When Liberty and Equality Conflict -- And When They Don't
by Michael Dorf
My latest Verdict column dives into the weeds of the Masterpiece Cakeshop case. To summarize and over-simplify, I argue that while there are hard cases that pit liberty against equality, Masterpiece Cakeshop should be deemed an easy case. That's not because the baker has no interests in this case. He may well have a substantial stake in the outcome of the case. It's just that he cannot win on his free speech claim without blowing up anti-discrimination law, and his claim that he has suffered discrimination based on religion does not find support in the record.
Here I want to address an issue I use to frame the discussion in the column: Should we understand apparent conflicts between liberty and equality as genuine--as value pluralists like Isaiah Berlin and Bernard Williams argued--or should we regard them as spurious--as Ronald Dworkin did? I'm not going to try to definitively resolve that question, but I am going to try to use it as a way of distinguishing two kinds of cases: those in which the (real or apparent) sacrifice of liberty for the sake of equality is a source of moral regret and those in which it is not.
There are many kinds of pluralism, but for present purposes it suffices to say: (1) moral pluralism is not moral relativism; (2) the kind of moral pluralism that I have in mind is value pluralism, under which one can simultaneously support multiple values; (3) for the type of value pluralist I have in mind, the domains of validity of these different values overlap; (4) therefore any resolution in favor of one or another value in this domain of overlap involves a "tragic choice;" and (5) by that I mean that the choice to favor liberty over equality or vice-versa is a source of moral regret, even if it is the right choice.
I am contrasting the foregoing account of value pluralism with a view elaborated by Dworkin, even though readers familiar with Dworkin's work principally because of his views about the nature of law may find that confusing. Dworkin famously argued against legal positivism by, among other things, insisting that law includes not only rules with an on/off quality but also principles that have weight and that therefore must be weighed against each other. That sounds like and is a kind of pluralism. However, in other writing with a more purely philosophical bent, especially in the last decade of his life but also before then, Dworkin wrote that he disagreed with the view of Berlin and others that political values--especially liberty and equality--conflict with one another. Dworkin did not deny that tragic choices can arise in life; he denied that the choice between liberty and equality, each properly understood, would ever be tragic. He thought that such conflicts seem to arise only because other thinkers (such as Berlin) use "flat" definitions, treating liberty as the liberty to do whatever one pleases and equality as fully equal distribution of goods. (I think this is not an entirely fair characterization of Berlin's view of liberty or equality, but let's put that to one side.)
Dworkin anticipated an obvious objection: We can grant that the flat conception is wrong, but how do we know that a thicker conception will make all of the conflicts go away? Sure, it is possible to gerrymander your definition of liberty and equality in such a way as to make any conflict appear to dissolve, but then you will have simply put the rabbit in the hat right before pulling it out. It's a neat trick if no one sees you putting the rabbit in, but a trick nonetheless.
I have made a version of this objection in another context. In our first major article on the debt ceiling, Prof. Buchanan and I explained that constitutional doctrine typically does not acknowledge that constitutional provisions can come into conflict with one another. As an illustration, we described Nebraska Press Ass'n v. Stewart, which we said involved a conflict between the Sixth Amendment right of a criminal defendant to a fair trial and the First Amendment right of a news organization to an open trial. Rather than openly acknowledging the conflict, however, the Supreme Court suppressed it, by defining the scope of the Sixth and First Amendments so that they did not conflict. We illustrated that move graphically with the following picture:
Prof. Buchanan and I said that the two approaches will often lead to the same bottom line, but we criticized the conflict suppression approach as less transparent than the conflict acknowledgment approach. Acknowledging the conflict between rights leads one to make explicit the choice one makes between them. Is Dworkin vulnerable to the same criticism?
He would be vulnerable if he defined liberty and equality for the purpose of avoiding their collision, but then ended up with definitions of each concept that sacrificed some of what makes it valuable. However, Dworkin denied that he was doing that. He said that when we articulate the best account of each value, we just happen to find that they end up as not conflicting.
Is that right? Here are Dworkin's definitions of equality (taken from a 2001 Arizona Law Review article that more or less restates what he wrote in his 2000 book Sovereign Virtue):
Dworkin's argument for his definitions is complex and irreducibly normative. I also find it ultimately unpersuasive, chiefly because the caveat on liberty--that one has no liberty to violate the rights of others--strikes me as the sort of conflict suppression that Prof. Buchanan and I criticized. The best way to make that point is with an example.
