Federal Courts Study Committee. See the Committee's Report (April 2, 1990), at pp. 55-58.

A denial of black lung benefits can be reopened within one year upon a showing of a change in the miner's condition or a mistake in a determination of fact. Section 22 of the Longshore and Harbor Workers' Compensation Act so provides, 33 U.S.C. § 922, and this section is incorporated by reference in the Black Lung Benefits Act, 30 U.S.C. § 932(a), and repeated in a regulation under the latter Act, 20 C.F.R. § 725.310. (See Eifler v. Office of Workers' Compensation Programs, 926 F.2d 663, 665 (7th Cir.1991), on this genealogy.) When Mr. Franklin first applied for black lung benefits he submitted no medical evidence that he had black lung disease and his application was therefore turned down with the notation that he did not have the disease. When less than a year later he applied to reopen the proceeding, he attached a letter from his personal physician, Dr. M.E. Tomak. The letter explained that Tomak had been Franklin's physician for more than thirty years, had last examined him twelve days earlier and observed a variety of pulmonary symptoms, and had concluded that Franklin was disabled from gainful work in the mines as a result of his "many years of exposure to the environment and atmosphere of the coal ming [sic ] industry." The administrative law judge ruled that the letter established a change of condition and so warranted a reopening of the proceeding. It established nothing of the kind. Nothing in Dr. Tomak's letter indicated a deterioration in Franklin's condition since his first application. Granted, the deterioration in the miner's condition need not be great to allow a reopening within one year. We may assume, indeed, that it need only be perceptible, unlike the case where the request to reopen comes more than a year after the denial of benefits: for then the regulations require a material change in conditions, 20 C.F.R. § 725.309(d), and we have interpreted this to mean that between the time of the initial denial and the time of the motion the miner went from not being disabled to being disabled. Sahara Coal Co. v. Office of Workers' Compensation Programs, 946 F.2d 554, 556 (7th Cir.1991). (The difference in standards reflects the fact that the interest in the finality of a judgment--the interest that underlies statutes of limitations, the doctrine of res judicata, and cognate policies--varies with the amount of time that has elapsed between the denial of benefits and the request to reopen.) But however slight the required showing of deterioration when the request to reopen is made within a year of the denial of benefits, Franklin failed to make it. Nor can the requirement of a showing of deterioration be waved aside with the observation true though it is, that black lung disease is progressive, so that there must be deterioration between the denial and the motion to reopen, however brief the interval is. That would rewrite the regulation to allow reopening at will.

Although there was no demonstrated change in this miner's condition, there was arguably a mistake of fact--a mistake as to whether he had black lung disease--and no more is required to reopen the proceeding within a year of denial. We say "arguably" in order to hold separate the question of mistake from the ultimate question of entitlement to benefits. Suppose, as actually happened here, that within one year of the original denial of benefits on the ground that the claimant does not have black lung disease, he introduces evidence (the letter from Dr. Tomak) indicating that he may well have had the disease, after all, on the date of the original application. Years pass, during which additional medical evidence accumulates, and finally a hearing is held. Logically there are two issues for the hearing: First, did the evidence submitted within a year of the original denial demonstrate that the claimant had black lung disease all along? If not, the proceeding cannot be reopened, and the initial denial of benefits is final. Second, does all the evidence demonstrate that the claimant has the disease and has satisfied the other requirements for an award of benefits (notably that he is totally disabled as a result of the disease)?

It may be logical to proceed in this way, but it is not sensible. It requires that the claimant within one year of the original denial show not only that the denial may have been mistaken, but that it was mistaken, and if he has failed to get all his ducks in order, no evidence obtained later can save him. This is not how the reconsideration process normally works, and we can see no reason why it should be made to work abnormally in black lung cases. Normally when a losing litigant moves for reconsideration, he presents only enough evidence to show that reconsideration is warranted; he is not expected to prove at that point that the judgment he wishes to see set aside was mistaken. Estate of Kraus v. Commissioner, 875 F.2d 597, 602 (7th Cir.1989) ("the evidence is such that a new trial would probably produce a new result") (emphasis added); Mumford v. Bowen, 814 F.2d 328, 330 (7th Cir.1986) ("the new evidence is likely to change the outcome") (emphasis added); McKnight v. United States Steel Corp., 726 F.2d 333, 336 (7th Cir.1984). Definitive proof is for the hearing. It is enough if the movant's submission establishes a high enough probability of error to warrant a hearing--at which both parties will be allowed to make full evidentiary presentations. Dr. Tomak's letter was, as we shall see, far from conclusive on whether Franklin had black lung disease, but it was sufficiently probative to justify reopening the case.

