It's not a logic game , Eric—it's simply good, old-fashioned judicial review. The role of the "C-Box"—the court—is not to determine whether the detainee (not a "criminal suspect," by the way) is in fact telling the truth, but instead to determine (i) whether the "M-Box" is relying on a valid legal theory to detain the prisoner; and (ii) if so, whether there is any reason to credit the M-Box's conclusion that the detention satisfies that legal standard, even after giving quite a bit of deference to the M-Box-that is to say, to determine whether the M-Box decision is supported by a preponderance of relevant evidence.
You might recognize this—it's what courts do all the time in reviewing executive conduct and what habeas courts have done for time immemorial.
In this case, Phil is right: When there was no prospect of judicial review, the government could detain Uighur prisoners for more than six years, based on a shockingly broad theory of what Congress has authorized, and on flimsy supposition about the facts of the case. But as soon as a court is in the picture—even a court that included David Sentelle and Tom Griffith—and puts the government to the modest burden of, gosh, explaining its decision, it becomes clear very quickly not only that the administration is acting on the highly implausible view that Congress has authorized the indefinite detention of all the world's suspected terrorists (even those whose target might be China), and not only that the military assumes that attending a Uighur training camp makes one a member or supporter of the ETIM, but also that the government's conclusions that the ETIM is part of the Taliban and has engaged in hostilities against the United States are based almost entirely on the government's own say-so (which in turn is likely based on unsupported representations offered by the Chinese, who are more than happy to have the United Statesimprisoning the critics of that government).
I can't do better here than quoting from "the C-Box":
The principal evidence against Parhat regarding the second and third elements of DOD's definition of enemy combatant consists of four government intelligence documents. The documents make assertions-often in haec verba -about activities undertaken by ETIM, and about that organization's relationship to al Qaida and the Taliban. The documents repeatedly describe those activities and relationships as having "reportedly" occurred, as being "said to" or "reported to" have happened, and as things that "may" be true or are "suspected of" having taken place. But in virtually every instance, the documents do not say who "reported" or "said" or "suspected" those things. Nor do they provide any of the underlying reporting upon which the documents' bottom-line assertions are founded, nor any assessment of the reliability of that reporting. Because of those omissions, the [CSRT] could not and this court cannot assess the reliability of the assertions in the documents. And because of this deficiency, those bare assertions cannot sustain the determination that Parhat is an enemy combatant.
The [M-Box] insists that the statements made in the documents are reliable because the State and Defense Departments would not have put them in intelligence documents were that not the case . This comes perilously close to suggesting that whatever the [M-Box] says must be treated as true, thus rendering superfluous both the role of the Tribunal and the role that Congress assigned to this court. We do not in fact know that the [M-Box] regards the statements in those documents as reliable; the repeated insertion of qualifiers indicating that events are "reported" or "said" or "suspected" to have occurred suggests at least some skepticism. Nor do we know whether the [M-Box] relies on those documents for decisionmaking purposes in the form in which they were presented to the Tribunal, or whether they supplement them with backup documentation and reliability assessments before using them to take actions of consequence.
That is to say, the C-Box asked the M-Box to explain its methods, and, after taking a look, concluded that the M-Box is, well, a Ouija board. Now, remind me: Why on earth should we credit the M-Box's conclusions, especially where the liberty of persons detained for more than half a decade is at stake?