On Legislative Restraint and Religious Accommodation: Replies on Chapter 6

Apologies for the delay in posting my reply to Chad’s really rich comments on Chapter 6 of my book. I decided to delay until after Thanksgiving, as I thought people might be more likely to read the post. But Chad’s arguments were also sufficiently challenging that it took me awhile to figure out how to respond, and I’m not entirely happy with my responses at the moment.

I’ll address Chad’s concerns about judicial restraint, legislative restraint, my general approach to religious accommodation and my take on key accommodation-related court cases in that order.

I. On Judicial Restraint

Chad, in contrast to almost everyone in the public reason literature, worries that I impose too many restraints on judges and legislators. For instance, I think a consensus requirement applies to judges, in part because their reasoning becomes the explicit basis for future law, such that the reasoning itself should be mutually acceptable. But Chad has concerns:

It seems to me that the best Kevin could get is that judges should not appeal to reasons in their decisions if they justifiably believe that such appeal will lead to unjustified laws in the future. Such a restriction, it seems, would not systematically call for or restrict shareable or unshared reasons.

Chad suggests that judges might be permitted to support laws for unshared reasons as long as they reasonably believe their reasoning won’t be the basis for future law. I’m willing to admit that exception in principle, but given how judicial reasoning can be revived decades and even centuries after the fact, I think it is hard for judges to reasonably believe their arguments will not be misused (though they can be misused even if they’re shared). At least if the reasons are shared, they are more likely to support publicly justified law in the future, given that many diverse reasons are unlikely to become shared reasons over time.

However, Chad has led me to see something I missed before, namely that if judges worry enough about the future, the shared reasons they appeal to today might not be shared in the future. You might think about the shared reasoning supporting freedom of contract in Lochner that became extremely controversial a few decades later as an example. So perhaps judges should only be concerned about the use of their reasoning within the foreseeable future. So perhaps there is less reason for judges to hew to shared reasoning than I thought, though it still seems to me plausible that sticking to shared reasoning is a better long-term strategy for judges to promote publicly justified law. But I lack an argument for this hunch.

II. On Legislative Restraint

I argue that legislators can act on whatever reasons they like so long as they believe the proposals they support can be publicly justified to multiple points of view. Chad worries that this excessively restricts the flexibility politicians need to support some proposals in order to promote others. Chad writes,

I wonder if Kevin has reason to think politicians must be more uncompromising than voters or if he would amend his principle to allow these tradeoffs.

I think legislators do need to fairly uncompromising in not promoting publicly unjustifed laws, but I could probably build in a clearer condition where a politicians could support publicly defeated law X in order to promote publicly justified law Y that is more important to members of the public.

Chad also worries that legislators might be entitled to promote bills that they know will never pass:

What objection could Kevin have to, say, Bernie Sanders introducing and voting for a bill, presumably unjustified if passed, to implement socialism knowing that he is the only avowed socialist in Congress and that the bill is doomed to fail?

I didn’t specify this in the book, but I think that legislative restraint applies to politicians if they have a non-trivial chance of affecting legal outcomes, not merely if the probability is low. Perhaps, for instance, legislative restriant kicks in when an unjustified bill has a 1/1000 chance of passing. But there are surely some laws whose chance of passing is much lower, so perhaps Sanders’ socialism bill could pass through.

III. On Accommodation Generally

Chad raises some very important questions about how to conceive of religious exemptions due to the complex nature of defeater reasons. A law is defeated when a suitably idealized member of the public prefers no law to the law in question, but in some cases where it appears that the member of the public prefers no law to the law in question, she in fact just finds the law very, very sub-optimal – just barely better than nothing. A few times in the chapter I don’t draw this distinction clearly enough. In fact, when the book was going to press, Chad and Jerry Gaus convinced me that I needed to make this distinction clearer, so I wrote an additional paper called “How to Publicly Justify Religious Exemptions” that I could share with readers, if you like. I like to think I’ve addressed Chad’s worries there.

IV. On the Cases

In Chapter 6, I apply my approach to religious exemptions to three famous cases of religious accommodation: Yoder, Smith and Mozert. Chad has some reasonable concerns about my approach. In general Chad worries that I’m not clear enough about which law the objectors have defeaters for. I think my implicit thought in the chapter is that they’re exempted even from fine-grained characterizations of the law and so exempt from more coarse-grained characterizations of the same law. But I now realize this isn’t quite right: objections to coarse-grained characterizations of a law might be less effective, as they’d undermine more potentially positive aspects of the more coarse-grained law. Things get complicated fast in this area, I think!