If this is right it might seem merely a harmless error that the administrative law judge granted the motion to reopen on the erroneous ground that the miner had shown a change of condition. It is not harmless. Motions to reopen or reconsider are appeals to the discretion of the judicial or other officer to whom the motion is made. INS v. Abudu, 485 U.S. 94, 105-07, 108 S. Ct. 904, 911-13, 99 L. Ed. 2d 90 (1988); In re Shondel, 950 F.2d 1301, 1304 (7th Cir.1991); May Department Stores Co. v. NLRB, 897 F.2d 221, 230 (7th Cir.1990); Mumford v. Bowen, supra, 814 F.2d at 329. So the Supreme Court expressly held with reference to 33 U.S.C. § 922--the basis, as we noted, for the statute and regulation governing reopenings of black lung cases within one year--in O'Keeffe v. Aerojet-General Shipyards, Inc., 404 U.S. 254, 92 S. Ct. 405, 30 L. Ed. 2d 424 (1971) (per curiam). A reviewing court cannot exercise the discretion of the responsible officer--it's his discretion to exercise--save in the unusual case where there is only one decision he could make that would be rational; and that is not a case where a discretionary judgment is permitted.

We said that a mistake about whether a miner had black lung disease at the time of the initial denial of his application for benefits is a mistake of "fact," but this could be questioned. As the Black Lung Benefits Act and the regulations promulgated under it are drafted, a finding of black lung disease is generated by a complex system of presumptions; it is not a straightforward medical judgment. It might better be classified as a "mixed question of fact and law" or an "ultimate fact" rather than as a plain-vanilla fact. But for many legal purposes (appellate review, for example) such facts, though they are not facts in the lay sense but are instead the outcome of applying legal standards to facts, are treated like facts, Mucha v. King, 792 F.2d 602, 605 (7th Cir.1986); United States v. McKinney, 919 F.2d 405, 419 (7th Cir.1991) (concurring opinion), and one of these purposes we think should be that of deciding whether a black lung case can be reopened. This conclusion is consistent with the Supreme Court's holding in O'Keeffe v. Aerojet-General Shipyards, Inc., supra, 404 U.S. at 256, 92 S. Ct. at 407, that the reopening provision is to be interpreted generously to the claimant, and with the basic logic of such a provision. The deadlines for appeals are short, and one reason is that an appeal is based on the record before the tribunal whose decision is being appealed, so that no additional fact-gathering is necessary for the appeal. Deadlines for reopening are longer, because the reason for allowing a case to be reopened after the decision in it has become final is that something new has come to light that wasn't in the original record. A change in condition is a dramatic example of something new, but another example is new evidence (though O'Keeffe holds that this is not essential to a reopening based on a mistake of fact) showing that the original decision was mistaken. Dr. Tomak's letter was new evidence, albeit evidence about the accuracy of the original determination of an ultimate rather than elementary fact. Such evidence, which could not be presented in an appeal, should be allowed to be considered as a ground for reopening the case.

Although the case must be remanded for a decision by the administrative law judge whether to reopen the case on the basis of a mistake of fact rather than a change in condition, we must not stop with this conclusion. Should the administrative law judge decide to reopen the case once again, undoubtedly he will reinstate his opinion granting benefits, the coal company will again appeal though this time only on the merits, and we will reverse on the merits--for the administrative law judge's opinion on the merits was unreasoned. It would be pointlessly coy, and would prolong this already unconscionably protracted proceeding unduly, to withhold our view of the merits and thereby invite the administrative law judge to reinstate a doomed opinion.

