There’s lots of great blogging about Hamdan around the blogosphere. I was particularly impressed by the interesting thoughts at Opinio Juris and Balkinization.
There’s lots of great blogging about Hamdan around the blogosphere. I was particularly impressed by the interesting thoughts at Opinio Juris and Balkinization.
My main concern about Hamdan seems to be unlike anyone else’s, so maybe Professor Kerr would be kind enough to tell me whether I’m the one who’s crazy, or everyone else is crazy.
If, as five Justices contended yesterday in Hamdan, 28 USC 2241(e) does not apply to pending cases, then why in the world would Congress have written an EXCEPTION to 28 USC 2241(e) that explicitly does apply to cases that are “pending on or after the date of the enactment”? It makes absolutely no sense to me.
In other words, how can an exception to a rule apply to pending cases, if the rule itself does not apply to pending cases?
Easy: the “exception” you identify isn’t, by its own terms, an exception to 2241(e).
DTA sec. 1005(h)(2) says this: Paragraphs (2) and (3) of subsection (e) [i.e., DTA sec. 1005(e)] shall apply with respect to any claim whose review is governed by one of such paragraphs and that is pending on or after the date of the enactment of this Act.
However, 28 USC 2241(e) was enacted by DTA sec. 1005(e)(1), not (2) or (3). Thus, as the Court states quite clearly (slip op. 9-10), “The Act is silent about whether paragraph (1) of subsection (e) ‘shall apply’ to claims pending on the date of enactment.”
Mark, 28 USC 2241(e) says:”EXCEPT as provided in section 1005 of the Detainee Treatment Act of 2005, no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action….”
Standing alone, 28 USC 2241(e) clearly applies to both pending and future cases. As Justice Scalia pointed out in dissent:
“the Court … cannot cite a single case in the history of Anglo-American law (before today) in which a jurisdiction-stripping provision was denied immediate effect in pending cases, absent an explicit statutory reservation.”
Anyway, 28 USC 2241(e) instructs us to look at 1005 of the DTA in order to find exceptions to the denial of jurisdiction. And, we find at 1005(h)(2) that the exceptions apply “with respect to any claim … that is pending on or after the date of the enactment of this Act.”
If the rule in the US Code does not apply to pending cases, then how could an exception to that rule apply to pending cases? It cannot, and thus 1005(h)(2) buttresses the meaning that 28 USC 2241(e) would have had standing alone, IMHO.
I might add that here are additional reasons for believing the Court had no jurisdiction in this case. The Court’s rationale would require that 1005(e)(2) and 1005(e)(3) are the ONLY provisions of the DTA that apply to pending cases. There is nothing in the DTA that would distinguish 1005(e)(1) from, say, 1005(e)(4)(f) in this regard.
P.S. Any chance Orin could delete the 4th and 5th comments in this thread?
Andrew:
There is no section (e)(4)(f) in 1005. There is a section 1005(f), but I’m not sure the relevance of that — it is not one of the three parallel jurisdiction-stripping provisions that are at the heart of 1005(e).
As to the broader question: the argument goes like this. I say that the courts don’t have jurisdiction over A, B and C. I then go on to say that, regarding B and C, this act applies to all cases pending on or after the date of enactment. If the act means what you say it means (that the first part is sufficient to withdraw all jurisdiction over cases pending on or after the date of enactment), the second part becomes superfluous, because the act would already say that in the first section. Generally, statutes shouldn’t be construed in a way that makes language superfluous, because there should be a presumption that every word is meaningful.
Thus, the reading is that the enumeration of B and C presupposes something not enumerated — which, by inference, is A. Ergo, Congress did not intend A to be treated in the same way as B and C, because it explicitly failed to include it when it was listing the classes of cases to which it wished the act to apply on or after the date of enactment. The court’s reading preserves a distinction that does seem embedded in the lanugage of 1005(h).
At least, that’s how I understand the argument. I was initially hostile to the court’s reading, but on reflection, I begin to think they have the better side of this argument.
Marghlar, thanks for your comment. Of course, you’re correct that there is no section (e)(4)(f) in 1005 of the DTA. What I meant to say is “1005(e)(4) or (f).”
In a previous comment at this blog, you said, “I think Scalia had it right on the jurisdictional question, but the merits being reached, I agree with the majority.” But I don’t want to even think about the merits, because I continue to think the Court’s decision on the jurisdictional question was preposterous.
In your example, you designate 1005(e)(1), 1005(e)(2), and 1005(e)(3) as A, B, and C respectively. I’m with you so far, but no farther.
“A” is a jurisidiction-stripping provision. It completely strips jurisdiction from the courts regarding Guantanamo detainees. B and C, by contrast, are limited restorations of jurisdiction. “A” amends the US Code to strip all jurisdiction, but says that the exceptions are to be found in the DTA, and sure enough those exceptions are spelled out in B and C. I fail to see how there is anything “superfluous” about this interpretation. It’s the plain meaning. Even Justice Stevens acknowledged that “subsection (e)(1) strips jurisdiction while subsections (e)(2) and (e)(3) restore it.”
You contend that Congress did not intend A to be treated in the same way as B and C, because A explicitly failed to include language about pending cases as it did for B and C. However, as someone else noted at SCOTUSBlog, “While Senator Levin succeeded in getting express application language removed [from "A"], he did not get express nonapplication language inserted.” More to the point, “A” is a general cancellation of jurisdiction, while “B” and “C” are listed exceptions; when a list of exceptions is provided, that list should ordinarily be deemed exclusive, under the principle of exclusio unius. And note that if those exceptions weren’t listed at all, then it’s undisputed that “A” would have applied to pending cases.
My point about 1005(e)(4) and (f) is simply that it would be absurd to suppose that they apply only to future cases, and yet that’s exactly the kind of spin that the Court has put on 1005(e)(1)=A. This is merely one additional reason why I still find the Court’s opinion virtually incomprehensible.
