It seems that governments sometimes act in ways that are good, but not required by any plausible conception of justice. Think of the lovely flowers on the highway median, or the block party thrown in your town each year, or the grant you received last summer. Or, more controversially, consider the victim compensation fund set up by the U.S. Congress after the September 11th attacks, or maybe even the $700 billion bailout of the U.S. financial industry going on right now.
When the government provides for these things, it may be engaging in what we can call government supererogation. Which acts we’d label as examples of government supererogation will depend on which theories of justice and goodness we accept. While there is disagreement over these theories, and thus disagreement over which specific acts are supererogatory, I believe most people would identify some acts as fitting into the category. Taken together, these acts may consume a substantial amount of public resources and attention.
In my contribution to the Political Philosophy Podcast Symposium, I argue that the kinds of acts that seem to be examples of government supererogation are only apparently supererogatory. Because of the kind of agent government is, the category of supererogation is unavailable to it. Acts of apparent government supererogation, I argue, are instead either wrong or required by justice, and for most such acts the latter option is implausible.
This paper is related to a larger project of mine on questions about the agency of justice. Some of these questions have to do with the agency of the state, since the state is typically thought of as the prime agent of justice. Other questions concern other possible agents of justice, such as private citizens and voluntary associations.
I look forward to getting your thoughts on the paper. Thanks for listening / reading. Thanks also to Simon for setting up the Symposium, and to Helena de Bres for her very helpful comments.
Click here to read the paper.
Click here to read Helena de Bres’s comments on the paper.
You must be registered and logged in to post a comment. Public Reason welcomes participation from members of the academic community with an interest in political philosophy and theory. Your registration as a participant is subject to approval. Please specify your academic institutional affiliation on the registration form.
10 comments
Comments feed for this article
1 - Friday, 17 October 2008 at 9:13 pm
Richard Chappell
I have some worries concerning the assumed relation between taxation and coercion. It’s true in some sense that taxation is coercive. But so is the alternative. Indeed, the enforcement of libertarian property rights may be even more (or more problematically) coercive. So the idea that coercion (understood in some purely descriptive, pre-political sense) necessarily requires special justification strikes me as a non-starter. I guess I’m more sympathetic to indirect consequentialist approaches precisely because they avoid this problem.
2 - Saturday, 18 October 2008 at 2:36 pm
Justin Weinberg
Thanks for your comment, Richard. Sure, when Gerry Cohen pitches a tent in my backyard and I call the police to complain about his trespassing, and they force him off my property, then he is being coerced. My argument doesn’t depend on taxation being uniquely coercive among the various activities available to government. I agree that all sorts of government activity, including the enforcement of property rights, can be coercive. But I don’t see how it follows from this that coercion wouldn’t require some kind of “special justification”.
This justification has typically come in the form of a theory of justice that provides us reasons for why the coercion under consideration, as opposed to other forms of coercion, is warranted. (In her comments, Helena offers some alternative justifications for coercion, and I’ll be posting a response to her in the next day or so.)
Even if we were to adopt an indirect consequentialist theory, it seems likely that we’d still want a special justification for coercion, presumably because of either the high instrumental or intrinsic value of allowing people to act as they please. The presumptive utility of liberty gives the indirect consequentialist a good reason to put pretty tall justificatory barriers in the way of restrictions on it, and this is true even if we have no choice but to restrict people’s liberties in some way or another.
3 - Saturday, 18 October 2008 at 9:57 pm
Richard Chappell
Hi Justin, thanks for your thoughtful response. On indirect consequentialism, however, I don’t think you have taken account of what (in my linked post) I called the distinction between ‘internal’ and ‘external’ questions. Certainly we should set up our basic institutions such that people have various liberties, protected by “tall justificatory barriers”. I do agree, for example, that a special justification would be required for ad hoc confiscation of (post-tax) property. But this is an internal question, of actions that take place within the institutional structure and are bound by the internal rules (or rights). Those rights may define a moralized conception of ‘coercion’, for use as an internal objection to certain actions. It is ‘coercive’ (in this moralized and morally relevant sense) to intrude upon another’s rightful holdings. But the basic structure (including taxation) sets the limits of our rightful holdings, so such ‘internal’ criticisms cannot be raised against it. It is instead beholden to the ‘external’ standard of consequentialism: does this basic system tend to promote the good? If so, that’s all the justification it needs. So, on this picture: taxation is only ‘coercive’ in a ‘non-moralized’ and morally irrelevant sense.