Consider claims by conscientious objectors to a right to avoid military conscription (perhaps by some alternative form of service) on the ground that they believe that killing in war is immoral (on religious or other grounds). Most democratic societies that have conscription allow exceptions for conscientious objectors, but I am not now interested in whether that resolution is correct.
My point is simply that the decision whether to allow conscientious objection to military service seems like a pretty clear and pretty high-stakes conflict between liberty and equality. Being told one must kill for the state when one believes killing is immoral is a very substantial limitation on one's liberty. At the same time, exempting some but not other people from military service and the substantial risk of death it poses is a very substantial unequal distribution of the burdens of citizenship. It is possible that the conflict dissolves when one properly applies Dworkin's definitions, but if so, I say, more the worse for his definitions. The force of the conflict--and the need for a tragic choice--can be felt in the example itself. Any definition of liberty or equality (or both) that dissolves the conflict is a definition that should not survive in reflective equilibrium. (I am grateful to my colleague Steve Shiffrin for a version of this point.)
I am left somewhere in between Berlin's pluralism and Dworkin's conception of "equal liberty." I agree with Dworkin that a choice between equality and the liberty to do whatever one wants need not pose a difficult, much less a tragic choice. But I disagree with his view that we can or should dissolve the conflict between equality and liberty by redefining those concepts as he would. A proper refinement will leave some room for conflict.
Now let's apply these ideas to the conflict between a right to same-sex marriage and the purported rights of business owners and others who deem same-sex marriage wrong to opt out by denying to same-sex couples services they would otherwise be obliged by anti-discrimination law to provide. My Verdict column acknowledges that there could be real conflicts of this sort -- between equality and speech or religion -- even as I also say that the Masterpiece Cakeshop case does not present a real such conflict.
Part of what I mean by saying that Masterpiece Cakeshop should lose is simply doctrinal. I explain why, under existing free speech and free exercise doctrine, the baker's claim should fail. That is not to deny, however, that there is at least a sacrifice of liberty in what Dworkin called Berlin's "flat" sense: Obviously, if the baker loses, he is not free to do what he pleases, which is to sell wedding cakes to opposite-sex couples but not to same-sex couples. But does the baker sacrifice liberty in a thicker sense?
Under Dworkin's definitions, no, because Dworkin's definitions of liberty and equality do not allow for a conflict. If the right answer in the case is that Dworkinian equality prevails, that necessarily means that there is no sacrifice of Dworkinian liberty.
But what about the middle ground between Berlin's flat sense of liberty and Dworkin's always-equality-conforming liberty--a middle ground that I'll self-aggrandizingly call "Dorfian" liberty? If the baker loses in Masterpiece Cakeshop, as I say he should, is that because of a necessary but nonetheless tragic choice to prefer equality? And if the answer is no in the case of the baker, might the answer be yes with respect to some business owners with stronger legal claims who nonetheless still lose?
Put differently, suppose that the right of a same-sex couple to purchase some goods or services free of sexual orientation discrimination prevails over the free speech and free exercise claims of religious business owners who oppose same-sex marriage; is that a source of moral regret in the way that any resolution of a conscientious objector's objection to military conscription is a source of moral regret? I don't think it's possible to answer that question without reference to one's views about sexual orientation.
Consider Justice Kennedy's majority opinion in Obergefell v. Hodges. He repeatedly tells the reader that the opinion should not be taken to imply that the opponents of same-sex marriage are benighted. Their view, he says, "long has been held—and continues to be held—in good faith by reasonable and sincere people here and throughout the world." He later goes on to say that many people "who deem same-sex marriage to be wrong reach that conclusion based on decent and honorable religious or philosophical premises, and neither they nor their beliefs are disparaged here."
Writing in dissent, Justice Alito was not reassured. He worried that recognition of a constitutional right to same-sex marriage was a step on the road to making opposition to same-sex marriage unspeakable in polite company. He wrote that the majority opinion would
The answer cannot be that all religious premises, much less all philosophical premises, are decent and honorable. Does sati derive from decent and honorable premises? How about jihad as understood by ISIS? More to the current point, when, as recently as the year 2000, Bob Jones University banned interracial dating among its students based on a particular fundamentalist Protestant reading of the Bible, were the university leaders reasoning from decent and honorable religious premises?
Even religions that preach fundamentalism tend to hold to beliefs and doctrines that actually change over time. And so, just as religious justifications for racism waned in the wake of the civil rights movement (which, we do well to recall, was itself led by people who drew inspiration from a different reading of the same religious traditions), we can expect that the dynamic Justice Alito anticipates will in fact occur.