As Mark noted above, the Court said: “The Act is silent about whether paragraph (1) of subsection (e) ’shall apply’ to claims pending on the date of enactment.” Yes, and the Act is also silent about whether it applies to “chauffeurs.”
P.S. The DTA says that 1005(e)(2) and 1005(e)(3) “apply with respect to any claim … that is pending on or after the date of the enactment of this Act.” Would anyone seriously contend that therefore 1005(e)(1) does NOT apply to claims pending on “OR AFTER” the date of enactment? Of course not. I want to be nice and polite to Justice Stevens, but IMHO this opinion is just lousy.
I tend to think that there are persuasive arguments on both sides of the jurisdictional argument regarding the DTA, as the posters here have outlined above. However, I find Scalia’s abstention argument particularly persuasive. Judicial restraint should have been employed.
However, that being said, the merits of this case are fascinating. One question that i have is whether the lack of presence of the accused under the UCMJ was actually joined by five Justices. Kennedy’s concurrence at page 16 talks solely about the evidentiary problems of the commissions and the last sentence of the first full paragraph on page 19 seems to sugest that he does not join the Court’s opinion regarding presence under the UCMJ. It seems fairly clear to me that this is the case, but did anyone else read the opinion in this fashion?
Also, Alito’s opinion regarding what is a “regularly constituted” court under Geneva seems to rebut easily the majority’s opinion on this issue. I don’t recall Clement arguing this point at oral arguments, nor do I recall it in the government’s brief.
Andrew:
I’d say that e(4) and (f) are clearly very different than is (e)(1) — they both qualify all other provisions of the act. By contrast, the act contains three central jurisdictional provisions.
First, it provides that, subject to enumerated exceptions, all jurisdiction is withdrawn.
Second, it says that the D.C. Cir. has exclusive jurisdiciton over one class of cases.
Third, it says that the D.C. Cir has exclusive jurisdiciton over a second class of cases.
It then goes on to say that provisions two or three apply to all cases pending on or after the date of enactment.
If you are right, and it was clear from the plain language of (1) that it applied to all cases pending on or after the date of enactment, it seems like it would be equally clear that the D.C. Cir had exclusive jurisdiction over all cases enumerated in exactly the same way. Yet Congress went ahead and defined the scope of time covered by two or three — apparently concluding that such a meaning was not unambiguous. If your reading is right, there is absolutely no reason for 1005(h) to exist at all, because that interpretation would obviously flow from (e)(2) and (3).
As to what Senator Levin may have wanted, I think that it is inapppropriate for courts to consider legislative history, or any other non-textual indica of legislative intent, when interpreting statutes. So I don’t consider such data relevant to the task at hand.
I’d agree that your reading is within the scope of ambiguity of the act — I just am coming around to the point of view that Justice Stevens’s reading is the better interpretation of the same ambiguity.
Marghlar, you make two related points:
A] Congress expressly applied 1005(e)(2) and 1005(e)(3) to pending cases because Congress believed 1005(e)(2) and 1005(e)(3) might not otherwise apply to pending cases.
B] Because Congress believed that 1005(e)(2) and 1005(e)(3) might not apply to pending cases, Congress must have also believed that 1005(e)(1) might not apply to pending cases.
For the sake of argument only, I will concede your point [A].
Regarding your point [B], there are very compelling textual reasons why Congress would believe that 1005(e)(2) and 1005(e)(3) might not apply to pending cases, without believing that 1005(e)(1) might not apply to pending cases. First and foremost, 1005(e)(2) and 1005(e)(3) comprise a list of exceptions to 1005(e)(1), and therefore the rule of expressio unius implies that there is no additional exception for pending cases. Congress had no similar reason to suppose that 1005(e)(2) and 1005(e)(3) would not have an exception for pending cases.
But there is an issue of semantics here that your certainty seems to belie — what does it mean to say that something is an exception? I’m not sure that a definition of the effect over time/cases of a rule is necessarily an exception to that rule. One can quite logically say that A, which strips all jurisdiction, is defined over a particular class of cases; it is also subject to exceptions B and C, each defined over a different class of cases which partially overlaps with the cases in A. To the extent that A would require otherwise, B and C act as exceptions to it. However, B and C function independly of A, and only act as exceptions to it when they conflict.
Hence, I’m not sure that expressio unius has the force you suppose here.
Marghlar, 28 USC 2241(e) says: “EXCEPT as provided in section 1005 of the Detainee Treatment Act of 2005, no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action….”
Do you deny that the exceptions referred to in that quote are expressly listed 1005(e)(2) and 1005(e)(3)? Why wouldn’t expressio unius therefore apply so as to rule out any other exceptions? I don’t see any semantics at work here at all.
Because I’m not sure that a definition of the breadth of a stripping provision is an exception to that provision.
If A states a general rule, and B and C are carveouts from that rule, that doesn’t mean that A doesn’t have a defined scope.
Marghlar, 28 USC 2241(e) announces that it has exceptions: “EXCEPT as provided in section 1005 of the Detainee Treatment Act of 2005, no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action….”
You’re actually denying that the exceptions referred to in that quote are expressly listed at 1005(e)(2) and 1005(e)(3)? My mind is now officially boggled.
Why is this so difficult to understand?
A has a defined scope, which includes a set of cases, A.
B & C are enumerated exceptions to that scope.
I don’t see why the defined scope of A is necessarily an “exception” to itself. Indeed, such a reading seems a bit odd. Section A either does or doesn’t apply to currently pending cases, but either way, stating what cases it does not by definition apply to is not stating an exception, it is defining a scope, I think.