In any case, I look forward to reading your response to Helena’s comments.
4 - Monday, 20 October 2008 at 11:54 pm
Simon Cabulea May
Thanks for the paper Justin and for the comments Helena.
The key claim in the paper is that it is wrong for governmental action is wrong unless justified by appeal to justice. Like Helena, I’m not persuaded that this claim is given enough support. I think there are a couple of ways to go about criticising it.
First, the discussion on p. 9 concerning the assumption that governmental action is always coercive seems to me to be too quick. Not only do governments do all sorts of things that do not coerce anyone, they can also support these activities through processes that do not involve coercion. Taxation is just one form of state revenue, even if the dominant form. In addition, governments could run lotteries, and use the income from those lotteries to subsidise various cultural (etc.) activities. This is not especially uncommon. Or state-owned industries could turn profits that could be used for ends unrelated to justice. It seems to me that there is still an interesting question of supererogation in these cases, because there could be an argument that justice demands all resources at a government’s disposal, however they may have been acquired, e.g. if justice demands that the position of the worst off be improved as much as possible, then any resources dedicated to projects other than that end may be unjustly allocated. This requires assessing the substantive content of the theory of justice to see how insatiable it is.
Second, turning to the question of state coercion, I also am not entirely persuaded that state coercion — in this case taxation — can only be justified by appeal to the ends of justice. On this question, there are two issues which are often conflated. One is a question of legitimacy, the other is a question of justifiability. Legitimacy concerns the conditions under which a government has the right to exercise power. On this point, Rawls is insistent, as I think he is right to be, that justice is not a precondition of the legitimate exercise of political power. Parties with the wrong platforms can be democratically elected and as a consequence govern legitimately, as long as this does not exceed certain broad limits. The function of political liberalism is to define these limits, at least on fundamental issues. Even then, on relatively peripheral issues, Rawls does not think that the justification of governmental action must be cashed out in terms of political values to be legitimate — ipso facto, governmental action need not be a requirement of justice to be legitimate. So I think the thesis of the paper is not going to get much support from anything Rawls says about legitimacy.
But legitimacy is a red herring here. The real question is whether state coercion in the form of taxation can be morally justified when not for any end of justice. Bracketing aside the above question about how demanding justice is — which I think really is a problem — it seems to me that it is relatively easy to override the default assumption against coercion. Taxation seems to me to be unlike other forms of coercion because it is part of the conditions under which one participates in a public endeavour, i.e. the economic system of a society. It’s not like having the police invade one’s bedroom or prohibit one from praying, i.e. the intrusion of the state into properly private affairs. So I’m not particularly troubled by democratic governments modifying the conditions of participation in the economic system so that various desirable ends other than justice may be promoted, as long as this modification does not violate justice.
The argument on pp. 6-7 for premise 4.1 on p. 8 doesn’t persuade me. In the case of the judge, we have an example of a direct conflict with justice, since the sentence is less than that required by justice. There’s a difference between saying (a): Justice does not require that the sentence be less than X, and (b): Justice requires that the sentence not be less than X. Proposition (a) does not imply proposition (b), because people can be given lenient sentences that they are not entitled to. This is not an injustice, even if a more severe sentence would also not have been an injustice. Similarly, in the case of excluding evidence, we have a claim of type (b): Justice requires that the evidence not be brought into the proceedings. This is quite different from a claim of type (a): Justice does not require that the evidence be brought into the proceedings. Again, we may have a lot of information in a trial the inclusion of which is neither unjust nor required by justice — e.g. information about how much the defendant’s parents love him. So I don’t see that 4.1 and hence 4.3 should be accepted on the basis of the discussion on pp. 6-7.
