Deborah , I too get the sense (from gay service members who have recently left the military) that "don't ask, don't tell" is generally disfavored by a broad segment of the military community. And I would be delighted for the Supreme Court to use this context as an occasion to rethink its stance on military deference if or when the case arrives there. But here is why I remain pessimistic.
I don't think the detainee cases are going to revise the court's stance on military deference in the "don't ask, don't tell" context. First, I think the detainee cases themselves exhibit a not insignificant amount of such deference. Second, like you, I think the DADT cases will cause to the court to look at cases involving the civil rights of service members, not the civil rights of detainees. Third, I think that if the court considers DADT, it will at most apply rational basis "with bite" (either under Lawrence v. Texas , which I take to be a rational-basis-with-bite case in the rights realm, or under Romer v. Evans , which could be read as a rational-basis-with-bite case in the classification realm).
The only Supreme Court equal-protection case I know where the plaintiff prevailed against the military is Frontiero v. Richardson (1973), where Sharron Frontiero successfully challenged a benefits scheme that facially discriminated on the basis of sex. That case is easily distinguished on the ground that it did not concern a core military function. In cases that did involve national security, military readiness, or unit cohesion, the court has been extraordinarily deferential, as I stated in my earlier post. Indeed, Rostker v. Goldberg (1981), which upheld the male-only draft, is also generically cited for the proposition that "judicial deference to ... congressional exercise of authority is at its apogee when legislative action under the congressional authority to raise and support armies and make rules and regulations for their governance is challenged."
It's worth dwelling on the sex discrimination context here. (Race and national origin are not going to be as probative, because the court will be uncomfortable citing Korematsu for military deference and because the racial integration of the armed forces was accomplished by Truman's executive order rather than through a court decision.) In 1981, when Rostker was decided, sex-based classifications already drew intermediate scrutiny under Craig v. Boren (1976). As Cass Sunstein has noted, even before United States v. Virginia (1996), intermediate scrutiny was a lot closer to strict scrutiny than it was to rational basis review. As a purely analytic matter, it seems that if military deference allows congressional legislation to survive intermediate scrutiny, it would a fortiori allow congressional legislation to survive the rational-basis-with-bite the court would apply under Lawrence or Romer .
As for the effect of a possible amicus brief from retired generals, I'm also not sanguine about the weight the court would give to such a brief, even if it came into existence. Some of the most visible retired generals, such as H. Norman Schwartzkopf or Colin Powell, testified in favor of DADT. So, they would have to have George Wallace moments to participate in such a brief. Moreover, even if such a brief were filed, I doubt the military's green brief would be as influential here as its brief allegedly was in Grutter v. Bollinger . In Grutter , the military brief cut for the court's inclination to defer to educational institutions; here, the military brief would cut against the court's inclination to defer to the political branches. More importantly, it would not just be the "civilian executive" (as you posit) defending the DADT policy, but also Congress. One of the awful things about DADT is that, unlike the prior 1981 executive order governing gays in the military, it is a joint creation of congressional legislation and executive regulation. And so one could argue that not only Article II military deference to the executive, but also Article I deference to the Congress, would be due from the court.
Don't get me wrong-I've written at length about the injustice of this policy. But unless the court revisits its military deference doctrine, the solution here is much more likely to be legislative repeal than judicial invalidation. Opponents of the policy should allocate their resources accordingly.