As a matter of set theory, we’d say that the total scope of the jurisdictional void created by the statute is the set A, minus the disjoint of the sets B and C (as defined). The exceptions for B and C need not control the content of the original set A, which is what the court was attempting to define, looking to the statute as a whole.
The sticking point seems to be that you believe the court was reading in an exception to the statute. They would say they were interpreting the text of the statute, in the context of all of its provisions.
Marghlar, as I pointed out at 3:58 AM, it is “undisputed” that 1005(e)(1) standing alone would have cancelled jurisdiction not only for future cases but also for pending cases. In other words, if there had been no 1005(e)(2) and 1005(e)(3) then the meaning of the statute would have been to completely cancel jurisdiction for both future and currently pending cases. No justice on the Supreme Court suggested otherwise. Are you suggesting otherwise?
I’m not sure that is right. I think that the provision in 1005(h), which obviously would not exist but for the listed exceptions, provided support for the court’s reading. However, I do think that the court’s reading is within the scope of ambiguity of the act. To say that no court shall have jurisdiction of a class of cases, begs several questions. Is an appeal a case? And does shall carry a temporal implication of future effect (in can have that connotation — “I shall go to the store,” for instance, would commonly be thought of as a statement of future intent/effect.
Now, I’d say that your construction would be equally plausible, if 1005(e)(1) stood by itself. (At that point, I think it is within the court’s discretion to decide either way, although I’d advise them to tread softly for prudential reasons). However, once 1005(h) enters the mix, it looks less balanced to me. Now, we have ambiguous language in the original statute, coupled with a structural reason to think that a construction other than “all cases pending on or after enactment” is to be preferred. At that point, I think Stevens’ position starts to read as a fairer construction than Scalia’s.
Furthermore, Justice Stevens notes this ambiguity in his opinion (so I take issue with your assertion that no SCOTUS justice took a similar view): “The Act is silent about whether paragraph (1) of subsection (e) “shall apply” to claims pending on the date of enactment.” (Slip. Op. at 9-10). Furthermore, he holds that the “plain language” argument of Justice Scalia necessarily depends on the Bruner presumption, which he holds is inapplicable to statutes that totally withdraw jurisdiciton (rather than merely transfering it).
Thus, I do think the plurality (and perhaps also Justice Kennedy, although I don’t have time to look right now) believe that the text of 1005(e)(1) is permissive of their reading. I don’t think anyone was trying to do something that would have been textually excluded by the DTA.
Marghlar, you’re correct that the Court said that the “Act is silent about whether paragraph (1) of subsection (e) ‘shall apply’ to claims pending on the date of enactment.” Indeed, there was no explicit statement in the DTA that (e)(1) “shall apply” to pending claims. Likewise, there was no explicit statement in the DTA that the Act applies to “former chauffeurs.” That does not preclude a plain meaning that does apply to former chauffeurs, or a plain meaning that does apply (e)(1) to pending claims.
Although the Court disparaged any “blinkered study of subsection (e)(1) alone,” the Court did acknowledge that “subsection (e)(1) strips jurisdiction while subsections (e)(2) and (e)(3) restore it in limited form.” Surely, this means that (e)(1) stripped jurisdiction over pending cases, seeing as how the Court held that (e)(2) and (e)(3) established jurisdiction over pending cases.
In any event, nowhere did the Court deny that (e)(1) standing alone would have applied to pending cases. Consider plain English. (e)(1) says, “[N]o court, justice, or judge shall have jurisdiction to hear or consider an application for a writ of habeas corpus filed by or on behalf of an alien detained by the Department of Defense at Guantanamo Bay, Cuba.” Surely, the US Supreme Court is a “court” that did “hear or consider” a “writ of habeas corpus” filed by Hamdan. The plain language is unequivocal that (e)(1) standing alone would have applied to pending cases.
As Justice Scalia correctly observed, the Court “cannot cite a single case in the history of Anglo-American law (before today) in which a jurisdiction-stripping provision was denied immediate effect in pending cases, absent an explicit statutory reservation. By contrast, the cases granting such immediate effect are legion….” No court (including the Hamdan Court) has ever construed free-standing language like that in (e)(1) as not applying to pending cases.
From all this, it is clear that the meaning of (e)(1) standing alone would have been to cancel jurisdiction in pending cases as well as future cases.
The Court inferred an unwritten exception to (e)(1) from the language of the rest of the DTA, despite the fact that the DTA explicitly lists the precise exceptions that were intended. Oddly, neither the Court nor the dissent mentioned the blatant disregard of the expressio unius maxim.
I know it’s not fashionable to delve into the legislative history, but for what it’s worth, Congress considered a version of the DTA that explicitly would have barred consideration of Hamdan’s case, and that version was not passed. That makes Mr. Hyman’s reading of the statute unreasonable. Carl Levin, a sponsor of the DTA, has repeatedly said so, and only the manufactured Congressional Record “exchange” between Kyl and Graham suggests otherwise (see the footnote in the majority opinion).
Andrew (and marghlar)–
Let’s ask a simple question. Why does paragraph (h) exist at all?
The existence of paragraph (h) suggests that there was some reason to conclude that (e)(2) and (e)(3) did not apply to pending cases.
A kind of schematic drawing of Marghlar’s logic might look like this:
Although the Court disparaged any “blinkered study of subsection (e)(1) alone,” the Court did acknowledge that “subsection (e)(1) strips jurisdiction while subsections (e)(2) and (e)(3) restore it in limited form.” Surely, this means that (e)(1) stripped jurisdiction over pending cases, seeing as how the Court held that (e)(2) and (e)(3) established jurisdiction over pending cases.
Actually, the court didn’t hold anything about (e)(2) or (3), which weren’t at issue — but it’s reasoning suggests strongly that the reason we know which cases they apply to is because 1005(h) explicitly tells us so. Hence, by themselves, all three would likely be equally ambiguous over whether jurisdiciton was stripped from only newly filed cases, or over all cases, including pending cases.