Where that leaves us at the current moment is not entirely clear. If a religiously motivated racist sought and was denied an exemption from anti-discrimination law on the ground that the government's compelling interest in barring race discrimination overrides his religious freedom, I doubt very much that the many people who would support that result would say it is a source of moral regret. People who hold religiously motivated views against same-sex marriage may eventually come to be seen in the same way.
I would not say that's categorically true yet, however, because religious opposition to same-sex marriage is quite common today in a way that religious opposition to interracial marriage is not.
The US has a relatively high conversion rate. A 2015 Pew survey found that 42% of American adults had switched religions at some point, and that number drops only a bit, to a still-fairly-high 35%, when one does not count switching from one Protestant sect to another. That means that if someone belongs to a religion (or sect) that does not permit same-sex marriage, it is relatively easy to switch to another religion (or sect) that does. Even so, in 2017 many people who generally accept religious authority as a source of moral guidance accept that guidance with respect to same-sex marriage as simply part of the prix-fixe menu of their religion, whereas one would have to go out of one's way to find a religion that condemns interracial marriage.
Thus, for the time being we might still regard the overriding of a religious business owner's opposition to same-sex marriage as justified but nonetheless a source of moral regret: the outcome of a necessary albeit tragic choice. Over time, however, as the process that Justice Alito predicted unfolds, fewer and fewer religions will condemn same-sex marriage. At some point, finding a religion that does not allow same-sex couples to marry might be like finding a religion that disallows interracial marriage today. At that point, only true homophobic bigots will voice religious objections to same-sex marriage, so we will then be able to say that overriding their objections is not a source of moral regret. At that point, their liberty claims, properly characterized, will not conflict with equality, because their liberty claims, properly characterized, will fail to get off the ground.
My latest Verdict column dives into the weeds of the Masterpiece Cakeshop case. To summarize and over-simplify, I argue that while there are hard cases that pit liberty against equality, Masterpiece Cakeshop should be deemed an easy case. That's not because the baker has no interests in this case. He may well have a substantial stake in the outcome of the case. It's just that he cannot win on his free speech claim without blowing up anti-discrimination law, and his claim that he has suffered discrimination based on religion does not find support in the record.
Here I want to address an issue I use to frame the discussion in the column: Should we understand apparent conflicts between liberty and equality as genuine--as value pluralists like Isaiah Berlin and Bernard Williams argued--or should we regard them as spurious--as Ronald Dworkin did? I'm not going to try to definitively resolve that question, but I am going to try to use it as a way of distinguishing two kinds of cases: those in which the (real or apparent) sacrifice of liberty for the sake of equality is a source of moral regret and those in which it is not.
There are many kinds of pluralism, but for present purposes it suffices to say: (1) moral pluralism is not moral relativism; (2) the kind of moral pluralism that I have in mind is value pluralism, under which one can simultaneously support multiple values; (3) for the type of value pluralist I have in mind, the domains of validity of these different values overlap; (4) therefore any resolution in favor of one or another value in this domain of overlap involves a "tragic choice;" and (5) by that I mean that the choice to favor liberty over equality or vice-versa is a source of moral regret, even if it is the right choice.
I am contrasting the foregoing account of value pluralism with a view elaborated by Dworkin, even though readers familiar with Dworkin's work principally because of his views about the nature of law may find that confusing. Dworkin famously argued against legal positivism by, among other things, insisting that law includes not only rules with an on/off quality but also principles that have weight and that therefore must be weighed against each other. That sounds like and is a kind of pluralism. However, in other writing with a more purely philosophical bent, especially in the last decade of his life but also before then, Dworkin wrote that he disagreed with the view of Berlin and others that political values--especially liberty and equality--conflict with one another. Dworkin did not deny that tragic choices can arise in life; he denied that the choice between liberty and equality, each properly understood, would ever be tragic. He thought that such conflicts seem to arise only because other thinkers (such as Berlin) use "flat" definitions, treating liberty as the liberty to do whatever one pleases and equality as fully equal distribution of goods. (I think this is not an entirely fair characterization of Berlin's view of liberty or equality, but let's put that to one side.)
Dworkin anticipated an obvious objection: We can grant that the flat conception is wrong, but how do we know that a thicker conception will make all of the conflicts go away? Sure, it is possible to gerrymander your definition of liberty and equality in such a way as to make any conflict appear to dissolve, but then you will have simply put the rabbit in the hat right before pulling it out. It's a neat trick if no one sees you putting the rabbit in, but a trick nonetheless.