If they may be coercively promoted, through taxation, what marks the desirable ends off from those of justice? I don’t think it is particularly revisionary to say that they are different. First, the kind of moral loss involved in not pursuing those ends would not amount to injustice. Not preserving old buildings is sad, and in some sense a moral loss, but it is hardly the kind of moral necessity that justice sets out. Saving old buildings is not the first virtue of a social system. Second, when we don’t save old buildings, I don’t think there has to be any particular person or group of people who have been wronged. Future generations have not been benefited by our wise preservation policies, but not every failure to benefit is an infliction of a wrong. In contrast, when the state does not try to ensure that each child is well-nourished, then I think it wrongs those children who go hungry, and this is an injustice. Third, similarly, no-one is entitled to a state policy that preserves old buildings, but I do think that everyone is entitled to the advantages that justice dictates they should receive. What this suggests to me is that one should not rely on the assumption that justice just is whatever may be coercively enforced, because not everything that it may be morally permissible to use coercion to achieve has to be a matter of just entitlement. Instead, there are two routes to the no-government-supererogation thesis: either 1. argue that the assumption against coercive taxation really is weighty enough to trump the relatively petty concerns of old building preservation, cultural goods, etc., or 2. argue that the demands of justice override any justification for using resources (however acquired) for desirable ends other than justice.
5 - Tuesday, 21 October 2008 at 3:15 pm
Justin Weinberg
Helena de Bres provided some great comments and a helpful reconstruction of my argument. She focuses largely on the claim that government coercion must be grounded on justice.
Here is a relevant passage on this from my paper:
With the claim that government coercion must be grounded on justice I took myself to be stating a widely held view. Here’s Rawls on the topic:
Let’s work backwards through the quote to see what Rawls is saying here. The last sentence says that coercion must be based on public reason. The sentence before that says that public reasons must come from our political conception of justice. So what Rawls is saying is that the case for coercion must be made on the basis of political justice. Now just because Rawls says something doesn’t mean that it is right, but it probably does mean that it isn’t far off from what a lot of other people think.
The kinds of examples I’ve used for discussing government supererogation may not look like they belong among the “fundamental” or “essential” concerns, but, as Rawls notes, they are closely related to them, along with “tax legislation and laws regulating property… laws establishing national parks and voting funds for museums and the arts” (JaF, p.91). It appears to me that the question of “for what may citizens be coerced” is a constitutional essential, and as such, the kind of question the answer to which must be backed up by political justice.
Helena asks, “Why…might not government coercion be legitimately justified on grounds other than justice?” This is a good question. She also offers some alternative grounds for government coercion: humanitarianism, culture, and solidarity.
One way to respond to this is by way of what I referred to in the paper as the “presto” approach (p.12). That is, the allegedly non-justice grounds for coercion could be built into one’s conception of political justice. I think this is a common move. For example, some communitarians support government’s coercive limitation of people’s choices so as to promote the common good. Does this provide a non-justice-based ground for coercion? I don’t think so. Instead, we get “communitarian theories of justice” that argue that justice requires the promotion of the common good, coercively, if need be. Similarly, we get consequentialist theories of justice that build the production of good into the conception of justice.
Regarding any good, the more plausible the case for its production through government coercion, the more plausible the case that it is actually part of one’s theory of justice.
