dissenting.
I readily concede that Mr. Eidukonis is not a sympathetic plaintiff. I nonetheless find myself unable to join in the majority opinion.
The result of this appeal depends on the legal standard applied. I believe that the language of 38 U.S.C. § 2204(d) and the Supreme Court’s interpretation of a similar provision compel the adoption of an approach akin to the one adopted in Gulf States Paper Corp. v. Ingram, 811 F.2d 1464, 1468 (11th Cir.1987)-an approach that looks solely to the reasonableness of *698the employee’s action and accords it a strong presumption of reasonableness. See 811 F.2d at 1468. Applying a modified Gulf States principle, which is more favorable to the reservist than that adopted by the majority, see infra, I would affirm. I do not think that the standards espoused in Gulf States and Lee v. City of Pensacola, 634 F.2d 886, 889 (5th Cir.1981), can be reconciled in the way that the majority has attempted, and, more important, I think that the Lee — Gulf States combination formula derived by the majority is not only legally incorrect but will be difficult for district courts to apply.
It is important to note that if SEPTA had established clear and reasonable leave policies, which Mr. Eidukonis violated, or if the district court had found that the pattern of absences had been the cause of his discharge, rather than finding that “[pjlaintiff was terminated by SEPTA solely because of his failure to report back from military duty on February 18, 1985,” Dist.Ct.Op. at 7 (May 17,1988), Mr. Eidukonis might have failed even my more generous standard. Faced with the present record, however, I see no alternative but to uphold the judgment of the district court.
I.
The majority attempts to reconcile the Lee and Gulf States cases as both adopting a “reasonableness” standard. I do not believe that they can be so reconciled. The Fifth Circuit’s approach in Lee is not completely clear, but it seems to involve an inquiry into the totality of the circumstances to ascertain whether the reservist-employee’s leave request was reasonable. See Bottger v. Doss Aeronautical Services, Inc., 609 F.Supp. 583, 585 (M.D.Ala.1985) (describing Lee as adopting a “totality of the circumstances” test). In Lee, the Court reasoned that in enacting 38 U.S.C. § 2024(d) (1982), Congress could not have intended “ ‘to permit employees who have been granted military leave to remain on such leave unnecessarily and at their own convenience to the detriment of the legitimate concerns of their employers.’ ” 634 F.2d at 889 (quoting district court’s opinion). With this in mind, the Lee court looked at the length of the leave, the amount of notice Lee gave his employer, whether Lee could have scheduled his training at other times, and the burden that Lee’s absence caused his employer. Id.
In Gulf States, the Eleventh Circuit, although purporting to apply Lee, tilted the Lee standard more in favor of the reservist-employee. See 811 F.2d at 1468-70. Under the Gulf States standard the court must “look for conduct akin to bad faith on the employee’s part in determining reasonableness.” Id. at 1469. Gulf States further held that the employee’s actions should be accorded a “presumption of reasonableness.” And it noted that a “burden to the employer alone is not enough to mark a leave request as unreasonable.” Id. There must also be “questionable conduct on the part of the employee.” Id. at 1470.
Though it adopts essentially the same factors to assess reasonableness which the Gulf States court did, see Maj.Op. at 697, the majority here expressly rejects the “akin to bad faith” approach of Gulf States. See Maj.Op. at 697. Although the majority holds that courts should consider “the legitimate needs of the employer,” in addition to the Gulf States factors, id. at 697, its opinion is, I believe, somewhat confusing on the issue of how much weight should be accorded to the employer’s interests. The majority adopts the language from Gulf States that burden on the employer alone is not enough to justify dismissal, but, unlike Gulf States, does not specify what extra factor is needed to establish unreasonableness. The majority has thus adopted a totality of the circumstances standard, but has given the district courts little guidance on how to assess the relevant information. Moreover, it has not made clear whether reasonableness should be presumed or whether the reservist has the burden of establishing it.
In sum, I believe that the majority’s effort to reconcile Lee and Gulf States has produced an elusive precedent which will be difficult to apply and which will make it impossible for reservists and their lawyers *699to predict with any degree of certainty how they will fare under any given set of circumstances.
II.
More significantly, I believe that the standard espoused by the majority is incorrect because it does not protect the reemployment rights of reservists to the degree required by Congress. Title 38 U.S.C. § 2024(d) is written broadly, mandating that leave for reserve duties “shall upon request be granted,” and that reservists “shall be permitted to return to [the] position^] [that they] would have had if [they] had not been absent for such purposes.” Furthermore, the Supreme Court, in construing a similar provision of the statute, 38 U.S.C. § 2021(b)(3) (1982 & Supp. IV 1986), stated that “[t]his Court does not sit to draw the most appropriate balance between benefits to employee-reservists and costs to employers. That is the responsibility of Congress.” Monroe v. Standard Oil Co., 452 U.S. 549, 565, 101 S.Ct. 2510, 2519, 69 L.Ed.2d 226 (1981). In Monroe, the Supreme Court noted that
[t]he frequent absences from work of an employee-reservist may affect productivity and cause considerable inconvenience to an employer who must find alternative means to get necessary work done. Yet Congress has provided in § 2021(b)(3) that employers may not rid themselves of such inconveniences and productivity by discharging ... employee-reservists solely because of their military obligations.
