There's a lot to say about Peter Margulies' reply to my and Kevin Heller's criticisms of the "former government officials'" amicus brief in al Bahlul--the military commission appeal currently pending before the en banc D.C. Circuit, where the central question is whether the commission lawfully had jurisdiction to try and convict Bahlul of "conspiracy." To put it plainly, Peter's brief concedes that the commission could not validly try conspiracy as a standalone offense, but nevertheless argues that the D.C. Circuit can affirm the conviction because Bahlul could have been tried and convicted for a completed war crime on a "joint criminal enterprise" (JCE) theory of liability. And as his response to me and Kevin suggests, Peter's basic defense of our criticisms is that the "errors" we've identified in this regard can all be dismissed as "harmless." Below the fold, I offer three independent reasons why Peter's response only serves to dig his hole deeper--and why case law, statutory text, and common sense preclude the after-the-fact bait-and-switch for which the new amicus brief argues. And in the interest of moving this conversation forward, I raise three specific questions for Peter to answer if and when he so chooses. I. The "Harmless Error" Doctrine Requires... You Know... Error! Let's start with first principles. The "harmless error" doctrine is a doctrine about trial-court errors that deprive the defendant of his constitutional rights. What, exactly, was the "error" in Bahlul's case that Peter now argues was "harmless"? As Kevin has shown, and as relevant here, all that the trial judge did in Bahlul's case was simply accept the government's own charging decision at face value--including its decision to omit any reference to JCE. Thus, as I explained in my initial post, the real "error" the brief identifies is with the government's (deliberate) charging decision. The amicus brief now seeks to have that prosecutorial choice set aside on a post-conviction appeal in favor of a charge and theory of liability that not only were not submitted to the jury, but were explicitly eschewed by the prosecution--and, thus, by the court. That's not "harmless error" if for no other reason than because there's no judicial "error" in the first place. And so, here's my first question for Peter: What is the specific trial-court error that, in your view, is subject to the "harmless error" doctrine on Bahlul's post-conviction appeal? II. Wrong-Offense, Wrong-Theory Rulings Are Never "Harmless" Maybe what Peter is trying to argue, albeit elliptically, is that even if the trial court committed no error, it would be harmless error now for the D.C. Circuit to affirm Bahlul's conviction on a ground other than that submitted to and found by the jury... Leaving aside the silliness of urging an appellate court to commit error on the ground that it would be harmless, there's a larger problem here: regardless of who committed the error, wrong-offense, wrong-theory appeals are not properly subject to the "harmless error" doctrine in the first place. I won't bore readers with the voluminous jurisprudence on the difference between "harmless" and "structural" errors in criminal cases; suffice it to say that the Supreme Court has always held that errors that “affect the framework within which the trial proceeds,” and are not “simply an error in the trial process itself" fall into the latter camp--even as it has expanded the scope of the former. On Peter's own theory, there was no flaw in the trial process in Bahlul's case. Instead, there was a flaw in the framework within which the trial proceeded--one in which Bahlul was charged with committing conspiracy as a standalone offense, as opposed to one in which he was charged with a separate, completed war crime under a JCE theory of liability. Thus, insofar as there was an error in Bahlul's case, it was a framework error, not a process error. But lest it seem like I am splitting hairs, the larger point here remains the more important one: There are serious ex post facto, due process, and jury-trial problems with affirming a defendant's conviction on an entirely different offense than that actually found by the jury. Peter seeks to blur this distinction by invoking the Supreme Court's 1999 decision in Neder v. United States. Neder was, as Peter describes, a "harmless error" case. But the distinctions between Neder and Bahlul are both obvious and decisive: There, the error was that the trial court itself, rather than the jury, adjudicated what turned out to be one of the elements of the charged offense. That is to say, Neder was convicted on the charged offense; the error was merely in who found which elements. As Chief Justice Rehnquist explained, "an instruction that omits an element of the offense does not necessarily render a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence." Needless to say, that's a far cry from a case in which the "guilt" that was determined was for an entirely different offense than the one on which the conviction is urged to be affirmed. And so, this leads to my second question for Peter: Can you identify a single Supreme Court case holding that it was harmless error for a conviction to be affirmed based upon a charged offense and a theory of liability not submitted to--or found by--the jury? III. Oh--and About those Pesky Merits: Why It's Not Clear the MCA Would Even Allow Peter's Theory Finally, without getting too deeply into the back-and-forth between Peter and Kevin, let me say hello to the elephant in the room: the Military Commissions Act of 2006. As relevant here, in Bahlul, the government only charged and convicted Bahlul of the standalone offense of conspiracy, as delineated in 10 U.S.C. § 950t(29). The argument at the core of the amicus brief is that the government could instead have prosecuted Bahlul for one of the other offenses listed in § 950t, under a JCE mode of liability, without running into the ex post facto issues at the heart of the D.C. Circuit's decision in Hamdan II. But neither the amicus brief nor Peter's reply identifies (1) which other charges under § 950t could have been charged; (2) why it is so clear that § 950q allows for JCE liability, as opposed to aiding-and-abetting; or, most importantly, (3) whether the jury in Bahlul's case actually found the requisite elements necessary to establish JCE in Bahlul's case. But let's put aside those pesky merits, because I suspect they'll prove too much of a distraction. There's also an enormous procedural problem: Under the MCA, a change in theory requires statutorily prescribed notice, and probably also requires re-charging and the approval of the new charges by the Convening Authority. (Even in the 9/11 case, the government has had to pursue these steps.) Neither Peter's brief nor his reply explains why an appellate court could so easily circumvent these rules in the interest of affirming a conviction on grounds different from those submitted to the jury. To put it bluntly, Peter is in effect asking the D.C. Circuit to not only correct the prosecution's strategic charging decision in a manner that case law prohibits, but to do so in outright defiance of the procedural and substantive rules Congress has imposed upon the commissions themselves. So here's question # 3: Do the MCA's procedural constraints with regard to notice and approval of charges not apply to the D.C. Circuit?
* * *
Part of me understands where Peter is coming from: If the government had charged Bahlul with a completed war crime -- and if it had submitted facts tending to establish his culpability under a JCE "mode of liability" -- then maybe the jury would have found the requisite elements and convicted Bahlul for what is, in effect, the same offense. If all that had happened, the story goes, the Hamdan II issue would almost certainly be moot in Bahlul's case. I'm not sure I agree, but I at least get the point. But post-conviction appeals are not the time to seek such re-writings of history. As I noted in my initial post and above, it looks like the government is now pursuing in the 9/11 case the exact course of action that Peter thinks it should have followed in Bahlul's case. But the fact that it didn't do so there is not Bahlul's fault or the fault of the trial judge--and it's not the province of an appellate court to excuse the government from the consequences of its conscious, strategic litigation choices.