In any event, nowhere did the Court deny that (e)(1) standing alone would have applied to pending cases. Consider plain English. (e)(1) says, “[N]o court, justice, or judge shall have jurisdiction to hear or consider an application for a writ of habeas corpus filed by or on behalf of an alien detained by the Department of Defense at Guantanamo Bay, Cuba.” Surely, the US Supreme Court is a “court” that did “hear or consider” a “writ of habeas corpus” filed by Hamdan. The plain language is unequivocal that (e)(1) standing alone would have applied to pending cases.
Is it so unambiguous? I’m not sure. First, note that the Supreme Court isn’t considering an “application for a writ of habeas corpus” — a lower court did that. Rather, they are hearing an appeal from a denial of such an application. There is a space of ambiguity there — is an appeal from an application, also an application?
Furthermore, note that the word “shall” often connotes a future temporal aspect (which has historically been the subject of a presumption against retroactivity in statutes). Thus, the language does seem permissive of a reading that because courts shall not have jurisdiction, that means that they cannot gain jurisdiciton over new cases — but in cases where jurisdiction was already proper, they may continue. (The alternative usage: “no court has jurisdiciton” would connote more clearly the meaning you read in the act.)
Note that (e)(2) and (e)(3) use entirely parallel language — “the United States Court of Appeals for the District of Columbia Circuit shall have exclusive jurisdiction….” — and nevertheless, Congress clearly felt these provisions left some ambiguity about the scope of the jurisdicitonal cut-off/grant, because it included an explicit jurisdicitonal provision in subsection (h).
And Justice Scalia’s reliance on precedent is somewhat besides the point — each statute must be construed on its own merits. I find it odd that an avowed formalist would champion a doctrine that creates presumptions about what stautory language means. I agree with what Justice Scalia said about statutory construction in A Matter of Interpretation: courts should construe acts neither strictly nor loosely, but fairly, to mean what they in fact say. The court looks at the act as a whole, and finds that the canon of expressio unius in fact compels the conclusion that Congress both felt that the jurisdictional grants/denials were ambiguous, and that they did not wish to treat 1005(e)(1) identically with 1005(e)(2) or (3).
Thus, as I’ve repeatedly said, I don’t think the court was creating an “exception” to 1005(e)(1) — I think they were determining its proper temporal scope. Hence, I don’t think the expressio unius canon applies. But you are free to disagree, of course.
Marghlar, you say that a lower court (rather than SCOTUS) is the one that “heard” or “considered” the application for the writ of habeas corpus. Even if I were to concede that you’re correct about that, it makes absolutely no difference, because in that case (e)(1) would deprive the lower court of jurisdiction, and the SCOTUS should have said so.
Moreover, if the Supreme Court has in the past authoritatively stated what a term or phrase means, and then Congress subsequently uses that phrase or term in a subsequent statute, it is absurd to suppose that the Court’s meaning should not be impputed to Congress. I really think it’s unrealistic to say otherwise. Provisions such as 1005(e)(1) have been construed by the Court for ages as being applicable to pending cases. I very much doubt that even the judges in the Hamdan majority would dispute that 1005(e)(1) standing alone would have applied to pending cases. Certianly Senator Levin would not dispute that, and he was the most adamant and influential congressional opponent of the administration regarding Hamdan. Levin has stated that the reason 1005(e)(1) does not apply to pending cases is not because of its own language, but rather because there was an exception elsewhere in the DTA. Levin has said: “As I pointed out when we passed the bill, the provision says that it ‘shall take effect on the date of the enactment of this Act.’ The meaning of these words is clear: the provision is prospective in its application, and does not apply to pending cases. The Administration is just plain wrong when it says otherwise.”
As I’ve repeatedly said, it was highly improper for the Court (and for Levin) to infer an unwritten exception to 1005(e)(1) despite the fact that the DTA expressly listed the desired exceptions. Why this expressio unius argument did not appear in the justices’ opinions is something that I’m still trying to figure out.
As a non-lawyer, I have read the above exchange with much interest, please forgive any lack of legal expertise in the questions asked. Does the argument come down to whether positive inference or negative inference is applicable when ‘express’ language is not provided on a specific issue ?
Would it not be just as valid to claim, knowing the political manipulation of floor statements by both parties, that had they wanted a pending cases exception to 1005(e)(1), they would have used specific language to do so ?
I find the argument that an earlier version, that had express language on pending cases, was defeated to be somewhat persuasive, but, it still leaves me to wonder why, if that was so important, the six words ‘will not apply to pending cases’ does not appear in 1005(e)(1) of the version that passed ?
Doesn’t using negative inference, in this case, lend itself to the impression of legislating from the bench by adding words to legislation does not exist, whereas positive inference relies strictly on the language present ?
Would it not be wiser to send the act back to Congress for clarification to correct that which is unclear ? Is there any process that would allow that avenue of action ?
Please let’s not forget the cannons of statutory construction that statutes must be read as a whole and each word must have meaning. This makes the whole if e(1) were by itself argument largely irrelevant. The second you add other sections they have to be construed together. The whole argument that negative inferences cannot be drawn from statutes is also bunk. There is a long line of precedent supporting this.
There are two fundamental tensions at play here: 1)Scalia’s insistence that a judicial presumption for retroactivity is superior than the negative inference; and 2)whether the language of e(1) trumps the negative inference.
These are both tough questions. As to 1, Scalia is probably not the best person to rely on for giving an accurate reading of the Lindh’s majority opinion considering that he dissented and would like to see the case marginalized. Moreover, Scalia arguing that a judicial presumption trumps a reading of the plain language of a statute is a little specious given his judicial philosophy. As to 2, I think Stevens (legitimately) relied on the drafting history and floor debate to break the tie between e(1) and negative inference. Of course if you don’t like LH, you would disagree.
