Wednesday, April 08, 2015

How RFRA is Like a Taking and Two Thoughts on Establishment

by Michael Dorf
Most of my latest Verdict column considers one of the two main objections that was lodged against the pre-amended Indiana RFRA and the failed version of the Arkansas RFRA bill that drew so much criticism last week. They were criticized on the ground that they would have applied in private litigation. I explain why a simple-minded version of this criticism misses the mark but why there are also more subtle versions that have more to recommend them. Here I want to focus on the second of the more subtle versions.

The core of the argument I describe in the column is that under a RFRA (or under a state constitutional provision protecting free exercise that is construed to grant protection against laws of general application, not just laws targeting religion), people whose religious exercise is substantially burdened should be accommodated but only when doing so does not impose the cost of the accommodation on particular third parties. Under this principle, a RFRA would just about never apply in private litigation because private litigation necessarily singles out particular individuals--the parties resisting the claims for religious exceptions--as those who must bear the cost of the accommodation.

Let me illustrate with a couple of examples that are far afield from anti-discrimination law. Suppose that a state has a RFRA and also has a general law reserving the use of certain state highways for motor vehicles. The law specifically forbids pedestrians, people on horseback, horse-drawn carriages, and bicycles. Now suppose that an Amish citizen of the state, whom I'll call Yoder, rides his bicycle on one of these state highways. A state trooper gives Yoder a ticket and Yoder goes to traffic court, where, if convicted, he would face a fine of $100. Yoder pleads the state RFRA in his defense. Does he prevail?

Maybe. Yoder would first have to show that keeping him off the state highway substantially burdens the exercise of his religion. Let's suppose he can get over that hurdle--perhaps because this state highway is the only road between his onion farm and the market where he sells the onions that he transports on the bicycle; keeping his bicycle off the highway would make him choose between violating his religion (by driving a car) and giving up his livelihood. The burden claim here is at least as strong as in Hobby Lobby.

The government would nonetheless be able to overcome Yoder's RFRA claim if the bicycle prohibition is narrowly tailored to advance a compelling state interest. Highway safety is undoubtedly a compelling interest but whether the law is narrowly tailored would depend on a number of factual issues. For my purposes, let's suppose that the law is not considered narrowly tailored because it turns out that the presence of Amish bicyclists on the state highways does not increase the aggregate risk on those highways. Or at least the government is unable to show otherwise, as is its burden under the state RFRA. Accordingly, the state court finds that Yoder does not have to pay the fine.

Now suppose that another Amish onion farmer, whom I'll call Miller, also rides his bicycle on the state highway. Miller is a safe bicyclist but he has the bad luck to hit a pothole that causes his onions to spill into the road, in turn causing a passing car to crash into the median. The driver of the car, whom I'll call Driver, sustains injuries and his car is damaged. Driver sues Miller in tort and argues that Miller's presence on the state highway on a bicycle is negligence per se. Let's suppose that a prior precedent involving a non-religious bicyclist holds that violating the no-bicycles-on-the-highway law is negligence per se in a tort suit. Can Miller defend based on RFRA?

My intuition here is that Miller has a much weaker case than Yoder. Admittedly, part of the basis for that intuition is that Miller actually caused injury, whereas Yoder did not, but I don't think that's the whole of it. In addition, allowing Miller to invoke RFRA means imposing the cost of his exception entirely on one innocent third party: Driver.

And that seems unfair in much the way that it seems unfair for the government to impose the cost of some socially beneficial project--building a needed highway or sewage treatment plant, say--on the people who happen to own the property that the government needs for the projects. If this is a collective good, then the government should pursue it by raising revenue from the general public, not by uncompensated exactions from the particular people who happen to have the bad luck to own the particular parcels of property that the government wants to use.

To be clear, I'm not saying that the Takings analogy is perfect. If it were, we might want to say that in a case like Driver's, Miller can implead the government, so that Driver would get his compensation from the general public. I don't think I would go that far. However, I do think that the anti-singling-out logic of Takings law underwrites the broader view that particular individuals should not bear the burden of religious accommodations.