Id. Similarly, in enacting § 2024(d), Congress has provided that employers may not rid themselves of the burden of having employees who are members of the Army Reserve by refusing to reinstate them after their leave.1
Despite this background, I agree with the majority that the right of reservists to take military leave should not be absolute. See Maj.Op. at 695. It is true that we generally bow to the plain meaning of a statute. See Piper v. Chris-Craft Industries, Inc., 430 U.S. 1, 24-26, 97 S.Ct. 926, 940-42, 51 L.Ed.2d 124 (1977). However, it has long been a maxim of statutory construction that “ ‘[g]eneral terms should be so limited in their application as not to lead to injustice, oppression, or an absurd consequence. It will always, therefore, be presumed that the legislature intended exceptions to its language, which would avoid results of this character.’ ” Government of Virgin Islands v. Berry, 604 F.2d 221, 225 (3d Cir.1979) (quoting United States v. Kirby, 74 U.S. (7 Wall.) 482, 486-87, 19 L.Ed. 278 (1868)). Were we to read § 2024(d) as creating an absolute right of reinstatement, reservists would be allowed to play fast and loose with the system in a way that Congress could not have intended. However, given the considerable breadth of the right accorded to reservists by the literal terms of the statute, and the Supreme Court’s pronouncement that the balance between the employer and the employee is for Congress to strike, I would create only a very narrow exception to the reservists’ general right to be reinstated after military leave.
More precisely, I would adopt a standard much like the one adopted by the Eleventh Circuit in Gulf States. I would hold that the relevant inquiry is whether the employee’s actions have been reasonable, and I would accord a strong presumption of reasonableness to them. An employer could rebut this presumption only by presenting evidence that the employee acted highly unreasonably. I use a “highly unreasonable” standard rather than the “akin to bad faith” standard articulated in Gulf States for two reasons. First, I find the term “akin” to be somewhat confusing in this context. Second, I think that adopting a *700“bad faith” formulation may put too much of a burden on employers by requiring them to offer proof about the mental state of the reservist.
The burden on the employer is not wholly irrelevant to this inquiry, but I agree with the Eleventh Circuit that a court should consider whether the leave has caused hardship to the employer only if the reservist has engaged in “questionable conduct.” Gulf States, 811 F.2d at 1470. For example, a reservist’s failure to notify his employer as soon as practicable that he would be taking leave could be a questionable act, as could an employee’s violation without good cause of an employer’s established reasonable leave policy. Such instances, if they are sufficiently egregious, may be in and of themselves highly unreasonable. Alternatively, they may be considered highly unreasonable even if they are somewhat less egregious, if the employee was aware that they would cause significant hardship to the employer.
III.
I accept the facts as described by the majority, which, as I have observed, do not paint a sympathetic picture of Mr. Eidukon-is. What is legally significant, however, under the standard I would apply is the district court’s finding that Mr. Eidukonis
was terminated by SEPTA solely because of his failure to report back from military duty on February 18, 1985, and not because of any previous military leaves, not because when he took those military leaves, not because of when he took the military leave that was — that ended on February 16, 1985, not because of notices given concerning those leaves, not because of the number of his prior military leaves, not because of the length of his military service ... and not because he broke any promises about seeking military duty.
Dist.Ct.Op. at 7. This factual finding is not clearly erroneous. Indeed, it is supported by statements made by SEPTA’s representatives at trial. See Trial Trans, at 86-87 (May 3, 1988).
In light of this finding, the only relevant issue is whether Mr. Eidukonis acted highly unreasonably in failing to report to work at SEPTA on February 18, 1985. I believe that he did not. Mr. Eidukonis notified SEPTA “within three days of learning that [the extension] had been approved” that he would be extending his leave. Dist.Ct.Op. at 8. He was in the midst of an important computer project, about which he had developed special expertise. See id. And SEPTA did not notify him that there were limits on the amount of leave a reservist could take, or give him any warnings that he could be fired because of his leave, until after his orders had been issued, and only one week before his February 18 assignment was to begin. I believe that under these circumstances, Eidukonis’s failure to return to work on February 18 did not constitute questionable conduct. Consequently, I would not consider the burden that his absence placed on SEPTA.
This might have been a different case if the district court had found that Mr. Eidu-konis was fired because of his pattern of leave-taking. That pattern viewed as a whole may well have been unreasonable even under the standard that I propose. But the district court made no such finding. Similarly, this might have been a different case if SEPTA had articulated clear and reasonable policies regarding military leave, and Mr. Eidukonis had willfully violated them. Again, that is not this case. On the record before us, I would affirm the judgment of the district court. I respectfully dissent.
. The legislative history of § 2024(d) is not particularly helpful in determining what standard we should adopt. As the majority points out, there are some references in the history to the fact that when deliberating about the statute, Congress was primarily thinking about military leaves of less than 90 days. See Maj. Op. at 693. But the majority itself concedes that these references are not dispositive, and there is no clear indication of congressional opinion with respect to the question of how great a burden employers should be required to bear. The legislative history simply does not discuss the possibility or contours of a "reasonableness test.”