This is why, it seems to me, that Helena’s humanitarianism example is her strongest. Helena claims that some people believe the following three claims:
It does seem that certain kinds of humanitarian aid are worth coercing for. Do we have here an example of acceptable government coercion that is not based on justice? I am somewhat skeptical. The answer depends on the meaning of “justice” in the second claim Helena lists. She mentions two possibilities: respecting the negative pre-political rights of the poor, or meeting a standard of global distributive fairness. It is of course possible to reject these two conceptions of justice and accept that government can coercively produce humanitarian aid. But to reject those two conceptions of justice is not to reject the idea that justice is what grounds the permissibility of government coercion. For there may be other conceptions of justice that we think do a better trick. Helena notes that Pogge would reject the second claim but accept the third. He argues that aid from wealthy to poor countries is a matter of justice because of the causal relationship between the wealth of some countries and the poverty of others. It is true that we can reject this causal-interaction conception of justice and still be committed to coercively-produced humanitarian aid. However, I am instead tempted to think that those who make this move don’t disagree that justice requires such aid; rather, they simply disagree with the particular conception of justice on offer. Someone with a more consequentialist conception of political justice could argue on the basis that we could achieve a tremendous amount of good with humanitarian aid that government is permitted to coerce it, if that’s what it takes to produce the required good, (regardless of any causal claims about what brought about the need for such aid). Alternatively, someone who holds that justice requires rough material equality would hold that if humanitarian aid is necessary for that equality, it may be produced coercively.
So where Helena sees persons rejecting justice as the grounding for a particular coercive measure, an alternative explanation is that persons are simply rejecting one conception of justice and embracing another as the grounding for a particular coercive measure.
Once we move away from humanitarian aid to something like fireworks, the plausibility of this alternative explanation fades. But so, too, does the sense that it is just to coerce people into paying for or attending such events.
So that is one way to respond. I am unsure of its strength. It alone does not show that one can’t coherently ground coercion on something besides political justice. It only shows how disagreement over conceptions of justice, instead of disagreement over the justificatory power of the concept of justice, might explain why some views that purport to call for coercion based on something besides justice actually do not.
There is more to say on this point, and to other points Helena, and now Simon, have made. Since this post is pretty long, I’ll end this one here and work up another to continue the discussion.
6 - Wednesday, 22 October 2008 at 4:59 pm
Justin Weinberg
Thanks to Simon for his comments. Here is a partial reply, focusing on the kinds of coercion involved in potential government supererogation, and his remarks on taxation.
First, Simon writes that not all government action is coercive. As I note in the paper, I am not making the claim that government is necessarily coercive. It is just that, as a practical matter, the kinds of acts I cite as examples of potential government supererogation happen to make use of government coercion. (Most people find government coercion in the service of these acts perfectly legitimate; so if I were forced to narrow my thesis to arguing that coercive government supererogation is not possible, this would still be a challenge to a widely held view.)
Could government perform the acts in question non-coercively? It may depend on the act. Historic preservation measures, by which government limits the architectural choices of homeowners and developers in order to maintain an area’s historic character, would seem to require coercion. This is because its success depends not just on funding but on certain kinds of behavior. To historically preserve an area you have to prevent people from, say, putting vinyl siding on their homes, rather than wood. For the block party example, success is not just procuring money to pay for the entertainment, but, presumably, preventing people from driving down the street during the party. Providing for poetry-writing grants, though, is different. For these kinds of government acts that require funding but no forced behavior in order to happen, then there may be, as Simon suggests, non-coercive means of generating this funding.
That said, I think we do need to be careful in our description of certain means as non-coercive. Take Simon’s suggestion of lotteries. State-run lotteries generate the revenue they do in part because they are the only such game in town. As far as I know, in the U.S. there are no large, regularly scheduled legal private lotteries. This is not some interesting result of people’s free choices. Rather, the government prohibits private citizens and businesses from running such lotteries. Those who attempt to do so will be charged with a crime and subject to the coercive power of the government. While most folks don’t consider their purchase of, say, Powerball tickets, coerced, if the success of Powerball depends on coercively excluding competition, I don’t think we can call it and other lotteries non-coercive sources of revenue. The same applies to Simon’s other example of state-owned industries. Sometimes governments protect state-owned firms from private competition, and sometimes they do not. If the former, it seems that to the extent that the profits of a state-owned firm are owed to its government-enforced monopoly status, such profits are the product of government coercion.