KMAJ2, the premise of your question is that the DTA is unclear. Marghlar has made a similar argument. However, I continue to believe that an objective look at the Detainee Treatment Act shows that it is not unclear at all regarding application to pending cases.
28 USC 2241(e) says categorically: “Except as provided in section 1005 of the Detainee Treatment Act of 2005, no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action….”
Marghlar argued that this is ambiguous, but I very much disagree. Putting aside any exceptions in section 1005 of the DTA, the quoted sentence plainly should have applied in Hamdan. Why use the word “justice” if Congress did not contemplate a future Supreme Court hearing? I don’t think the quoted sentence is susceptible to application of any presumption of non-retroactivity. The only way to get around the quoted sentence is by way of exceptions in section 1005 of the DTA, and the only express exceptions in section 1005 of the DTA are (e)(2) and (e)(3).
Skeptic, of course statutes must be read as a whole and each word must have meaning. This doesn’t make the whole “if e(1) were by itself” argument irrelevant. The question is whether the Court has applied a non-retroactivity exception to (e)(1), and the only way to determine that is to consider what (e)(1) would mean without any exceptions.
Also, contrary to what you said, Scalia did not argue that a judicial presumption trumps a reading of the plain language of a statute. The plain language of (e)(1) is that it applies to pending cases. (e)(1) does not say that “no judge shall have jurisdiction to begin hearing or considering … habeas corpus … or … any other action.” It says that “no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action ….” The plain language is that it applies to pending cases. Scalia merely pointed out that a long line of cases support that plain reading.
Bryan, you’re correct that Congress considered a version of the DTA that explicitly would have barred consideration of Hamdan’s case, and that version was not passed. In other words, Senator Levin did succeed in getting express language removed barring Hamdan’s case, but Levin did not get express language inserted allowing Hamdan’s case to go forward. Also, you’re incorrect that the exchange between Kyl and Graham was the only indication in the record that Congress meant to bar Hamdan’s case from going forward — see notes 3 and 4 in Scalia’s opinion.
Some Guy, you say that the existence of paragraph (h) suggests that there was some reason to conclude that (e)(2) and (e)(3) did not apply to pending cases. That’s true. In contrast, there was no reason to conclude that (e)(1) did not apply to pending cases, for at least three reasons: first, (e)(1)’s plain language indicates that pending cases are included; second, the courts’ precedents regarding provisions like (e)(1) say that pending cases are included; and third, the exceptions to which (e)(1) refers are expressly listed elsewhere in the DTA without mentioning any exception for pending cases.
Thank you for your reply, Andrew, I apologize if it appeared I was agreeing on ambiguity. My experience in debate/argumentation, while not legally oriented, was the reason behind the statement about the defeated version being ‘somewhat persuasive’. It is also why I wondered why the inverse was not equally persuasive, the lack of ‘express’ language excluding pending cases. I agree with you that the language of the DTA is not ambiguous, it is about as straight forward, no hedge words, as language in any act can be.
The line of argumentation of the majority seem to be along the post hoc ergo proptor hoc or modus ponens style, in other words, because Congress defeated one version, the passed version must mean this, which is a logical fallacy, because of the lack of express language, thus that reasoning does not necessarily hold true. That was the basis for my question of why negative inference should have held more weight than positive inference, due to the lack of express language exemping pending cases, when it could have easily been inserted if it was that strong of a point.
If you think that Congress is always paying careful attention to SCOTUS precedent when drafting statutes, you have far more confidence in their abilities than I do. I certainly think the background tradition of interpretation deserves to be looked to in determing meaning, but I can’t agree that it should always be dispositive, especially when their are countervailing signals of intent running counter to the precedent.
I have very little interest in what Levin said or thought — his unenacted understanding and intention have no place in the construction of a statute, because it wasn’t his intent that became law, it was the wording of the statute that he helped to shape. Furthermore, I think Stevens’ opinion was clear on the point that he didn’t think that 1005(e)(1) clearly spoke to the question presented, nor did he believe that the Bruner line of precedent should properly be extended to statutes that totally cut-off all jurisdiction, as opposed to transfering the locus of jurisdiction.
Why use the word “justice” if Congress did not contemplate a future Supreme Court hearing?
Because the SCOTUS can hear habeas applications directly, in some cases. See 28 U.S.C. 2241(a), 2242. Thus, the hear both “applications,” and appeals from applications. Ergo, the use of the word “Justice” is not dispositive on the question.
Marghlar, you say: “Stevens’ opinion was clear on the point that he didn’t think that 1005(e)(1) clearly spoke to the question presented.” Yes, Stevens was clear, but he was wrong.
Indulge me by taking another look at the Detainee Treatment Act:
“[N]o court, justice, or judge shall have jurisdiction to hear or consider (1) an application for a writ of habeas corpus filed by or on behalf of an alien detained by the Department of Defense at Guantanamo Bay, Cuba; or (2) any other action against the United States or its agents relating to any aspect of the detention by the Department of Defense of an alien at Guantanamo Bay, Cuba, who–(A) is currently in military custody….”
Look at those last five words. Thus, section (e)(1) of the DTA manifestly, unambiguously, incontrovertibly applied to preenactment conduct. How can it possibly be denied?
“Where the statute in question unambiguously applies to preenactment conduct, there is no conflict between the anti-retroactivity presumption and the principle that a court should apply the law in effect at the time of decision.”
—Landgraf v. USI Film Products (headnote), 511 US 244 (1994).
Well, a person who is currently in military custody may well continue to be in custody in the future, right? Furthermore, even if they were released the moment the act was passed, they still could try and bring an action against the U.S. for damages, which would be barred by 1005(e) if it applied only prospectively.
