Breaking News

Textualism and open questions about S.B. 1070

The following is an essay for our symposium on Arizona v. United States by Roderick Hills, Jr., William T. Comfort, III Professor of Law at NYU. Professor Hills teaches Constitutional law, among other courses, and writes on federalism, local government law, and other areas relating to decentralization of governmental power.


In Arizona v. United States, Arizona claims the power to enforce the federal Immigration & Naturalization Act (“INA”) by performing background checks of aliens’ immigrant status, arresting aliens believed to be removable from the United States under federal law, requiring aliens to carry federally specified ID, and barring aliens from seeking employment when they are ineligible to do so under federal law.  In response, the United States declares that the proffered state assistance wastes federal resources and endangers national foreign relations.  In effect, the case pits the Article II concept of the unitary executive against a state’s assertion of federalism. If its recent decisions are any guide, however, the Court will probably evade these big issues by pretending that the “plain” text of the INA resolves the dispute.  Both for the sake of legal predictability and intellectual honesty, it would be better if the Court dropped the pretense that the INA’s text really settles the case and instead confronted the deeper structural question of whether the states or the President should receive deference in construing immigration law.

  1. Does the plain text of the INA settle the case?

In Chamber of Commerce v. Whiting, the Court rested its decision against preempting Arizona’s sanctions of employers who employ undocumented aliens on the theory that the term “licensing and similar laws” in the Immigration Reform and Control Act of 1986 (“IRCA”) unambiguously included Arizona’s suspension of the right to do business in the state.  Whiting lightly side-stepped textual ambiguity by scoffing at, but not answering, Justice Breyer’s complaint that the majority’s literalistic reading of “licensing … laws” covered an absurd swathe of state legislation, from dog-licensing ordinances to laws regulating the marriage license.  The Court briskly observed in a footnote that Arizona’s assertion of power did not involve any such extravagant claim of power to de-license its citizens from owning dogs or marrying, conveniently ignoring Justice Breyer’s point that, if the term “licensing” could be narrowed extra-textually to exclude marriage licensing, then it could also be narrowed to exclude licensing of businesses unrelated to employment.

The INA provides the raw material for the Court to attempt the same faux textual resolution of the dispute in Arizona v. United States.  The INA requires federal authorities to respond to “inquiries” by non-federal officials about the immigration status of any individual (8 U.S.C. § 1373(c)) and allow “communication” between non-federal and federal officials about any individual’s immigration status (8 U.S.C. § 1644).  These provisions might be cited to uphold the command of S.B. 1070 (Arizona’s challenged statute) that state and local law enforcement officers perform background checks for otherwise rightfully detained persons.  The INA also generally authorizes non-federal governments to “cooperate with the Attorney General” in immigration enforcement even absent any specific “287(g) agreement” between ICE and a state or local agency (8 U.S.C. §  1357(g)(10)).  The Court might emphasize that the plain meaning of “cooperat[ion]” implies mutual agreement by both sides of the cooperative bargain and strike down S.B. 1070’s authorization for state and local officials to arrest aliens when the Attorney General objects.  This last argument needs a little help from the expressio unius canon:  The Court would have to hold that Section 1357(g)(10)’s specific authorization for state “cooperat[ion]” implicitly excludes all other inherent or implied powers that states might otherwise possess to enforce federal law.

This sort of cut-and-dried textual argument would resolve the case (and incidentally, eliminate most of the recent efforts by states to enforce the INA outside of “Section 287(g) agreements”).  It would, however, hardly be a satisfying way of justifying the outcome.  Arizona v. United States calls for a more expansive exposition of the role of the states and the President’s role in implementing federal immigration law.

Consider, for instance, the following three questions that such a textual analysis ignores.

  1. 1. Do states have inherent power to enforce immigration law?

Arizona and its amici repeatedly have invoked the idea that states have inherent power to enforce federal immigration law, an idea most controversially associated with an OLC memo written by Jay Bybee in 2002.  Does such a power exist, and what is its scope?  Even the proponents of the “inherent state power” agree that it is not co-extensive with the immigration laws:  No one asserts, for instance, that states have the power physically to deport aliens from their territory.  But why exclude this concededly extravagant assertion of state power, if state officials have the general right to enforce the federal statute?  One could argue that the states’ inherent power must give way to the extent that Congress has provided a specific enforcement mechanism – say, deportation hearings before an Immigration Judge – by which a federal statute is to be carried out.  But then it is not obvious why, under this reasoning, the ICE’s power to perform traffic stops and arrest aliens lacking proper ID should not likewise preempt the inherent powers short of deportation that Arizona claims.