On the basis of pulmonary-function studies conducted after the motion to reconsider was filed, the administrative law judge invoked a presumption that Franklin had black lung disease, 20 C.F.R. § 727.203(a)(2), a presumption which however the coal company could rebut by proving that (for example) his pulmonary problems were unrelated to exposure to coal dust. 20 C.F.R. § 727.203(b)(3). To this end the company introduced two reports by, and the lengthy deposition of, Dr. J. Timothy Deppe, a pulmonary specialist who has conducted many examinations of black lung claimants at the request of the Department of Labor, although in this case he was retained by the coal company. Dr. Deppe examined Franklin and elicited from him the information that Franklin had smoked a pack of cigarettes a day or more for more than 40 years. Deppe concluded from the examination that Franklin had a classic case of emphysema induced by cigarette smoking and no characteristic symptoms of black lung disease (even generously defined, as the Black Lung Benefits Act has been interpreted to define it, 20 C.F.R. § 727.202; Freeman United Coal Co. v. Office of Workers' Compensation Programs, 957 F.2d 302, 304 (7th Cir.1992); Peabody Coal Co. v. Lowis, 708 F.2d 266, 268 n. 4 (7th Cir.1983), as any lung disease caused in whole or part by exposure to coal dust), except that Franklin's exposure to coal dust may, Deppe opined, have contributed to his cough. At his deposition Deppe pointed out that Franklin had worked in surface rather than underground mines and had therefore been exposed to less coal dust than if he had worked underground, which in Deppe's opinion made it even less likely that Franklin's lung problems were related to coal dust.

In evaluating the record, the administrative law judge pounced on Deppe's comment about exposure to coal dust. Because Deppe had failed to "state how much less dust Claimant would have been exposed to [as a surface rather than underground miner] and did not equate his years as a miner at surface mines to a number of years of underground dust exposure," the administrative law judge concluded that Deppe's "opinion on the issue of causation is deserving of minimal weight." Not so Dr. Tomak's letter. Because Dr. Tomak had "treated and observed the miner's condition for over thirty years, I find that Dr. Tomak's opinion is deserving of great weight." The conclusion was that the presumption of black lung disease had not been rebutted.

The administrative law judge's treatment of the two doctors borders on the irrational. The opinion of Dr. Tomak, who so far as appears has no more than a general practitioner's knowledge of pulmonary diseases, who conducted no x-ray, arterial gas, or pulmonary function tests on Franklin, who was not deposed, and whose one-page letter does not even mention Franklin's long history of heavy cigarette smoking, is given great weight, while the longer reports, and full deposition, of a pulmonary specialist retained in other black lung cases by the Department of Labor itself, who studied the medical tests that had been made of the miner's lungs and who properly considered the bearing of the miner's heavy smoking, are nitpicked apart because the specialist failed to quantify the miner's coal dust exposure--something never done in these cases.

A possible explanation for this difference of treatment is the traditional but unwarranted preference of some administrative law judges in disability cases--a preference that has persisted in the face of such decisions as Richardson v. Perales, 402 U.S. 389, 91 S. Ct. 1420, 28 L. Ed. 2d 842 (1971); Garrison v. Heckler, 765 F.2d 710, 713-15 (7th Cir.1985), and DeFrancesco v. Bowen, 867 F.2d 1040, 1043 (7th Cir.1989)--for the opinion of a treating physician over that of a nontreating specialist regardless of circumstances that make the specialist's opinion more probative. Especially where, as in this case, the specialist both examines the claimant and is on the Department's roster of consultants in black lung cases, his opinion deserves more weight than that of a nonspecialist treating physician who knows so little about pulmonary diseases that he does not even see the relevance of a man's smoking habits to the causality of his pulmonary condition. Id. at 714; Peabody Coal Co. v. Helms, 901 F.2d 571, 574 (7th Cir.1990). It is no good saying that Dr. Tomak, having treated Franklin for more than thirty years, must have known how much and how long he had smoked. He may or may not have known, and whether he did or not nothing in his report indicates the slightest awareness of the significance of a history of heavy smoking; as there is no indication that Dr. Tomak is a pulmonary specialist, such awareness cannot be assumed.

Black lung disease, at least when broadly defined, is a progressive disease, so the opinion of a physician who, like Tomak, had treated the miner over a period of many years might on that ground deserve some weight in comparison with the report of a specialist who had seen the miner only once. The treating physician might learn something from being able to study the course of the developing disease. But that is provided that "ability to observe the claimant over an extended period of time is essential to understanding" his condition, Garrison v. Heckler, supra, 765 F.2d at 715, and that the treating physician knows something about the disease in question, and neither precondition has been established here.

There are other medical reports in the record, and we cannot say that no rational finder of fact could reject the company's attempted rebuttal. All we can say is that the administrative law judge's opinion is inadequate. There must be a remand, first to decide whether the case should have been reopened because of a mistake of fact and second to decide, if so, whether the company rebutted the presumption of black lung disease. It should go without saying that both inquiries are to be conducted in conformity with the discussion in this opinion.

REVERSED AND REMANDED, WITH DIRECTIONS.