This does not mean that there are no non-coercive ways for the government to generate revenue. But I do think these ways are a bit less common than one might think.
On taxation, Simon writes:
Here Simon provide an account of the coercive structure of taxation along the following lines: in order for an agent, A, to partake in a beneficial public system, practice, or institution, S, she must meet condition C, where S is the economic system and C is the paying of taxes. In short, in order for A to S, she must C. And it seems Simon intends this account of the coercive structure of taxation to provide support for the idea that taxation is not very coercive, if at all, or not a troubling form of coercion. If taxation isn’t very coercive, Simon argues, then we needn’t be too worried about its use to fund government activities not required for justice.
The coercive structure Simon provides for taxation may be accurate, but I do not think it gives us a reason to be less concerned about coercion through taxation, or to think that this coercion isn’t very coercive. This is because most forms of government coercion have the structure he describes, and that doesn’t lead us to think of them as any less coercive. So suppose that the government imposed the following condition on air travel: all passengers must submit to a drug test and a strip search. This, too, has the same structure as taxation. In order for a traveler, A, to participate in the system of air traffic, S, she has to take a drug test and remove her clothes to be searched, C: that is, in order for A to S, she must C. Nonetheless, most people would find C, here, quite coercive.
If there is a difference in the kinds of coercion in these cases, then, it is not a difference in structure. Rather, it is a difference based on a substantive moral disagreement about how bad it is to be coerced (or how wrong it is to coerce) one way rather than another.
So what is the takeaway from all of this? First, many forms of supposed government supererogation require not just (i) securing funding but also (ii) the coercive enforcement of certain kinds of behavior. So even if the government could take care of (i) without coercion, some forms of supposed government supererogation would still be coercive. Second, there aren’t as many noncoercive means of (i) as it seems, as some versions of lotteries and state profit-making are coercive. And third, those who want to deny that taxation is a coercive form of (i), or that it is coercive but not in a troubling way, cannot rely on arguments about the structure of coercion; instead, they must make substantive arguments about why taxation is less bad, or less wrong, a type of coercion, such that the usual conditions on the use of coercion do not apply to it. In short, there is no quick and easy way to convincingly argue that (many? most?) acts of potential government supererogation are not coercive, or not coercive in a worrisome sense. Of course, this is only one piece of the puzzle I am trying to figure out.
7 - Tuesday, 28 October 2008 at 3:53 am
Joshua Miller
On March 9th, 2006, the US Senate passed S.RES.397 , which was a resolution “recognizing the history and achievements of the curling community of Bemidji, Minnesota.” This seems a case of supererogation, as does the naming of airports and highways after extraordinary individuals.
You write of non-coercive government actions that while we can imagine such actions would be performed after the state’s duty to justice is discharged, so long as they are unfunded and optional, nonetheless, “this model does not depict how governments actually perform the kinds of acts that are typical candidates for supererogation, nor is it clear how, without coercion, governments would be able to achieve the aims they set out to with these kinds of acts.” (9) However, naming and recognizing are acts of state that need not be funded or coerced. The only condition on which this would be coercive is your first provision: they occur while justice has yet to be achieved. Yet this seems far too demanding a standard: why ought we to celebrate anything so long as human beings suffer?
Another possible response is that naming and recognition are in fact duties. Things must have names if we are to refer to them, after all, and we are duty bound to name them under some set of conditions. These conditions would perhaps require that we choose from a pool of names with enough diversity to distinguish the referents, while granting the best place names to the best of our citizens and allowing more frequent repetitions for particularly well-respected citizens.
The same duty-bound argument is possible for recognition: we might claim that some groups, like curlers, justly deserve recognition. Perhaps the Senate also has a duty not to recognize some groups. (Ice fishers maybe?) On a more serious note, we’ve talked before here about whether the Senate should recognize genocides like that of the Armenians, and that discussion really does boil down to a conflict among duties: to the truth, to the Kurds, etc. I’d distinguish that question from my allegedly supererogatory recognition.
