Image: Committee chairman Sen. Chuck Grassley (R-IA), ranking member Sen. Dianne Feinstein (D-CA), and Sen. Patrick Leahy (D-VT) talk with each other during a Senate Judiciary Committee hearing on firearm accessory regulation, Dec. 6, 2017 in Washington, DC. (Drew Angerer/Getty)

UPDATED 1/4:

A number of President Trump’s judicial nominations have collapsed in recent days after the candidates’ utter lack of qualifications was revealed, signaling (at a minimum) poor vetting on the part of the administration. The most stunning failure was Matthew Peterson, who became an internet sensation and meme generator when he could not answer even basic questions about U.S. law.  Also in the candidates’ graveyard is Jeff Mateer, who in 2015 stated that transgender children are part of “Satan’s plan.” Mateer also prophesied that same-sex marriage would lead to bestiality and lamented the banning by some states of harmful and bogus “conversion therapies” for gay and lesbian persons.

One additional problematic candidate is scheduled for a hearing this month: Howard C. Nielson, who has been nominated for the district court in the District of Utah. The intrepid Alliance for Justice (AFJ), which has worked assiduously to defeat several judicial nominees, has published a worrisome blog on Nielson calling into question his fitness to enjoy life tenure as a federal judge.  (AFJ normally focuses its work on appellate judges; there are so many problematic district court judges in play that it has expanded its vetting role to help Senate staffers and others evaluate candidates).

AFJ’s research reveals much about Nielson that is problematic, including his work on behalf of the National Rifle Association, his spurious motion to have a judge recused because of the latter’s sexual orientation, and his involvement in impermissible politicized and ideological hiring at the U.S. Department of Justice during the administration of George W. Bush.  (The fault-finding 2008 report by the DOJ’s Office of the Inspector General and Office of Professional Responsibility is here).

I raised a concern about Nielson’s fitness as a lawyer on these pages a year ago when I brought readers’ attention to a memo Nielson had written in 2005 while at the DOJ’s Office of Legal Counsel (OLC). Nielson’s memo is here in the OLC’s Freedom of Information Act (FOIA) reading room.  At the time he wrote the memo, Nielson worked under Stephen Bradbury, one of the infamous authors of the “torture memos,” which Senator John McCain has described as “permission slips” for torture.  Bradbury’s role providing legal cover for custodial abuses re-emerged during his confirmation hearings for the Department of Transportation. In signaling he would vote against Bradbury, McCain tweeted:

I will not support a nominee who justified the use of torture. Our enemies act without conscience—we must not…

Nielson earlier wrote to the Washington Post defending his ex-boss and essentially admitting to his own involvement in producing the torture memoranda.

In my previous post, I gave a close critique to Nielson’s own memo “to the file” setting forth a crazy (and dangerous) theory about the applicability of the Geneva Conventions that would also countenance the extraterritorial torture of civilians.

The argument gets rather technical (I’ve reproduced my full legal critique below), but the short of it is that Nielson argued that the Geneva Convention devoted to the protection of civilians in enemy custody or detention only applies to civilians held on U.S. territory.  That treaty (the 4th Geneva Convention governing international armed conflicts) requires signatories to treat all civilian detainees humanely and prohibits torture and other forms of mistreatment.  If Nielson’s theory of the treaty were to prevail, United States personnel could torture civilians—so long as they did so outside the United States—without breaching the treaty. Under the same reasoning, our adversaries could harm U.S. civilians in their custody, so long as the victims were not brought back to the territory of the belligerent in question.  In short, the memo advocates for the removal of any constraints on how treaty signatories (which now number all the states of the world) treat civilians outside their territories. This warped interpretation finds no support in international or domestic jurisprudence, the treaties’ drafting history, the treaties’ humanitarian object and purpose, or legal scholarship (even scholarship advancing conservative readings of the treaties).

Separate and apart from its flawed substance, the memo reveals exceedingly bad lawyering. Most importantly, like many of the Bush administration’s post-9/11 legal analyses, the memo contains results-driven reasoning that fails to engage (or even cite) all (or any of) the opposing precedent and contrary interpretations out there. This approach makes a mockery of the proper role of the OLC, which is to give candid, apolitical, and accurate legal advice to the White House. Furthermore, Nielson cites no support for his position in jurisprudence, scholarship, or the legislative history of the treaties.

In next week’s hearings, the Senate Judiciary Committee should ask Nielson pointed questions about:

  1. His role in the production of the suite of torture memos produced by the OLC, including those ostensibly authored by his boss.

  2. Whether he still stands by his opinion that the Geneva Conventions do not constrain how the United States treats civilians in its custody or control outside of the United States.