Furthermore, the act’s language only takes effect upon its becoming law — which is when 1005(h) would make all provisions take effect. So that section would be totally consistent with a forward-looking denial jurisdiction.
I think your argument fails to the degree it depends on those five words.
Marghlar, you apparently would like to apply the anti-retroactivity presumption to 1005(e)(1) of the DTA, as Justice Stevens may have done, even though that presumption has never before been applied to a jurisdiction-stripping provision. Right?
So, fine, let’s apply the presumption, despite SCOTUS precedent to the contrary. How can the presumption be overcome? According to the headnote in Landgraf, “Where the statute in question unambiguously applies to preenactment conduct, there is no conflict between the anti-retroactivity presumption and the principle that a court should apply the law in effect at the time of decision.”
So then the question becomes whether or not 1005(e)(1) applies to preenactment conduct, right?
Obviously, 1005(e)(1) applies to pre-enectment conduct. It denies federal court jurisidiction over detainees “in military custody” at the moment of enactment. Right? Does that not obviously refer to detainees who are being prosecuted for conduct that occurred prior to enactment?
First, I’m not a big fan of the presumption against retroactivity, or any presumption, when it is used to thwart apparent intent. But leave that to one side.
Take the following scenario.
A is in military custody at Time 1.
DTA is enacted at Time 2.
A files a petition for habeas corpus at Time 3. Obviously barred, even though he was in custody at the moment of enactment, right?
The majority is saying that the act does apply to persons in military custody at the time of enactment, when they file petitions subsequent to the Act. But if Detainee A had filed at time 1.5, the court would say that his suit is not covered by the DTA, because such a construction of 1005(e)(1) would make Section (h) an irrelevancy, which is not a favored way to construe a statute.
Which part of section (h)?
Section (h)(2).
Marghlar, I’m aware that the Court would say all that about (h)(2), because that’s basically what the Court did in fact say, and it’s still preposterous.
The Court’s theory is based on the notion that Congress had just as much reason to fear that (e)(1) wouldn’t be applied to pending cases as to fear that (e)(2) and (e)(3) wouldn’t be applied to pending cases. But that notion is false. Congress had no reason at all to fear that (e)(1) wouldn’t be applied to pending cases.
The exceptions to which (e)(1) expressly refers are expressly listed elsewhere in the DTA, and that listing does not include any exception for currently pending cases. Therefore expressio unius applies, and rules out any exception for currently pending cases.
You’re correct that that shouldn’t end the inquiry, because we also have to look at the temporal scope of (e)(1) itself to see if it might exclude currently pending cases. It doesn’t. The sort of language used in (e)(1) has ALWAYS been construed by the courts as applying to currently pending cases, absent a presumption of non-retroactivity. Such a presumption would be overcome in the present context, because (e)(1) explicitly applies to preenactment conduct (i.e. to aliens “currently in military custody”). Without a presumption of non-retroactivity, neither the Court nor anyone else has ever cited a single case where the words “shall have jurisdiction” were deemed to mean “shall gain jurisdiction over a new case,” as you have suggested.
In sum, Congress had every reason to expect that (e)(1) would be applied to pending cases, just like jurisdiction-stripping provisions have traditionally been applied to pending cases. As Justice Scalia put it, the Court “cannot cite a single case in the history of Anglo-American law (before today) in which a jurisdiction-stripping provision was denied immediate effect in pending cases, absent an explicit statutory reservation. By contrast, the cases granting such immediate effect are legion….”
Now contrast the jurisdiction-allocating paragraphs (e)(2) and (e)(3). No principle of expressio unius prevents exceptions to those two paragraphs from being implied or inferred, and that fact should have caused Congress concern that an exception for currently pending cases might be judicially implied or inferred. Indeed, as the Court mentioned at footnote 5 of its opinion in Hamdan, “The fact that courts often apply newly enacted jurisdiction-allocating statutes to pending cases merely evidences certain limited circumstances failing to meet the conditions for our generally applicable presumption against retroactivity….” (quotinga previous opinion). This fact too should have caused Congress concern that an exception for currently pending cases might be implied or inferred, due to a presumption against retroactivity. After all, neither (e)(2) nor (e)(3) explicitly applies to preenactment conduct, unlike (e)(1).
I think this is a slam-dunk against jurisdiction in this case. And let’s use some common sense here. Both (e)(2) and (e)(3) are exceptions to the rule of (e)(1). Who ever heard of a rule that does not apply in certain cases, but that has exceptions applicable in those very same cases. That would be VERY weird.
Don’t buy it. I don’t think it is an issue of extra-textual exceptions (which would be equally unauthorized in the case of e2 or e3).
Here’s the problem. The key language of e1, e2 and e3 is entirely parallel. The language is that the courts either “Shall not have jurisdiciton” or “shall have exclusive jurisdiction.” If this language is clear, it is equally clear in both cases. And if it is clear in e2 and e3, there was no reason for h2 to ever exist.
Congress clearly didn’t feel that it was clear, or there would have been no need to define a scope. Scalia says otherwise, but an analysis of the actual language and structure of the act makes his position unconvincing. The act as a whole evinces a clear understanding that the language, “shall have/not have jurisdiction,” is not considered clear in its scope. Scalia only reads differently by depending on presumptions established through case law which may be distinguishable, and regardless, the text and structure of the act are more important than the manner in which the court has previously interpreted other acts, with different texts and structures.
Finally, I’d note that the two sections granting exclusive jurisdiction in certain cases to the D.C. Cir. are not titled as exceptions. All that e1 says is “except as provided in 1005″; that need not mean that other sections of 1005 can’t have independent jurisdictional force. To me, the structure of the act seems to be quite clear: e1 completely denies jurisdiction in one class of cases. e2 and e3 create exclusive jurisdiction for the D.C. Cir. over a second, partially overlapping class of cases. To the extent that these principles conflict, e1 provides that e2 and e3 take precedence. That is perfectly consistent with the language and structure of the act, I think.