These conundra could all be avoided by the simple, crisp position that immigration law, narrowly defined as the business of excluding and removing unauthorized aliens from the national territory, is an exclusively federal field in which non-federal governments play a role only by statutory grace rather than by virtue of their reserved police powers.  But such a view has weaknesses of its own.  If the federal agencies fail to enforce immigration laws effectively, then exclusion of state enforcement prevents states from filling the gap left by federal incompetence.  By contrast, if federal law could eliminate such state gap-filling only by expressly prohibiting some sort of inherent state power, then the effort to amend federal law to include such a prohibition could help put the spotlight on the gap between brave statutory words and inadequate budgetary resources.

In short, both the recognition and exclusion of the ill-defined inherent state power have costs.  Neither position is foreclosed by precedent.  It is high time, therefore, for the Court to rule one way or another on the existence of such a power.

  1. 2. Does any presumption against preemption apply to states’ enforcement of federal law?

Suppose that the states do have an inherent power to enforce federal immigration law where federal law is silent: Should federal law be construed with an eye towards minimizing the preemption of this power?  This presumption against preemption has generally been regarded as protecting the “historic police powers of the state,” but the Court has generally assumed that the power to enforce federal law is not included among these “historic” powers.  As Justice Scalia noted for the majority in AT&T v. Iowa Utilities Bd., “[i]f there is any ‘presumption’ applicable to this question, it should arise from the fact that a federal program administered by 50 independent state agencies is surpassing strange,” such that “appeals …to what might loosely be called ‘States’ rights’ are most peculiar….”  The Court in Buckman v. Plaintiffs’ Legal Committee has rejected the idea that state tort law can impose duties on companies not to commit fraud against federal agencies during the rule-making process by noting that “the relationship between a federal agency and the entity it regulates is inherently federal in character because the relationship originates from, is governed by, and terminates according to federal law.”

Arizona and its amici, however, asserted before the Ninth Circuit that the Court ought to “start with the assumption that the historic police powers of the States were not to be superseded by the Federal Act unless that was the clear and manifest purpose of Congress.” Putting aside whether there is something especially national about immigration law, the Court should clarify whether there is something especially national about national law, such that state claims to implement national law should be viewed skeptically and recognized, if at all, only if they have an unambiguous home in a specific clause of a federal statute.

  1. Should the President receive deference for interpreting federal immigration law to exclude unsupervised state implementation?

The United States has asserted that Arizona’s unsolicited assistance endangers America’s foreign policy, wastes federal resources, and interferes with ICE’s enforcement priorities.  How much of an administrative record does the federal government need to develop in order to obtain judicial deference for these assertions?

In Wyeth v. Levine, the Court rejected the FDA’s assertions about how state tort law frustrated federal objectives of protecting consumer safety by noting that these assertions had not been tested through public deliberation of notice-and-comment rule-making, had shifted over time without explanation, and lacked the “thoroughness, consistency, and persuasiveness” necessary for either Chevron or Skidmore deference.  Like the FDA’s claims in Wyeth, the assertions of the United States are not backed by data tested through notice-and-comment rule-making but rather by affidavits of various federal officials cited in briefs.  Moreover, because the United States filed a facial challenge to S.B. 1070 before the state law was ever enforced, there is a paucity of information about the state law’s effects or even its meaning.

Yet Wyeth is not entirely germane to immigration law, if only because the states’ power to carry out federal statutes might be more tangential to states’ interests than the states’ powers to preserve their own common law.  Moreover, requiring the feds to make APA-tested findings before they can oust state intermeddling with the INA could eat up more resources than state enforcement could ever save.  One could imagine a scenario in which judges afford special deference to federal agencies’ baleful predictions about the consequences of state enforcement citing (say) the President’s special role in foreign affairs or the Department of Homeland Security’s special expertise regarding national security.  On the other hand, if the federal agencies are deliberately neglecting the enforcement of immigration law, either because of incompetence or ideology, then the same agencies might use their own ipse dixit to exclude state supplementary efforts that highlight federal inadequacy or contradict the federal executive’s agenda.

In short, the degree to which the courts should force the federal agencies to produce specific process-tested data as a condition of preempting state enforcement efforts turns on general judicial presumptions of trustworthiness.  Do the courts generally trust the federal agencies to carry out federal law, using state officials when their employment is cost-justified?  Or do the courts generally believe that forcing federal agencies to parry state enforcement with specific data will prod the feds into taking a careful look at federal shortcomings?  These sorts of general presumptions about institutional competence, not the fly-specking of dictionary definitions and intrinsic canons, will likely drive the result in Arizona v. United States, whatever the Court might officially declare in its opinion(s).  The only question is whether the relevant presumptions will remain hidden behind the machinery of textualism or see the light of day in the Court’s published reasons.

Recommended Citation: Roderick Hills, Textualism and open questions about S.B. 1070, SCOTUSblog (Jul. 12, 2011, 2:24 PM),