  3. Whether he still thinks the torture memos articulate valid legal advice for the White House.

Incidentally, this scenario reveals why memos such as Nielson’s should have been repudiated and withdrawn by OLC prior to the end of the Obama Administration, as I advocated.  As it stands, the memo remains in the file as a presumptively valid piece of legal analysis and a legally-available interpretation of the Fourth Geneva Convention.  Although it was not as salient as the other torture memos, which were rejected and withdrawn, Nielson’s memo could still be resurrected and put the United States’ treaty compliance—and the well-being of civilians in our custody—in jeopardy if implemented.

UPDATE: A further angle on this memorandum brought to my attention by someone in the State Department’s Legal Adviser’s office at the time this memo was written – why was it written “to the file” in the absence of any live legal controversy? And was it cleared through the ordinary inter-agency process? Was it written to the file precisely to avoid the clearance process (which clearly would have resulted in the rejection of this reading of the Conventions)?  During the first term of the Bush Administration, OLC had a practice of not sharing opinions on international law issues with State/L for comment. This practice apparently ended once Condoleeza Rice became Secretary of State.

My full analysis of the Nielson memo appears below.

The December 6, 2016 post:

The election of Donald Trump has triggered an anxious conversation about how President Obama can entrench some of his accomplishments before January 21, 2017.  Importantly, given the Trump campaign’s embrace of waterboarding and “worse”, President Obama would do well to ensure that he has fully dismantled the Bush Administration’s post-9/11 detention and interrogation system. A series of interlocking memoranda emanating from the Office of Legal Council (OLC) of the Bush Administration’s Department of Justice undergirded this system, although some of these memos were actually written after-the-fact to rubber-stamp decisions already made and conduct already underway.  Indeed, as legal ethicist Professor David Luban testified before the Senate Judiciary Committee, rather than providing sound, defensible, and balanced advice for policies under consideration, these memoranda were actually part of an executive branch cover-up for torture and other abusive tactics already underway:

[The memos] read as if they were reverse engineered to reach a pre-determined outcome: approval of waterboarding and the other CIA techniques.

All of these memos and more on the U.S. torture program are available in the Torture Database maintained by the ACLU, which won the public release of many of these memoranda through its dogged litigation, and in the OLC’s FOIA Reading Room

Many of the most problematic memoranda regarding executive power and authorizing conduct that constitutes torture or other forms of prohibited cruel treatment have been withdrawn by the OLC or invalidated by President Obama by way of Executive Order.  At least one, however, has escaped this deserved fate: an August 15, 2005, memorandum to the file entitled “Whether Persons Captured and Detained in Afghanistan are ‘Protected Persons’ under the Fourth Geneva Convention” (Memo).  This Memo argues that the Fourth Geneva Convention, which protects civilians in the hands of a state of which they are not nationals, applies only to individuals on U.S. territory—an interpretation that would significantly truncate, if not virtually gut, the protections of that treaty.  President Obama and the OLC should repudiate and withdraw this memorandum before the change of administration in order to ensure that there is no risk that the spurious reasoning it contains ever governs U.S. law-of-war practices in the future. This would confirm that Convention protects civilians in the hands of a party to which they are not nationals in all international armed conflict situations, including enemy territory.

The rest of this post sets forth the history of the politicized and impugned OLC memoranda; explains the origins of the Geneva Conventions and the scope of their protected persons regime; deconstructs and repudiates the reasoning in the Memo; and provides the correct interpretation of the Fourth Geneva Convention.

The History & Context

The OLC under President Bush became deeply politicized and, in the words of one commentator,

confused its role as a neutral expositor of the law with the role of legal policymaking.

That this once-independent institution had become politically-captured eventually came to light when a series of memoranda authorizing various forms of torture and other mistreatment were leaked to the press. In an almost unprecedented move, the Bush Administration’s OLC itself actually withdrew one of the worst examples of the OLC’s results-oriented reasoning: the infamous Bybee memorandum. Among other disgraceful arguments, that “legal opinion” sought to raise the bar on what conduct constitutes torture under U.S. law to cover only acts causing pain equivalent to “organ failure, impairment of bodily function, or even death.”  The OLC replaced this memo with a 2004 legal opinion, submitted to none other than James Comey. That text began with the bold and unequivocal statement that

Torture is abhorrent both to American law and values and to international norms.

It then proceeded to dismantle most of Bybee’s efforts to get around the long-standing ban on torture in international and domestic law.