With respect to Mozert, I think more needs to be said about how to individuate laws in that case. In Mozert, the question was whether they would be excused from class, so the exemption request was in fact very narrow. What part of the law did the Mozert parents have defeaters for? The unit of coercion characterized as “the power of the schoolboard to require that all children in their school district be educated in accord with textbook X” or perhaps the unit of coercion characterized as “the power of the schoolboard to require that all children in their school district be educated in accord with some reasonable range of textbooks” or perhaps even “the power of the schoolboard to set the terms of public education.”

But here’s another tricky issue: the Mozert parents don’t have defeaters that undermine the justification of the law for everyone to whom the law applies. They may merely have defeaters for the application of the law to themselves, in which case their defeaters for more coarse-grained characterizations of the law would not undermine the potential benefits of the coarse-grained version of the law that accrue to non-objectors.

So perhaps I can say this: the Mozert parents have defeaters for fine-grained characterization of the law in question, and for all the more coarse-grained characterizations, so long as their defeaters are understood as defeaters that exempt them from the law rather than defeating the law for everyone. Does that work? Perhaps.

With respect to my Smith discussion, Chad can’t see which law Smith and Black have defeaters for. In my view, they have publicly justified entitlement to unemployment compensation because they did not break a publicly justified law (namely the law prohibiting narcotic use in religious ceremonies, but perhaps that’s only part of a law). And the entitlement, by assumption, is based in a publicly justified law. So I could have been clearer that the defeaters relevant to assessing the Smith case are a bit more complicated than in Yoder or Mozert.

Chad’s comments on my work typically leave me with more insights than I had before I received and worked through them, so I thank him for pressing me to be clearer.

This entry was posted in Discussion, Politics, Posts, Reading Group. Bookmark the permalink.

5 Responses to On Legislative Restraint and Religious Accommodation: Replies on Chapter 6

  1. Many thanks go to Kevin offering those very helpful clarifications. It is good to know, for instance, that his view makes significant concessions for legislators pursuing legislation they know has extremely little chance of going through. I will note that I think Kevin makes a huge concession for practice when he above writes: “a politicians could support publicly defeated law X in order to promote publicly justified law Y that is more important to members of the public.” This seems to mean that we could see legislators directly voting, even as the decisive vote, for legislation they justifiably believe to be publicly defeated, and yet this may not violate Kevin’s principles if the legislator was, as is very often the case in real politics, only supporting that policy for strategic reasons to really support some other legislation (at least if that legislation is not defeated too). That is important enough for me to draw attention too, but not the main thing I wish to address.

    Regarding judicial restraint, I should correct Kevin on a minor point. He puts my view thus: “Chad suggests that judges might be permitted to support laws for unshared reasons as long as they reasonably believe their reasoning won’t be the basis for future law.” That is not quite right. The view I put forward is that judges should be able to appeal to whatever reasoning as long as they justifiably believe it will not be used as the basis for unjustified laws. Some reasoning might not be shared, but also not particularly prone to support unjustified laws. For instance, certain religious or philosophic doctrines might not be shared but also not particularly pernicious (e.g., appealing to the claim that people are “endowed by their Creator with certain inalienable rights”). Likewise, some reasons will be unshared but systematically support the elimination of laws rather than their creation, so systematically bring about more of the default liberty without generally producing defeated laws.

    Moreover, if Kevin wants to say that judges cannot appeal to reason R is R might be used as the basis of a defeated future law, then he does not have sufficient reason to allow shared reasons. As Kevin knows, since it is central to his project, many laws supported by the small set of shared reason are nonetheless publicly defeated. This defeat is most obvious when you have idiosyncratic defeaters, as from the religious convictions Kevin writes about, but also come readily enough from differences in the way people weigh the diverse shared values. Thus, as in my example in my original commentary, the appeal to the importance of equality in one decision may become the basis for future decisions that generate unjust laws. If judges really cannot appeal to reasons that may create defeated laws, then, strikingly, they may not be able to appeal to any reasons at all.

    It seems to me that it is actually too strict to say that they cannot appeal to reasons that might at some point be used to support defeated justification. We have to allow, as I think Kevin was getting at with the consideration of time, that some defeated laws may at some point result. I think minimally we should say that the judges are allowed, as Kevin allows others, to essentially consider tradeoffs. If the reasons can be expected to systematically produce a lot of defeated decisions, then they should not appeal to those reasons. If, on the other hand, the judge is justified in believing that the reasons would systematically produce justified laws, though might in some cases produce defeated laws, then the judge should be able to appeal to that reason. I think this will not generally even pick out reasons that are permissible or impermissible, but will actually depend on the kind of case being considered and the way the reason is appealed to in that case.

