by Michael Dorf
In Fallback Law, a 2007 article in the Columbia Law Review, I explored (among other things) the limits, if any, on coercive fallback laws. A fallback law is a provision of a law that states that, in the event some provision of a law is held invalid, a substitute provision (i.e., a fallback) will take its place. Congress and state legislatures sometimes include such substitutive fallbacks and, even when they don't, they usually implicitly or explicitly provide for a de facto fallback in the form of the original law minus the invalid part--i.e., they provide for severability. The point of both substitutive fallback provisions and severability is to ensure that the inclusion in a statute of one constitutionally questionable piece does not doom the whole law in the event that the courts strike down the questionable piece. Substitutive fallbacks and severability clauses aim to preserve as much of the legislative output as possible. By contrast, a coercive fallback provision is designed to frighten judges into upholding laws they would otherwise strike down by threatening them with dire consequences.
When I conducted the research that culminated in Fallback Law, I found a fair number of substitutive fallback provisions but virtually no coercive fallbacks. I was able to find and discuss exactly one real-world example of an arguable coercive fallback (beginning at p. 329 of the article), involving a California law that seemed aimed to coerce the courts into upholding an "impact fee" on cars owned by non-Californians against a dormant Commerce Clause challenge. However, even in this example, it was possible to see the law as not exactly a coercive fallback, and so most of my discussion of coercive fallbacks involved hypothetical cases. For example, I imagined a federal statute that cut funding for parachutes for members of the Air Force in the event that the courts invalidated (the then-still-operative) Don't-Ask-Don't-Tell policy. (I referred to this example in a different context earlier this year.)
Because of the hypothetical character of nearly all of my discussion of coercive fallbacks, a recent Emory Law Review article by Yale Law School PhD candidate Eric Fish dismisses my conclusion that coercive fallbacks can be invalid in virtue of their coerciveness as unrealistic. Fish also writes that "[n]othing in the Constitution protects any of the three branches from bullying," but this pronouncement is overstated. It's true that nothing in the Constitution explicitly protects any of the three branches from bullying but then nothing in the Constitution explicitly protects any of the states from bullying by Congress; yet the Supreme Court has held that Congress may not use its conditional funding power coercively. As I argue in Fallback Law using that example and others, there are sound reasons to conclude that the Constitution implicitly protects the judiciary from some kinds of bullying by the other branches.
Although I am unpersuaded by Fish's rejection of my substantive argument, I nonetheless have some sympathy for his view that political considerations will generally prevent coercive fallbacks. But generally doesn't mean always. Which brings us to Kansas.
As reported last week, Kansas Governor Sam Brownback recently signed into law a coercive fallback that would eliminate funding for the state judiciary if the courts invalidate another provision of Kansas law that reorganizes the Kansas judiciary in a way that is itself seen as hostile to the state supreme court because of its earlier ruling on education finance. The coercive fallback is thinly disguised as a non-severability clause (Sec. 29) of the funding measure, but once one traces the references, it is easy to see through the charade. The legislature and the governor have told the Kansas courts that they will lose their funding if they rule as they otherwise think they ought to.
Is the Kansas coercive fallback invalid? I think so. Just as we may infer a federal constitutional principle of judicial independence from Article III's life tenure and salary protection, as well as due process, so we might look to the Kansas Constitution for similar principles. Although Kansas judges lack life tenure, they do have some salary protection (Art. III, Sec. 13) and are forbidden from engaging in political activity (Art. III, Sec. 8). In addition, the federal Constitution's 14th Amendment Due Process Clause entails an impartial adjudicator (see cases like Goldberg v. Kelly), and an adjudicator who is worried about losing funding is not impartial. Perhaps Sec. 18 of the Kansas Bill of Rights, which guarantees a "remedy by due course of law, and justice administered without delay" for persons suffering injuries, could be given similar effect.
The best argument against finding coercive fallbacks invalid rests on the assumption that a legislature can do via fallback whatever it can do directly. By hypothesis, Kansas could cut funding to the courts (subject to limits I'll assume away for this analysis), so why can't Kansas do so through a fallback?
The short answer draws an analogy to blackmail. To use an example currently in the news, "Individual A" had a right to go public with his accusation against Dennis Hastert. But threatening to go public unless Hastert paid Individual A hush money amounted to blackmail.
Some legal scholars question the legitimacy of blackmail prohibitions. How can it be wrong to threaten to do something that one has a legal right to do? I don't have an all-purpose answer, but it does strike me that in the current context the harm is obvious. Although a state legislature is entitled to cut funding to the courts, it cannot exercise that power in a manner that violates constitutional norms. It can't cut funding to the courts for cases involving members of a particular race, for example. And it can't cut funding in a way that violates other constitutional norms, such as those ensuring adjudication by an impartial adjudicator.
In Fallback Law I nonetheless puzzled over the question whether making a threat in a law was distinctively problematic. To put it slightly differently, did my argument prove too much? If it is impermissibly coercive to threaten to cut funding to the courts in a fallback law (or inseverability clause), then wouldn't it also be impermissibly coercive for politicians merely to announce, in advance, that if the courts rule in a certain way, they'll be made to pay a price?
I conclude in the article that there is a distinction. Putting the coercive measure in the law, rather than merely announcing an intent to punish the judiciary, makes the threat both more credible and easier to detect after the fact. So, even assuming that the Kansas legislature and governor could lawfully carry out a threat to cut funding after a decision by the state courts, requiring them actually to do so makes them pay a political price, which in turn reduces the likelihood that they will do it. A coercive fallback leaves no doubt that the legislature aims to coerce the courts and eases the political pressure on the elected officials, because they vote for it at a time when it looks like the threat may not have to be carried out.
Accordingly, my analysis of the Kansas real-world example of a coercive fallback ends up pretty much where I was with respect to my hypothetical examples. Truth may not always be stranger than fiction but it's often as ridiculous.