Immediately upon taking office, President Obama escalated this process of recalling problematic OLC memoranda.  Executive Order (EO) 13491 (“Ensuring Lawful Interrogations”), dated January 22, 2009, revoked an earlier EO issued by President Bush in 2007 (EO 13440) that had asserted that members of al Qaeda, the Taliban, and their associated forces were not entitled to the protections provided under the Third Geneva Convention (concerning prisoners of war) and that the CIA Detention and Interrogation Program would satisfy Common Article 3 of the Geneva Conventions so long as the program complied with the U.S. laws on torture as interpreted by the OLC and Department of Defense.

President Obama’s 2009 EO also:

  • mandated that all detention centers and U.S. agents ensure that individuals subject to U.S. custody enjoy the protections of Common Article 3, at a minimum, and refrain from any interrogation technique not authorized by the applicable Army Field Manual (FM 2–3, Human Intelligence Collector Operations);
  • directed the CIA to close all detention facilities and refrain from opening any more in the future;
  • granted access to U.S. detention centers to the International Committee of the Red Cross (ICRC); and
  • asserted that no U.S. government employee or agent could rely upon any interpretation of the law governing interrogation issued by the Department of Justice between September 11, 2001, and January 20, 2009.

Meanwhile, the newly-constituted OLC also withdrew more of its prior memoranda. This January 15, 2009, memorandum, for example, confirmed that the OLC would no longer rely upon the legal reasoning contained in a number of Bush-era memoranda and set forth alternative reasoning that would henceforth govern law-of-war detentions and a range of other issues, including the exercise of the commander-in-chief power. Likewise, this 2009 memorandum formally withdrew four of the most notorious memoranda.

Notwithstanding this cascade of repudiations of Bush-era legal reasoning, there is one OLC memorandum on the concept of civilian “protected persons” that has been neither repudiated nor withdrawn.  President Obama and/or OLC should do so now to definitively reject the erroneous legal reasoning and conclusions contained therein.

The Memorandum Left Behind

This 2005 Memorandum—written by Deputy Assistant Attorney General Howard Nielson (who was associated with politicized hiring at the Department of Justice) “for the files”—ostensibly addressed the legal status of civilians (vice combatants) captured or detained in Afghanistan and the territorial reach of the Fourth Geneva Convention Relative to the Treatment of Civilian Persons in Time of War (GCIV).  Oddly, the Memo is dated 2005, long after the conflict in Afghanistan had shifted from an international to a non-international armed conflict following the fall of the Taliban in late 2001. Thus, at the time the Memo was written, the applicability of the Geneva Conventions to the conflict in Afghanistan was reduced considerably since the bulk of their provisions address international armed conflicts (IACs) rather than non-international armed conflicts (NIACs)—with Common Article 3 being a notable exception.  This suggests that the Memo was not really meant to govern operations in Afghanistan, but was rather laying the argumentative groundwork for future IACs.

The Memo takes as its premise that the Geneva Conventions apply to the conflict with the Taliban, as determined by President Bush, but then concludes that individuals “in the hands of” the United States would not benefit from the protections contained in GCIV on the specious theory that those protections do not apply beyond U.S. territory, even if the United States were engaged in an extraterritorial armed conflict.  This conclusion is reminiscent of the reasoning around extraterritoriality that has undermined the United States’ adherence to the human rights treaties it has ratified (see our coverage of the extraterritorial reach of the Convention Against Torture, the International Covenant on Civil and Political Rights, and other human rights treaties). Although the United States’ arguments around extraterritoriality are weak when it comes to human rights treaties—which by their own terms govern the treatment by a state of those individuals on its “territory” and those individuals subject to its “jurisdiction”—the reasoning in this 2005 OLC memorandum is even more tenuous given the text, history, structure, and object/purpose of the Geneva Conventions and of GCIV in particular.

Background on the 1949 Geneva Conventions & The Centrality of the Concept of “Protected Persons”

By way of background, the four Geneva Conventions (GCs) were finalized in 1949 with the intention of setting forth rules to protect various vulnerable groups from the consequences of war.  Accordingly, each of the four Geneva Conventions protects a different class of “protected persons”: the wounded and the sick (GCI), the shipwrecked (GCII), prisoners of war (GCIII), and civilians (GCIV).  Collectively, they apply from the outset of any conflict between High Contracting Parties (HCPs), i.e. to IACs, and to situations of occupation.  Prior law-of-war treaties, such as the 1899 and 1907 Hague Convention Regulations, were concerned only with the treatment of combatants and had, in the words of the ICRC’s commentary, proven

insufficient in view of … the problems relating to the protection of civilians in enemy territory and in occupied territories.

GCIV defines its category of “protected persons” as those individuals

who, at a given moment and in any manner whatsoever, find themselves, in case of a conflict or occupation, in the hands of a Party to the conflict or Occupying Power of which they are not nationals.