    The position I just defended does not necessarily require excluding unshared reasons or permitting shared reasons. If Kevin thinks his own proposed restriction to shared reasons will be effective at promoting justified law, that seems to depend upon further significant empirical claims that have to be backed up. I have doubts that such empirical claims would pan out, but I invite Kevin to make the case for thinking that policy would be effective in practice despite the fact that shared reasons are neither necessary nor sufficient conceptually for the task.

    Regarding the Mozert case, first, I think Kevin mischaracterizes the law in question, since he puts it in terms of “the power of the schoolboard to require….” But let us leave that aside, for at least that could be reworded in various ways and the real issue regards how to construe the scope and extent of the law.

    I cannot understand a part of Kevin’s reply that seems particularly important. He writes: “the Mozert parents don’t have defeaters that undermine the justification of the law for everyone to whom the law applies. They may merely have defeaters for the application of the law to themselves, in which case their defeaters for more coarse-grained characterizations of the law would not undermine the potential benefits of the coarse-grained version of the law that accrue to non-objectors.” Let me separate out two issues. First, there is the question of how to understand a law when it is defeated. In general terms, we might say that a law is publicly justified when every member of the public to whom the law applies has sufficient reason to endorse it, publicly defeated when at least some member to whom the law applies, and justified or defeated relative to some agent depending only on whether or not that agent has sufficient reason to endorse it. We might think that, for instance, when it is revealed that the Yoder family has sufficient defeaters for compulsory high school then the law is revealed to be publicly defeated. Even if the law remains in place, however, we might think that there is nothing wrong with continuing to enforce the law against those to whom it is justified. All of that is well enough, but we should note that this question of defeat is very different from a second question.

    The second question regards what law it is that we are asking is justified. Note, for instance, that most laws are not justified when we ask about them being applied just to one person. A speed limit is not justified to me if it is narrowly construed as a restriction on the speed that I personally can drive. Instead, the speed limit is justified to me when it is conceived as a restriction on the speed of all drivers, including others and myself. When Kevin says that “the Mozert parents have defeaters for fine-grained characterization of the law in question, and… their defeaters are understood as defeaters that exempt them from the law rather than defeating the law for everyone[,]” it sounds like he is taking the test for the defeat to be a test of the narrowly construed law as it applies to some individual. That is, I claim, the wrong test. The question should regard a single law, thought to cover all people, and the question is whether the Mozerts have reason to accept or reject that law, rather than accept or reject just the application of that law to themselves. I’m sure we would all exempt ourselves from speed limits, taxes, and many other laws, as long as that did not change whether or not the laws applied to others, but the public reason project involves compromise in accepting restrictions on ourselves in part justified by the value of like restrictions on others. Whatever the scope of the law in question, it cannot be that the relevant test of justification regards whether or not be have reason to endorse the law applying to us holding fixed whether or not it will apply to others.

    Regarding the Smith case, I think Kevin needs to say a lot more about why we would understand the relevant law to be the one he indicates. In the book he claims to accept, for the discussion, that the laws prohibiting use of peyote are publicly justified, which must entail that they are publicly justified to Smith and Black themselves. But above Kevin claims that Smith and Black “did not break a publicly justified law (namely the law prohibiting narcotic use in religious ceremonies, but perhaps that’s only part of a law).” The law in question, I think clearly, is not a law against narcotic use in religious ceremonies, but a law against narcotic use. It is, in that way, like laws against murder which happen to include, without special provisions to target, cases of murdering apostates, human sacrifice, or the like. I can see no reason to think that we should test as an independent law the set of cases in which the peyote-prohibition happens to apply to religious ceremonies. What it seems to me Kevin must say if he is to defend a relevant exemption for Smith and Black is that the religious use of peyote is so important to them that they do not have sufficient reason to endorse the general ban on peyote. They would prefer a state of liberty, in which all members of the public are allowed to use peyote for whatever reason they see fit, be it religious ceremony or wild weekend, rather than have a law that would both stop the use by the potential abusers as well as their own religious use. That seems true of Smith and Black, and avoids trying to carve up the contexts of a law in ways that seem suspect and devastating to the possibility of having any justified legal order.

