“Federalism was our Nation's own discovery,” Justice Anthony Kennedy wrote in 1995. “The Framers split the atom of sovereignty. It was the genius of their idea that our citizens would have two political capacities, one state and one federal, each protected from incursion by the other.”
Splitting atoms is often a mistake; the Framers’ “genius” idea eventually shattered their antebellum Republic over the matter of slavery. Today, a century-and-a-half after that debacle, the federal-state boundary is so poorly marked that it’s the site of constant guerrilla legal warfare. Nobody ever quite wins or loses; in legislatures and courts, the war goes on.
In those skirmishes, the most striking theme is pervasive hypocrisy. In fact, the idea of “principled federalism” is, with the exception of a few lonely oracles in faculty lounges, slightly rarer than Sasquatch. From James Madison to Jeff Sessions, political leaders and judges routinely defend broad federal power when their party controls the federal government, and sweeping state authority when it does not.
Case in point: United States v. California, a massive immigration lawsuit filed last week by the Trump administration. As civil conflict goes, it’s not quite firing on Fort Sumter, but it does signal all-out legal war.
California is one of America’s most diverse states, and its elected leaders are among the bluest of America’s blue politicians. They made it known early in 2017 that, Trump or no Trump, California would not be a part of a massive new wave of detentions and deportations by the federal division of Immigration and Customs Enforcement. As a result, the California Assembly enacted:
- Assembly Bill 103, which concerns federal immigration detention. It forbids states and localities from entering new contracts (or expanding existing ones) with Immigration and Customs Enforcement to house immigrant detainees, and requires state inspection of state, local, or even private facilities that are currently housing detainees. The inspection is to assess not only the conditions of detention, but also the “due process provided” to inmates and “the circumstances around their apprehension and transfer to the facility.”
- Assembly Bill 450, which governs private employers. It bars them from admitting ICE inspectors to “nonpublic” areas of a jobsite unless the inspectors have “a subpoena or court order,” thus impeding ICE’s surprise workplace raids. It also requires employers to notify their workers when ICE asks to inspect a company’s employment records—records that may reveal that some employees are in the U.S. without permission. Finally, it provides that employers (who must by federal law verify a worker’s eligibility to work at the beginning of employment) may not re-verify the same worker later.
- Senate Bill 54, the “California Values Act,” which governs state and local law enforcement. State and local agencies are sharply limited from assisting directly in federal immigration enforcement; in addition, in most cases, they may not ask individuals for their immigration status; honor ICE “detainers” requesting local jails to hold individuals so that ICE can pick them up; provide personal information or addresses of prisoners to ICE (unless the information is already public); notify ICE of the scheduled release of most prisoners; or transfer most prisoners to ICE unless required to do so by a “judicial warrant or judicial probable cause determination.” (The bill makes an exception: state and local authorities can notify and transfer when releasing prisoners convicted of a designated set of serious crimes.)
Leave aside whether these are good ideas. The question is how far California can go in disassociating itself from federal immigration enforcement, and where disassociation, which is often permitted, turns into obstruction, which is not. The search for an answer begins with two Sibylline constitutional texts—the “Supremacy Clause” and the Tenth Amendment—and one even more puzzling statute, 8 U.S.C. 1373.
Article VI § 2 of the Constitution provides that the Constitution and valid federal statutes “shall be the supreme Law of the Land … anything in the Constitution or Laws of any State to the Contrary notwithstanding.” This language has given rise to a doctrine called “preemption,” which provides that, when using its enumerated powers, Congress “preempt” state laws, rendering them null. “Preemption” happens in three ways; first, Congress can write a valid statute that explicitly nullifies state laws on the subject; second, Congress can create a scheme of regulation so detailed that it has “occupied the field,” leaving no room for states; and, third, courts can invalidate state laws that conflict with or obstruct valid federal laws.
On the state side of the equation, the Tenth Amendment provides that “the powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.” The states have an undefined mass of “police powers,” most definitely including control over their own law-enforcement apparatus. The court has also enunciated the “anti-commandeering principle”: Even when acting within its proper powers, the federal government cannot reach down and require state legislators or executive officials to carry out federal programs.
In other words, under the Constitution, relations between state and federal governments focus on what each party cannot do to the other, rather than the reverse. Congress can pass laws, and the feds can enforce them, but they can’t make the state governments enforce them; states, conversely, can refuse to enforce federal law, but can’t stop the feds from doing so in their territory.
This dance of federalism is thus a little like the strange playground version of “Simon Says” I played as a kid, in which the “Simon” player issued instructions containing double and triple negatives to confuse the players about whether to step forward or stop. (“Do not not step forward—not!”) That weird negative standoff gives rise to the third significant text in this case, 8 U.S.C. § 1373: State and local governments and officials “may not prohibit, or in any way restrict, any government entity or official from sending to, or receiving from [ICE] information regarding the citizenship or immigration status, lawful or unlawful, of any individual.”
Can the federal government require states to turn over information they have? Probably not; that would be commandeering. Can the states bar the federal government from getting information that’s otherwise available? Almost certainly not; that anti-federal discrimination would trigger pre-emption. But is saying, “States shall not have a policy of not doing something because not doing it obstructs federal law” really just a sneaky way of saying, “You shall have a policy of doing that thing because doing it helps federal enforcement”? The constitutionality of 1373 is not yet settled.