I would argue that recognition of the sort granted to the curlers cannot be a duty. While we ought to recognize some members of our community, if we only do so out of duty the advantages of recognition are lost. These kinds of affirmations are unnecessary by design. Like supererogation itself, such gratuitous celebrations help us confront the sense that justice is an over-demanding measure against which we must always fall short.
8 - Tuesday, 28 October 2008 at 4:51 pm
Justin Weinberg
Joshua– I think you are right to note a form of supererogation I overlooked, and which need not be coercive, that is, when the government honors or recognizes persons, groups, events, etc. This presumably could be done with no additional taxation (it is not as if we weren’t going to pay the Senators that day) or ordering about. I do recall when National Airport was renamed after Reagan many people (especially air traffic controllers) complaining about being forced to use this name, but I think that is a special case and not one we could generalize from.
Additionally, you are right to note that it would be too strong to require that society be just as a condition for government supererogation. I wrote that government supererogation requires that:
I realize now that the first part of this sentence isn’t right; as you say it is too demanding. It is really the second part that is critical. If it is because I am performing an apparently supererogatory act that I fail to do my duty, then my act is not really supererogatory (e.g., I abandon my post guarding the highly sought-after weapons in order to help someone across the street).
Thanks for this help.
9 - Tuesday, 28 October 2008 at 5:43 pm
Joshua Miller
Oh! I see you’re at Georgetown. (I teach in the DC area as well.) Yes, I was thinking of Reagan National as a possible problem, here, in the same way that the refusal to recognize Martin Luther King strikes me as potentially unjust. Anchorage, Alaska’s Ted Stevens International Airport is another great example, and on a related note, so are those state flags that incorporate the Confederate Flag.
I’m surprised by your response, actually. Insofar as we can submit the various place names to deliberative judgment, I think you could consistently demand that such decisions all be submitted to some minimal procedural or substantive testing in order to maintain the government’s legitimacy.
I have in mind here Rawls’ “offices and honors” provision. It seems to me that you’d have to be committed to the equitable distribution of names and recognitions in order to maintain your position, because it seems false to claim that these recognitions and honors stand in the way of Senators doing their just duties. The absurdity of honoring Minnesota curlers may be distracting, but I expected you to argue that this absurdity requires some serious explanation or else it undermines the legitimacy of such actions by a JBA.
10 - Wednesday, 29 October 2008 at 3:36 pm
Justin Weinberg
I’m not sure I follow you, Joshua. Justice does not seem to require naming airports after people (to use one example). Since having an airport named for you is considered an honor, such naming provides a good. Since it is not required by justice, but good, it strikes me as a possible example of government supererogation.
Now when I say that justice does not require naming airports after people, I don’t have specific people in mind. The idea is that justice does not require that any honoring-via-airport-naming take place.
That said, once the honoring-via-naming game commences, there might be certain things we should keep in mind, or build into the process, such as fairness and desert in the distribution of the honors. There will be rules for the game. But the playing of the game itself is not required. It wouldn’t be unjust to not name anything for anyone (see the related discussion on p.16 of the paper). So I think I can maintain that honoring-via-naming is an example of possible government supererogation while still holding onto the idea that once government starts doing it, certain rules will apply.
In some cases, it might be that when the government takes the time to honor someone, they are using public resources in a way that undeservedly harms someone (say, the honoree’s victims), or forces people to engage in speech acts that could be read as honoring people they loathe. In these cases, honoring might indeed be unjust, as it would be the government infliction of a harm not required by justice. In that case, following my argument, the honoring would not be supererogatory; it would be wrong. Perhaps I am going too easy on honoring when I say I don’t think this is the typical case. But if I am wrong, that’s alright with me, as it just provides me with a new load of examples.