As long as I'm analogizing to other constitutional provisions, I'd also add that imposing the cost of affording religious accommodations on particular individuals who are not part of the accommodated religious group implicates anti-Establishment values. I don't go so far as Justice Stevens, who thought that RFRA as a whole violates the Establishment Clause, but I do think that it is especially problematic for the state to make particular individuals bear the whole burden of accommodating other people's religious opt-outs, because that begins to look like a kind of coercion of non-believers into following, or at least facilitating, the practices of believers. Indeed, understanding the issue as one of coercion suggests there is a Free Exercise issue here as well: A RFRA applied in private litigation can undermine the freedom to choose non-religion or a different religion of the party against whom the RFRA claim is asserted.

Thinking about how the burden of providing accommodations is distributed also may provide some insight into the politics of RFRA. The original impetus for RFRA was bipartisan but in the last few years it has split ideologically, with conservatives supporting RFRA(s) and liberals wary. Partly that's a result of the particular contexts in which the claims arise: Liberals value contraceptive access (as in Hobby Lobby) and anti-discrimination protection for LGBT persons more highly than conservatives do.

But in addition, as others have noted, although the federal RFRA was inspired by a case involving a minority religion, RFRAs lately been used by people who adhere to conservative branches of mainstream religions. As the number of people who seek eligibility for a RFRA exemption increases--or as it increases in particular socially conservative communities--the burden of providing accommodations becomes more concentrated. Thus, even in a case where RFRA is being applied in litigation with the state, the cost of the accommodation now falls on a non-believing subset of the population who are either themselves a minority or, if a majority, a bare one--at least in particular communities. Accordingly, instead of virtually the whole community absorbing the marginally small cost of providing an accommodation to a handful of people with idiosyncratic religious beliefs, in the new contexts a RFRA begins to look like it enables one large faction of the population impose its religious beliefs on the rest of the population. And so liberals, who hold anti-Establishment values much more strongly than conservatives do, have another reason to be wary of RFRAs.


Joe said...

The connection between the two examples (Yoder? name seems a bit familiar*) to me is that if you allow bikes on the highway, it increases the chances of harms to third parties. More accidents akin to letting people play in the street but only penalizing them if they get hit by a car there. So, line drawing issues.

I'm sympathetic to free exercise claims more than some I agree with on certain issues, but am wary about it being applied evenhandedly. So, certain blogs discuss how the birth control mandate is a burden on religion.

But, I suggest, doesn't it also help by broadening the choices of those who let's say want to have marital intimacy (important to them as part of a religious sacrament) w/o having a child they aren't ready for (something they again see as having religious significance). The idea to some people simply seems a bit absurd.

How abortion is selectively handled shows this too. There is a special rule in place (Hyde Amendments) to address the conscience concerns of certain people. This at some point to me is an Establishment Clause concern.

But, I also respect free exercise values and think Smith was wrongly decided. Too broad RFRAs to me is not a good compromise & I simply don't believe it will be evenly applied. Thus, e.g., "sure blacks need to be served, but well gays? you know. kinda different." Right.


* Wisconsin v. Yoder. It also is a popular Amish name.

Hashim said...

But even when the Govt enforces the law, it's often doing so for the benefit of specific private parties, not just the general "public" at large. For example, in a situation where a business refuses to serve a customer in violation of anti-discrimination law, what conceivable difference does it make whether that law happens to be enforced through a private right of action, a public action by state officials, or both?

And more to the point, since state law chose to create a private right of action rather than allow only public enforcement, it seems perfectly appropriate for the state to revoke that private right of action when it interferes with what a state deems to be a more important state interest -- namely, religious freedom.

Finally, I suppose you could respond that RFRA also shouldn't apply to public enforcement that seeks to protect specific private parties. But it's hard to understand why the identifiable nature of the private party matters, since *virtually all* laws seek to benefit private parties and some of them will inevitably be harmed ex post if an accommodation is granted, even if they're not identifiable ex ante: the mere fact that alternatives are available is no guarantee that the alternatives will be adopted; and private remedies will often be under-compensatory for a variety of reasons. To use your initial hypos, if the State itself can't enforce its no-bikes-on-highway rule, it might not adopt an alternative safety measure, some Drivers will get hurt, and state tort law might not allow (or might cap) certain types of damages (and at a minimum likely doesn't cover attys' fees). So specific individuals will of course bear the brunt of the accommodation, but not applying RFRA against the State in these circumstances would essentially mean that RFRA never applies, since you can *always* tell this type of story.