Marghlar, I agree that the word “shall” has the same basic meaning in each and every one of the forty (40) times it is used in the DTA. However, the basic meaning of that word does not determine whether it is being used in connection with pending cases as opposed to future cases (or in connection withany cases at all). Instead, we must look to the context in which the word “shall” is used. And, it’s not sufficient to merely look at the two or three words following the word “shall” in order to ascertain whether pending cases or future cases (or both or neither) are being discussed.
The context of the word “shall” in e1 is very different from the context of the word “shall” in e2. Just to mention one of many differences, the word shall is immediately preceded by the words “no court, justice, or judge” in e1, whereas the word “shall” is immediately preceded by the words “the United States Court of Appeals for the District of Columbia Circuit” in e2. Those prefatory words don’t determine whether pending cases (as opposed to future cases) are being discussed, nor do the two or three words following the word “shall” make that determination. The broader textual context must be considered.
For example, you have overlooked a plain textual distinction: neither e2 nor e3 explicitly applies to preenactment conduct, whereas e1 does explicitly apply to preenactment conduct. Therefore e2 and e3 would be much more susceptible to the Court’s longstanding presumption against retroactivity. Hence h2.
Additionally, you have overlooked another plain textual fact: e2 and e3 list exceptions to which e1 refers, and not vice versa. This is not a symmetrical situation, and therefore the text has non-symmetrical consequences. A list of express exceptions casts doubt upon — if not rules out — the existence of implied exceptions, per the expressio unius maxim. Therefore, h2 would be especially useful to rule out unwanted exceptions to e2 and e3, rather than e1.
Incidentally, implied exceptions are not necessarily “extra-textual,” as you suggested. But in any event, Congress had reason to fear unwanted exceptions to e2 and e3 that are implied, inferred, or otherwise conjured up. There are solid textual reasons to believe that Congress did not have those same reasons with respect to e1.
Andrew, at this point you are just repeating yourself, and making points I have already said I do not find persuasive.
I see nothing in the text of e1 as compared with e2 or e3 that suggests that it should be treated differently regarding pending v. future cases (see above).
LIkewise, I have told you why I don’t buy your expressio unius argument. Thus, your restating it and saying it again isn’t likely to convince me, absent something new.
I didn’t say that “implied exceptions” were extra-textual — I said I don’t like extra-textual exceptions, to the degree you were suggesting any. But I would note that the text of 1005e1 is part of the text of 1005, and hence, any “implied exceptions” (although I disagree that a determination of the scope of e1 constitutes an exception to it) would be within the texual reference of e1.
Thanks for a fun chat. This has really helped me refine my understanding of Hamdan and the DTA. I may still disagree with you (and you are welcome to raise new arguments that might change my mind), but I think the exchange has been very profitable.
Marghlar, sure, it’s been an interesting discussion. Happy July 4, by the way.
Unfortunately, you haven’t said one word as to why you find no significance in the fact that neither paragraphs e2 nor e3 explicitly applies to preenactment conduct, whereas e1 does explicitly apply to preenactment conduct. That’s a critical factor (according to the Court’s jurisprudence) in determining whether those paragraphs could or should be applied to pending cases (i.e. whether a presumption of non-retroactivity could or should be applied to those paragraphs).
If one concedes that no presumption of non-retroactivity should be applied to e1, then the only way that e1 could be limited to future cases only would be by conjuring up some inferred or unwritten exception to e1, and that is clearly barred by expressio unius, since e1 already refers to a list of exceptions that is provided by e2 and e3.
I really fail to see why you insist that a presumption of non-applicability to pending cases should be applied to e1. Without h2, such a presumption might well have been applied to e2 and e3, but not to e1 (e1 expressly applies to pre-enactment conduct). And, without h2, an exception for pending cases might well have been applied to e2 and e3, but not to e1 (e1 rules out exceptions beyond those expressly listed in e2 and e3).
I.e., the Court wanted jurisdiction, so it took it.
I see no express application to pre-enactment conduct — the provisions regard jurisdiction, not conduct. The examples you’ve adduced as a regulation of pre-enactment “conduct,” do not appear to prove that such a point is made.
I said as much several times above.
All the rest of your argument seems to flow from a premise that the sections are distinguishable based on their conduct. I still see no relevant distinction within the language.
Furthermore, please note that I have no desire to see any presumption of anything applied as a means of interpreting e1. Rather, I want the text interpreted for its plain meaning, in context of the surrounding language and structure of the act. I think the majority’s holding can stand fairly on this ground, and you haven’t convinced me otherwise.
If you want to change my mind, you’ll have to find either some affirmative indica of retroactive intent in e1 (what you’ve done so far is not persuasive), a relevant difference in the texts of e1 and e2-e3, or some explanation of h2 that makes logical sense while still according with your position. So far, you haven’t done so.
Marghlar, on Monday (yesterday) at 8:40 PM, I directed your attention to subparagraph (e)(2)(A) of 28 USC 2241, which says that the denial of jurisdiction applies to detainees “currently in military custody.” In other words, the denial of jurisdiction in 1005(e)(1) applies to preenactment conduct. Apparently, you’re denying that it does.
Nonetheless, 28 USC 2241(e)(2)(A) is manifest evidence of retroactive intent, it is also a manifestly relevant difference in the texts of e1 and e2-e3, and it therefore manifestly explains why h2 would apply to e2-e3 but not e1.
As I said, even if 28 USC 2241(e)(2)(A) did not exist, still the DTA contains no express exception for retroactive application, and an implied exception is ruled out by expressio unius for e1 but not for e2-e3.