  2. Chad writes:

    [M]ost laws are not justified when we ask about them being applied just to one person. A speed limit is not justified to me if it is narrowly construed as a restriction on the speed that I personally can drive. Instead, the speed limit is justified to me when it is conceived as a restriction on the speed of all drivers, including others and myself. When Kevin says that “the Mozert parents have defeaters for fine-grained characterization of the law in question, and… their defeaters are understood as defeaters that exempt them from the law rather than defeating the law for everyone[,]” it sounds like he is taking the test for the defeat to be a test of the narrowly construed law as it applies to some individual. That is, I claim, the wrong test. The question should regard a single law, thought to cover all people, and the question is whether the Mozerts have reason to accept or reject that law, rather than accept or reject just the application of that law to themselves. I’m sure we would all exempt ourselves from speed limits, taxes, and many other laws, as long as that did not change whether or not the laws applied to others, but the public reason project involves compromise in accepting restrictions on ourselves in part justified by the value of like restrictions on others. Whatever the scope of the law in question, it cannot be that the relevant test of justification regards whether or not be have reason to endorse the law applying to us holding fixed whether or not it will apply to others.

    Chad, what difference does it make whether we think of the Mozerts having (a) defeaters for the application of the law to themselves, or (b) defeaters for the whole law but no defeaters for a more narrowly scoped law? That looks like the same thing.

    On the other hand, I think you’re right that we can’t generally apply the public justification test to the application of laws to specific persons. “I’m sure we would all exempt ourselves from speed limits…” I would put the issue differently. The problem is that many reasonable people won’t be able to express a preference about the application of the law to themselves without knowing whether it will apply to others too. A common view will be ‘I think the speed limit should apply to my driving if it applies to everyone else’s driving too, but not if it applies to me alone’. That’s clearly a reasonable view, so we can’t separate the two issues (speed limit for me, speed limit for you): we have to apply the public justifiability test to the proposal to have a speed limit for everyone (following Gaus’s independence criterion, OPR 495-7).

    To answer my own question from my first paragraph, the difference is that the process works top-down, by considering more general laws first then restricting their scope, but not bottom-up, i.e. figuring out what the scope of the law should be by asking whether A has a defeater, whether B has a defeater, etc.

    • Andrew, regarding the difference between thinking “the Mozerts having (a) defeaters for the application of the law to themselves, or (b) defeaters for the whole law but no defeaters for a more narrowly scoped law[,]” I think people will nearly always have defeaters for the laws applying to themselves if we can hold fixed that the law will continue to apply to others. Or, they will at least much more often have a defeater for the application to themselves than to the law generally. That is so with speed limits, pollution controls, taxes, and even laws restricting violence (even a generally non-violent person will often prefer to leave open the option of assaulting people with impunity as long as she would be the only one with that privilege). So, minimally they are different because the first generates vastly more defeat than the latter. So, I think the first way of understanding the object of justification leads to unacceptable outcomes.

      Thanks for providing your own thoughts on how the scope of the law would be picked out. I have no objection to what you say, but perhaps Kevin will wish to say more about why he seems to thinks it is better to test fine-grained or even specifically the application of a law.

  3. It’s not obvious that we only need to justify laws to those subject to them. Colin Bird argues that Joe’s reasonable objection to a pro-life anti-abortion law should count even though the law will never restrict his conduct.

    At the same time, even a law that doesn’t directly restrict my conduct may indirectly do so, because I am called upon to support the enforcement of the law, and also not to enforce whatever law I may think ought to be in place.

    • Andrew raises some very good points. If, as he suggests we might, we must justify the laws even to those they are not enforced against, a number of difficult issues will arise. First, something would have to be said about how to pick out the justificatory public if it is not the set of people who will be coerced via the law. Second, such a route will not, on Kevin’s account, generate laws without exemptions. Instead, it will tend to produce a lot more defeated laws. For instance, the Mozert parents may prefer that no one be forced to read the offensive book. If the Mozerts are part of the justificatory public for the accommodating law, despite the fact that it does not (directly) apply to them, then it too will be defeated. Third, as Andrew briefly notes, the fact that someone is not directly coerced does not mean that they are not relevantly impacted. Suppose we set the justificatory public in terms of who is coerced by a law, but the enforcement of all laws is funded by general taxation. Those who have defeaters for law L may well have defeaters for a law requiring them to fund enforcement of L.

      On those points, I suspect that Kevin will want to maintain that it really is only the coerced who are owed justification, but Kevin may be inclined to accept the last point. That is, Kevin is likely to say that at least often times the funding of enforcement will also have to come only from those to whom the law is justified. I am much less sure about what Kevin could say about other sorts of indirect coercion arising from a law.

Leave a Reply