I have made a version of this objection in another context. In our first major article on the debt ceiling, Prof. Buchanan and I explained that constitutional doctrine typically does not acknowledge that constitutional provisions can come into conflict with one another. As an illustration, we described Nebraska Press Ass'n v. Stewart, which we said involved a conflict between the Sixth Amendment right of a criminal defendant to a fair trial and the First Amendment right of a news organization to an open trial. Rather than openly acknowledging the conflict, however, the Supreme Court suppressed it, by defining the scope of the Sixth and First Amendments so that they did not conflict. We illustrated that move graphically with the following picture:
He would be vulnerable if he defined liberty and equality for the purpose of avoiding their collision, but then ended up with definitions of each concept that sacrificed some of what makes it valuable. However, Dworkin denied that he was doing that. He said that when we articulate the best account of each value, we just happen to find that they end up as not conflicting.
Is that right? Here are Dworkin's definitions of equality (taken from a 2001 Arizona Law Review article that more or less restates what he wrote in his 2000 book Sovereign Virtue):
Equality is satisfied when any differences in people's resources reflects the different costs to others of choices they have made. Liberty consists in being able to do what one wishes, short of violating the rights of others, with the resources assigned by a reasonably just distribution of resources.A moment's reflection will reveal that these definitions make conflicts between equality and liberty requiring a tragic choice essentially impossible. However, Dworkin thought that these definitions could be justified without stipulating in advance that liberty cannot conflict with equality.
Dworkin's argument for his definitions is complex and irreducibly normative. I also find it ultimately unpersuasive, chiefly because the caveat on liberty--that one has no liberty to violate the rights of others--strikes me as the sort of conflict suppression that Prof. Buchanan and I criticized. The best way to make that point is with an example.
Consider claims by conscientious objectors to a right to avoid military conscription (perhaps by some alternative form of service) on the ground that they believe that killing in war is immoral (on religious or other grounds). Most democratic societies that have conscription allow exceptions for conscientious objectors, but I am not now interested in whether that resolution is correct.
My point is simply that the decision whether to allow conscientious objection to military service seems like a pretty clear and pretty high-stakes conflict between liberty and equality. Being told one must kill for the state when one believes killing is immoral is a very substantial limitation on one's liberty. At the same time, exempting some but not other people from military service and the substantial risk of death it poses is a very substantial unequal distribution of the burdens of citizenship. It is possible that the conflict dissolves when one properly applies Dworkin's definitions, but if so, I say, more the worse for his definitions. The force of the conflict--and the need for a tragic choice--can be felt in the example itself. Any definition of liberty or equality (or both) that dissolves the conflict is a definition that should not survive in reflective equilibrium. (I am grateful to my colleague Steve Shiffrin for a version of this point.)
I am left somewhere in between Berlin's pluralism and Dworkin's conception of "equal liberty." I agree with Dworkin that a choice between equality and the liberty to do whatever one wants need not pose a difficult, much less a tragic choice. But I disagree with his view that we can or should dissolve the conflict between equality and liberty by redefining those concepts as he would. A proper refinement will leave some room for conflict.
* * *
Now let's apply these ideas to the conflict between a right to same-sex marriage and the purported rights of business owners and others who deem same-sex marriage wrong to opt out by denying to same-sex couples services they would otherwise be obliged by anti-discrimination law to provide. My Verdict column acknowledges that there could be real conflicts of this sort -- between equality and speech or religion -- even as I also say that the Masterpiece Cakeshop case does not present a real such conflict.
Part of what I mean by saying that Masterpiece Cakeshop should lose is simply doctrinal. I explain why, under existing free speech and free exercise doctrine, the baker's claim should fail. That is not to deny, however, that there is at least a sacrifice of liberty in what Dworkin called Berlin's "flat" sense: Obviously, if the baker loses, he is not free to do what he pleases, which is to sell wedding cakes to opposite-sex couples but not to same-sex couples. But does the baker sacrifice liberty in a thicker sense?
Under Dworkin's definitions, no, because Dworkin's definitions of liberty and equality do not allow for a conflict. If the right answer in the case is that Dworkinian equality prevails, that necessarily means that there is no sacrifice of Dworkinian liberty.
But what about the middle ground between Berlin's flat sense of liberty and Dworkin's always-equality-conforming liberty--a middle ground that I'll self-aggrandizingly call "Dorfian" liberty? If the baker loses in Masterpiece Cakeshop, as I say he should, is that because of a necessary but nonetheless tragic choice to prefer equality? And if the answer is no in the case of the baker, might the answer be yes with respect to some business owners with stronger legal claims who nonetheless still lose?