I grant you that the DTA does not explicitly say, “Tony, David, Ruth, John, and Stephen, this Act requires you to refuse to hear or consider the Hamdan case until there’s been a final judgment from a military tribunal.” However, what the DTA does say is equivalent.
Here is additional proof (though none is really needed)….
Suppose that a military commission had rendered a final decision prior to the date of enactment of the DTA. Would e3 give jurisdiction to the DC Circuit to review that final decision? e3 suggests that the answer is “yes”:
“the United States Court of Appeals for the District of Columbia Circuit shall have exclusive jurisdiction to determine the validity of any final decision rendered pursuant to Military Commission Order No. 1, dated August 31, 2005…”
However, h2 says that the answer is actually “no’:
“[e3] shall apply with respect to any claim whose review is governed by [such paragraph] and that is pending on or after the date of the enactment of this Act.”
If the miltary commission reached its decision and adjourned prior to enactment of the DTA, then h2 say that e3 is inapplicable instead of applicable. So, h2 is not superfluous, and could not be superfluous regardless of how e1 is construed.
Andrew, I don’t read the military custody language to refer to pre-enactment conduct, as I’ve already explained. Since that language only took effect upon enactment, it is perfectly consistent for an act that takes effect at enactment to refer to conduct currently existing at the time of enactment (and thereafter).
As to your other point, I’d posit that it is likely that the “cases pending” language includes within the meaning of “pending,” “awaiting review on appeal.” It is an ambiguity, to be sure, but I think that your reading is less plausible.
Marghlar, the DTA was signed by the President and went into effect on December 30, 2005 thereby immediately amending 28 USC 2241 by adding a new section (e) which said that it applied to detainees “currently in military custody.” You’re apparently asking me to believe that new section (e) applied not to the conduct which landed those detainees in custody, but only to conduct which the detainees engage in after being detained. That is an extremely farfetched interpretation, and it’s rebutted by the language of section (e) itself, which bars jurisdiction over “ANY other action against the United States or its agents relating to ANY aspect of the detention by the Department of Defense of an alien at Guantanamo Bay, Cuba.” Surely, an action for wrongful imprisonment would be barred by section (e), and such an action plainly would involve preenactment conduct.
Regarding the scenario I outlined in which a military commission has imposed sentence on a detainee prior to enactment of the DTA, you’re apparently suggesting that e3 and h2 give exclusive jurisdiction to the DC Circuit to review that final decision, even though h2 says that the DC Circuit doesn’t have exclusive jurisdiction over claims unless those claims were “pending on or after the date of the enactment of this Act.” You seem to be saying that every claim governed by e3 is “pending on or after the date of the enactment of this Act.” That’s not plausible.
Surely, an action for wrongful imprisonment would be barred by section (e), and such an action plainly would involve preenactment conduct.
But that is an irrelevant point. The question isn’t whether any conduct that occurred pre-enactment might give rise to a suit, it’s when the courts have jurisdiciton to hear that suit. You need to find language that suggests, not just that some facts that occurred before enactment might be the subject of a suit, but rather, that Congress wanted to strip the courts of jurisdiciton of claims already pending, not just to strip courts of any new jurisdiction that would arise after the act was passed.
And no, I’m not saying that every possible claim is pending on or after the date of enactment. I’m saying that claims that are still undergoing an appeal are still pending. Obviously, claims that have been the subject of a fully-appealed judgment, and affirmed, are no longer pending.
Marghlar, regarding the scenario where a military commission rendered a final decision prior to the date of enactment of the DTA, let’s also suppose that executive branch review of that decision was completed before enactment, in accordance with Military Commission Order No. 1. In that scenario, h2 modifies e3 by cancelling exclusive jurisdiction for the DC Circuit. There’s no reason why Congress should have been just as concerned about including e1 in h2. e1 should be construed on its own merits, and its language is clear.
As you know, e1 also expressly covers at least some claims that involve preenactment conduct of detainees “currently in military custody.” This difference between e1 and e2-e3 is not irrelevant. If Congress enacted h2 partly in order to remove doubt about whether any preenactment conduct could be covered by e3, then Congress had no reason to use h2 in order to remove such doubts about e1.
There is nothing in the language of e1 that suggests an exception for cases pending at enactment, and h2 does not suggest one. On the contrary, e1 says to look elsewhere in the DTA for exceptions to e1, and those express exceptions elsewhere in the DTA do not include any exception for pending cases.
e1 says that “no court, justice, or judge shall have jurisdiction to hear or consider … habeas corpus … or … any other action….” This is broad language, and it ought to be construed on its own merits, because h2 is perfectly understandable without using it to narrow the meaning of e1. If the Court had done so, it would have arrived at a different conclusion.
Now you are just repeating yourself. I wasn’t persuaded by this line of argument before, and this is not changing anything. If you want to understand why I don’t agree, read my previous posts, which have certainly elaborated the points ad nauseum.
Marghlar, you previously posited that it’s likely that the “cases pending” in language in h2 includes within the meaning of “pending,” “awaiting review on appeal.” Therefore, I began my previous comment by saying, “let’s also suppose that executive branch review of that decision was completed.” That’s hardly repeating myself.
Anyway, I’m glad to drop this subject now. I don’t find your defense of the Hamdan decision persuausive, and I find the Court’s treatment of the jurisdiction issue outlandish. The Court disregarded the plain meaning of e1, by reaching out to another provision (h2) that didn’t even mention e1, and the Court proceeded to puff up that other provision (h2) into an exception to e1. In so doing, the Court overturned a 50-year-old precedent (Bruner) that had instructed Congress how to strip jurisdiction. I believe the Court also disregarded the maxim of expressio unius, disregarded the effect of h2 on cases no longer pending at enactment, and disregarded the language in e1 that applies e1 to preenactment conduct. But there’s no point in going on ad nauseum.