The concept of “occupation” has a wider meaning within GCIV than it does in other law of war treaties, including the Hague Conventions mentioned above.  Indeed, the commentary makes clear that:

The relations between the civilian population of a territory and troops advancing into that territory, whether fighting or not, are governed by the present Convention. There is no intermediate period between what might be termed the invasion phase and the inauguration of a stable regime of occupation. Even a patrol which penetrates into enemy territory without any intention of staying there must respect the Conventions in its dealings with the civilians it meets. …The same thing is true of raids made into enemy territory or on [its] coasts. … Thus troops advancing into enemy territory cannot under any circumstances execute a civilian without trial, no matter what crime he has committed. The person in question must be tried and sentenced in accordance with Article 64 and the Articles which follow it. … (Emphasis added).

Some groups of civilians are excluded from GCIV’s definition of “protected persons”, including

  • “nationals of a State that is not bound by the Convention” (which is now a null set given universal ratification of the GCs);
  • nationals of a neutral State; and
  • nationals of a co-belligerent state so long as normal diplomatic relations exist between the states in question.

The theory behind these latter two exceptions is that the person’s home state will be in a position to provide diplomatic protection for its nationals caught up in the armed conflict (and presumably offering a more powerful defense than the GCs). Thus, as explained by the ICRC commentary:

The definition [of civilian] has been put in a negative form; as it is intended to cover anyone who is not a national of the Party to the conflict or Occupying Power in whose hands he is. The Convention thus remains faithful to a recognized principle of international law: it does not interfere in a State’s relations with its own nationals.

Had the drafters wanted to, they could easily have added “persons in unoccupied enemy territory” to this list of non-protected persons.  Needless-to-say, they did not precisely because it was intended that such persons benefit from the terms of the treaty.

In addition, IHL provides that if a person is protected by GCI – GCIII, his or treatment is governed by those regimes rather than GCIV, which operates as a safety net to ensure no one is without status or protection.  As the ICRC’s commentary states:

[e]very person in enemy hands must have some status under international law: he is either a prisoner of war and, as such, covered by the Third Convention, a civilian covered by the Fourth Convention, or again, a member of the medical personnel of the armed forces who is covered by the First Convention. There is no intermediate status; nobody in enemy hands can be outside the law.

It continues:

The Convention is quite definite on this point: all persons who find themselves in the hands of a Party to the conflict or an Occupying Power of which they are not nationals are protected persons. No loophole is left.

This concept of blanket coverage, it should be noted, is embraced by the U.S. Army Field Manual, FM 27-10, ¶¶73, 247 (1956).

By contrast, the 2005 Memo would leave persons in unoccupied enemy territory entirely without protections under the GCs—in other words, in a legal black hole—which was precisely the opposite result that GCIV is designed to achieve.

The Fourth Geneva Convention Regime of Protection

GCIV is divided into four main parts of relevance to this discussion:

Part I—devoted to general provisions.

Part II—setting forth protections for the general population against certain consequences of war, which

cover the whole of the populations of the countries in conflict [not just protected persons], without any adverse distinction based, in particular, on race, nationality, religion or political opinion, and are intended to alleviate the sufferings caused by war.

Part III—enumerating rules for the treatment of protected persons generally and in occupied territory. Rights to be accorded to all protected persons include:

This Part also contains rules governing certain security measures, subject to various protections, that may be utilized against protected persons who pose a threat to the HCP, including rules governing an internment regime.

Another set of rules in this Part govern occupied territory, which has not been at issue in either Afghanistan or Iraq.

Finally, Part IV—containing the enforcement provisions, including the extension of universal jurisdiction over breaches.

So, Where Does The 2005 OLC Memo Part Go Astray?

The Memo spills a lot of ink arguing the uncontroversial point that the treaty’s various provisions apply primarily to two classes of persons:  persons in the hands of HCP of which they are not nationals in “the territory of a party to the conflict” and persons in occupied territory.  The memo’s reasoning and interpretation of the treaty is largely accurate to this point. (Technically, Part II of GCIV applies to the entire population of the HCPs to the conflict, not just to individuals who would qualify as “protected person”).

But then the real mischief begins.

The memo goes on to pose what it identifies as an interpretive dilemma (p. 8):

[t]o say that ‘protected person’ status is available to persons finding themselves ‘in the hands of a Party to the conflict’ in ‘the territory of a party to the conflict,’ however, does not fully resolve the status of persons captured by a party to the conflict in territory belonging to its enemy. (Emphasis added).