Within that framework, how do California’s three bills play out? The state can certainly ban its own facilities from contracting with ICE to detain employees; the state almost certainly can’t bar private facilities from doing the same. Can it require inspections of state facilities being used for federal immigration purposes? Probably. What about private facilities? The state can generally supervise or inspect public or private facilities—jails, hospitals, nursing homes—but how sharply, if at all, can it specifically target private facilities that deal with ICE? There are reasons to do so—these private detention centers are rife with abuses—but the U.S. argues the inspections are intrusive and obstructive.
What about private employers? Can a state tell a business that it may not voluntarily cooperate with ICE? Federal law provides for voluntary cooperation, but California bans it. Federal law allows for inspection of employment records without notice to workers; California now requires notification. Federal law encourages use of the federal work-eligibility system; California law now restricts it. California doesn’t ban any conduct that federal law requires—but it takes the “voluntary” out of the rest. Some of this seems obstructive.
Finally, the state can certainly forbid its law-enforcement personnel from gathering immigration information from those they encounter. Based on the police power, that’s a good-faith law-enforcement strategy grounded in the idea that victims and witnesses should be able to interact with police without fearing deportation. The state can also bar its personnel from participating in ICE enforcement. A federal requirement that they participate would be classic “commandeering.” Refusing ICE “detainers” also seems pretty clearly within state power—these aren’t judicial warrants, they are phrased as requests, and they often seek detention of individuals who are otherwise constitutionally entitled to release.
Can the state refuse to turn over information about prisoners? A general policy of not turning over prisoner release dates might survive review—as long as it restricted only information that the state generally keeps confidential from the public. What, though, about a policy that gives ICE the release dates of only prisoners whom the state deems a threat? That seems awfully close to constructing an independent state immigration policy in this area, contradicting federal priorities.
The U.S. also claims that information about, let’s say, the home address or prison release of a detainee can’t be withheld under any circumstances. The reason? It argues that, under § 1373, these facts are “information regarding the citizenship or immigration status” of the individual. Indeed, almost anything about an alien that ICE wants to detain would, in this view, be covered by § 1373. California responds that, in fact, the language of the statute is much narrower—that “regarding the citizenship or immigration status” literally means whether the individual is a citizen and, if not, the individual’s immigration status. Location and duration of detention, physical location, home address, etc., is not “regarding,” the state argues.
Reading the statute as broadly as the U.S. does would seem, to me, to raise serious constitutional issues; some immigration authorities would argue the reverse—that a narrow reading would allow intentional state obstruction of federal policy.
Everyone agrees that the case will turn powerfully on the language of Arizona v. United States, the 2012 Supreme Court decision that invalidated most of Arizona’s aggressively anti-immigrant S.B. 1070. Arizona v. United States engaged the same emotions the new case does, but in reverse: A state government was seeking to regulate and even punish undocumented immigrants in ways that the federal government regarded as counter-productive.
Immigrant-rights advocates sought a broad reading of federal power, in order to forestall xenophobic measures like 1070. Restrictionists, by contrast, rejected the idea that states could not monitor and detain the undocumented without federal permission. The court’s opinion is generally seen as a win for immigrants’ rights; the majority struck down most of 1070 because, it said, the states were trying to enforce federal law more strictly than the federal government wanted to. That, the majority said, would strain limited federal resources, run the risk of creating conflict with foreign governments, and subvert the executive’s flexibility to set immigration policy. But the majority also strained to read parts of Arizona’s statute in a way that wouldn’t require striking it down. So it was a split decision.
What about a state that wants less, not more, enforcement? The majority’s language could be adapted to target that policy as well: By authorizing state officers to decide whether an alien should be detained for being removable, § 6 violates the principle that the removal process is entrusted to the discretion of the Federal Government. A decision on removability requires a determination whether it is appropriate to allow a foreign national to continue living in the United States. Decisions of this nature touch on foreign relations and must be made with one voice.
If the federal government wants to be more aggressive than a state, a court could reason, that decision may also “touch on foreign relations” and need “one voice.”
Who will win? Some commentators have predicted a romp for California, but that seems far-fetched. The case seems destined to make new law, probably at the Supreme Court level. If I were betting, I’d wager on a split decision like the one in Arizona: Some provisions will be upheld, and others struck down. And the decision will tell you a lot more about the attitude of the judges than about the spongy doctrines of federalism. The majority so far hasn’t shown a lot of skepticism about Trump-era immigration enforcement.
But the Arizona case is not the only precedent worth considering. Is there a difference between Arizona’s statute—which was explicitly aimed at making life so hard for the undocumented that they would leave the state—and California’s, which is designed to protect its undocumented population? Is there a difference between the Obama approach to immigration—which aimed chiefly at a set of “criminal aliens” and sought to leave the law-abiding in peace—and the Trump approach, which invades the states to treat individuals as illegitimate on the basis of status alone, that affords them lesser rights than others, that tears peaceable families apart and sends parents or children away from their homes, perhaps never to return? The current federal policy is, as historian Eric Foner wrote last year, eerily reminiscent of the decade-long battle over the Fugitive Slave Act, in which the anti-slavery movement defended “state’s rights” and the slave South insisted on total federal supremacy.
Jeff Sessions has, late in life, discovered federal power; California Attorney General Xavier Becerra is now a state’s rights man. For all the solemn rodomontade about the Framers and the atom of sovereignty, in the end, federalism battles are always really about something else.
We want to hear what you think about this article. Submit a letter to the editor or write to firstname.lastname@example.org.