Put differently, suppose that the right of a same-sex couple to purchase some goods or services free of sexual orientation discrimination prevails over the free speech and free exercise claims of religious business owners who oppose same-sex marriage; is that a source of moral regret in the way that any resolution of a conscientious objector's objection to military conscription is a source of moral regret? I don't think it's possible to answer that question without reference to one's views about sexual orientation.
Consider Justice Kennedy's majority opinion in Obergefell v. Hodges. He repeatedly tells the reader that the opinion should not be taken to imply that the opponents of same-sex marriage are benighted. Their view, he says, "long has been held—and continues to be held—in good faith by reasonable and sincere people here and throughout the world." He later goes on to say that many people "who deem same-sex marriage to be wrong reach that conclusion based on decent and honorable religious or philosophical premises, and neither they nor their beliefs are disparaged here."
Writing in dissent, Justice Alito was not reassured. He worried that recognition of a constitutional right to same-sex marriage was a step on the road to making opposition to same-sex marriage unspeakable in polite company. He wrote that the majority opinion would
be used to vilify Americans who are unwilling to assent to the new orthodoxy. In the course of its opinion, the majority compares traditional marriage laws to laws that denied equal treatment for African-Americans and women. The implications of this analogy will be exploited by those who are determined to stamp out every vestige of dissent. Perhaps recognizing how its reasoning may be used, the majority attempts, toward the end of its opinion, to reassure those who oppose same-sex marriage that their rights of conscience will be protected. We will soon see whether this proves to be true. I assume that those who cling to old beliefs will be able to whisper their thoughts in the recesses of their homes, but if they repeat those views in public, they will risk being labeled as bigots and treated as such by governments, employers, and schools.Although I disagree with the normative spin, I think Justice Alito had this about right. If denial of the right to same-sex marriage is like denial of the right to interracial marriage--as the majority in Obergefell says it is, rightly in my view--then how can one derive opposition to same-sex marriage from "decent and honorable religious or philosophical premises"?
The answer cannot be that all religious premises, much less all philosophical premises, are decent and honorable. Does sati derive from decent and honorable premises? How about jihad as understood by ISIS? More to the current point, when, as recently as the year 2000, Bob Jones University banned interracial dating among its students based on a particular fundamentalist Protestant reading of the Bible, were the university leaders reasoning from decent and honorable religious premises?
Even religions that preach fundamentalism tend to hold to beliefs and doctrines that actually change over time. And so, just as religious justifications for racism waned in the wake of the civil rights movement (which, we do well to recall, was itself led by people who drew inspiration from a different reading of the same religious traditions), we can expect that the dynamic Justice Alito anticipates will in fact occur.
Where that leaves us at the current moment is not entirely clear. If a religiously motivated racist sought and was denied an exemption from anti-discrimination law on the ground that the government's compelling interest in barring race discrimination overrides his religious freedom, I doubt very much that the many people who would support that result would say it is a source of moral regret. People who hold religiously motivated views against same-sex marriage may eventually come to be seen in the same way.
I would not say that's categorically true yet, however, because religious opposition to same-sex marriage is quite common today in a way that religious opposition to interracial marriage is not.
The US has a relatively high conversion rate. A 2015 Pew survey found that 42% of American adults had switched religions at some point, and that number drops only a bit, to a still-fairly-high 35%, when one does not count switching from one Protestant sect to another. That means that if someone belongs to a religion (or sect) that does not permit same-sex marriage, it is relatively easy to switch to another religion (or sect) that does. Even so, in 2017 many people who generally accept religious authority as a source of moral guidance accept that guidance with respect to same-sex marriage as simply part of the prix-fixe menu of their religion, whereas one would have to go out of one's way to find a religion that condemns interracial marriage.
Thus, for the time being we might still regard the overriding of a religious business owner's opposition to same-sex marriage as justified but nonetheless a source of moral regret: the outcome of a necessary albeit tragic choice. Over time, however, as the process that Justice Alito predicted unfolds, fewer and fewer religions will condemn same-sex marriage. At some point, finding a religion that does not allow same-sex couples to marry might be like finding a religion that disallows interracial marriage today. At that point, only true homophobic bigots will voice religious objections to same-sex marriage, so we will then be able to say that overriding their objections is not a source of moral regret. At that point, their liberty claims, properly characterized, will not conflict with equality, because their liberty claims, properly characterized, will fail to get off the ground.