It then argues that the phrase “territory of the party to the conflict” might be read to mean only the home territory of the party to the conflict in whose hands a person finds him- or herself (p. 8).  Note the linguistic sleight of hand – “the territory of a party to the conflict” (the treaty language) suddenly becomes “the territory of the party to the conflict” (the memo language).  Note also the speculative reasoning without any grounding in the treaty’s commentary.

Although the memo acknowledges that the treaty might logically be read to mean that GCIV applies to any civilian in the hands of any Party to the conflict, even outside that Party’s territory, it the endeavors to refute this plain language (and also the standard reading) of the treaty by pulling snippets from the treaty’s drafting history and parsing treaty references that “address[] governmental actions that would typically occur in a state’s domestic territory, rather than foreign enemy territory,” such as rules governing detention for security purposes or providing for gainful employment (see above). It also gets all tangled up in structural arguments about potential divisions of labor between various sub-parts of the treaty.  Even if some of these legal artifacts lend some support the memo’s ultimate conclusion, the memo does not engage in a good faith interpretation of the treaty, as required by the Vienna Convention on the Law of Treaties. Nor does it even acknowledge the commentary’s unequivocal explanation that GCIV applies as soon as troops of one HCP come into contact with civilians on enemy territory (see quotations above) whether during a raid or a full-scale invasion.

The Memo suggests that the only treaties regulating unoccupied enemy territory are the Hague Regulations Respecting the Laws and Customs of War on Land, which contains rules governing prisoners of war and the conduct of hostilities (e.g., rules on weapons causing superfluous injury, attacks on undefended localities, spies, flags of truce, etc.).  These rules would not apply as a matter of treaty law because Afghanistan has not ratified them and they contain a si omnes clause rendering them applicable only if all states involved in the conflict are parties. (Many of their provisions constitute customary international law, but they are not addressed to the treatment of civilians in the power of a party to the conflict, so offer no substitute for GCIV).

The Implications of the Memo

The impact of this interpretation would be that the United States (and other HCPs) would be bound by GCIV only with respect to actions taken on the territory of the United States (or territory that qualifies as occupied under GCIV).  Left out is unoccupied enemy territory.  Thus, in situations like Afghanistan and Iraq, where the United States was engaged in an IAC wholly on the territory of another HCP, the United States would have no GCIV obligations towards persons in its custody (or otherwise “in its hands” in the lexicon of the treaty) unless those territories were under formal occupation.  As a result, if the United States were to detain Afghan or Iraqi civilians, those individuals would not benefit from GCIV’s dictates about humane treatment, non-discrimination, etc. unless they were brought back to the United States.  Although the memo was ostensibly written with Afghanistan in mind, its strained interpretation would not easily apply to multilateral conflicts, where states are operating within the boundaries of multiple states, such as in World War II—the set of conflicts that inspired the promulgation of the Geneva Conventions in the first place.

As outlined above, GCIV contains dozens of rules protecting civilians who find themselves “in the hands of a Party to the conflict … of which they are not nationals.” The point of many of these provisions is to regulate the treatment of individuals who would not otherwise enjoy diplomatic protection from their state of nationality and who would thus be vulnerable to mistreatment if they were in the hands of a party to which they were not nationals. If the memorandum’s reasoning is correct, these individuals would benefit from no GCIV protections until they were “in the hands of” a HCP on its own home territory.

Because embattled states are unlikely to wholesale import large swaths of an enemy civilian population, the impact of GCs would be significantly diminished and would apply in only the narrowest of circumstances.  Indeed, were a HCP to forcibly bring enemy civilians into its own territory, it might actually constitute the unlawful transfer or confinement of protected persons (as noted by the commentary), which constitute grave breaches of the treaty subject to universal criminal jurisdiction. The OLC Memo also overlooks Article 1 of GCIV, which clearly states that

The High Contracting Parties undertake to respect and to ensure respect for the present Convention in all circumstances. (Emphasis added).

Add the Memo to the Legal Scrapheap

The 2005 Nielson Memorandum should be withdrawn because its conclusions are inconsistent with the text, the original intent of the treaty, and the prevailing legal interpretation of the GCs.  The Memo should also be withdrawn because it runs counter to the OLC’s normal practice, which is to

render[] formal opinions addressed to particular policy proposals and not undertak[e] a general survey of a broad area of the law or addressing general or amorphous hypothetical scenarios involving difficult questions of law.

Given that the Afghan conflict was no longer an IAC, the Memo is clearly unmoored from any live policy proposal.

The OLC should repudiate this Memo itself or President Obama should do so by way of Executive Order.  Indeed, before President Obama leaves office, the DOJ would carefully review its records to ensure that no specious vestiges of Bush-era legal reasoning remain in the files.