Michael C. Dorf said...

Joe: Yoder and Miller are indeed the respondents in Wisconsin v Yoder.

Hash: As I say in the column, there are many instances in which allowing a RFRA claim in litigation between a religious claimant and the state does indeed burden identifiable third parties, and I say in the column that if so, that may be a reason not to allow the RFRA claim to succeed there (as Justice Alito hints at but does not decide in Hobby Lobby). But if one takes that view then one will also note that RFRA claims in private litigation virtually ALWAYS burden identifiable third parties and therefore one could categorically forbid them.

Now, you say that all laws are for the benefit of private parties but I think that's false in the relevant sense. Take a law forbidding certain kinds of air pollution, the point of which is to benefit the public as a whole. Granting a religious exception to someone whose religious ritual requires the release into the atmosphere of an otherwise banned pollutant burdens the public generally, or, if you prefer, burdens every member of the public equally. (Let's imagine that the pollutant quickly diffuses over an area the size of the state so that it affects everyone in the state more or less equally). Another, more prosaic example would be where affording the religious claimant an exception adds to the administrative cost of government by some dollar amount, which will be marginally felt by all taxpayers. The point is that sometimes the cost of the exception will be distributed roughly evenly, while in private litigation it almost never will be. One still might conclude that RFRA ought to apply in private litigation because religious freedom is more important than other policy goals. The point of my column and post was simply to explicate the intuition that it ought not apply in private litigation.

Joe said...

I appreciate btw that Prof. Dorf provides audio for his columns though wonder why only he does it.

Hashim said...

Fair enough Mike, though I think in the real world most cases involve laws that do not even roughly benefit all members of the public equally. Certainly that's not the case for the paradigmatic type of RFRA exemption -- a zoning variance, which burdens the surrounding property owners more than others. Also not true for RLUIPA, where the burden of accommodation will be felt disproportionately by other prisoners, not by the public at large.

egarber said...

I am curious about government entanglement with religion in this context. Is that still a binding doctrine, stemming from Lemon? And if so, is there not a point where determining undue burden case-by-case invokes it, all the way to the conclusion that RFRA laws might be unconstitutional?

Joe said...

Concern about entanglement is weaker these days [e.g., Mitchell v. Helms] though Scalia did in Oregon v. Smith cite how problematic:

"horrible to contemplate that federal judges will regularly balance against the importance of general laws the significance of religious practice"

He also opposed reliance on some test where the courts would try to determine what is "central" to one's religious beliefs. He noted in an oral argument he was against RFRA but guess he in effect washed his hands of the matter. Oregon v. Smith recognized but accepts religious exemptions might in some fashion favor certain religions.

Joe said...
This comment has been removed by the author.
Michael Yuri said...

Professor Dorf, I'm curious how you would respond to some slightly different hypotheticals:

1) This time, suppose that the state does not have a RFRA. It has a general law reserving the use of certain state highways for motor vehicles, but this law has an explicit statutory exemption to the bicycle prohibition for persons who cannot travel by car for religious reasons.

Now the same onion incident occurs with Miller & Driver. Can Miller defend against the negligence per se precedent on the basis of the statutory exemption?

2) Now, suppose instead of a religious exemption, the bicycle prohibition has a grandfather provision for one particular stretch of road that exempts anyone who regularly used that particular road for bicycle commuting prior to its conversion into a major highway five years ago.

Miller has been regularly commuting on this road by bicycle for 15 years. The same onion incident occurs with Miller & Driver. Can Miller defend against the negligence per se precedent on the basis of the grandfather exemption?

3) Finally, same scenario as #1, except the statutory exemption applies to persons who can't travel by car for reasons of religion *or conscience* (precedent has applied this exemption to an atheist environmentalist who was conscientiously opposed to the use of fossil fuels).

Miller, who is Amish, is involved in the same onion incident. Can Miller defend against the negligence per se precedent on the basis of the statutory exemption?

Michael C. Dorf said...

Michael: I pose the hypos around here! More seriously, I think that your examples mostly pose the question of what the negligence per se standard does.