PEA Soup is pleased to announce an upcoming ethics discussion of Julian Jonker’s A Dilemma for Expressive Arguments Against Markets, accompanied by a critical précis by Brookes Brown.
The discussion is now open.
Some preliminary thoughts on “A Dilemma for Expressive Arguments Against Markets”
Julian has written a careful and engaging piece. I’m excited to get the chance to participate in this discussion.
Julian’s subject is expressive arguments against commodification, that is, arguments which hold that there is something problematic about what is said by participating in a market for the relevant good. These kinds of complaints, he suggests, are appealing because they advance “necessary, intrinsic, and a priori” objections to markets rather than relying on contingent empirical judgments. But, building on work by Peter Jaworski and Jason Brennan, Julian wants to show that these objections do rely on contingent empirical judgments and are thus not quite so appealing.
To show this, Julian considers three ways that we could understand the contention that markets in x mean something problematic: indication, representation, and performance.
First, we could take markets to indicate some troubling fact p (like lack of respect for x) because the market causes such attitudes. But this would require that we show empirically that markets really are reliably associated with the objectionable attitude in a way that cannot be blocked.
Second, we could take markets to represent something concerning by being part of a communicative practice according to which the relevant market or market participation has a problematic meaning. But this meaning too would be contingent because the relevant communicative practice is in principle revisable. Even when it is not, since the badness is found in the link between the communicative practice and the market, it is an empirical question whether triggering the practice is worse than avoiding the market. Consequently, it is not necessary that the market have the concerning meaning and whether we ought to revise the practice or the market is an empirical question.
Finally (and this is the move to which he devotes much of his time), Julian notes that we might take markets to involve problematic performance. On this account it is a feature of a person’s performing particular actions (such as apologizing or thanking) that they commit to having certain attitudes or intentions in the sense that they are liable for not doing so. Performances, you might think, avoid the vulnerability to collapse that besets representational or indication relations because they involve commitments to having the relevant actions and intentions that persist in their wrongfulness even if we revise the conventions of performance, and (more importantly) because markets necessarily involve particular performances in a way they need not necessarily involve particular representations.
Julian accepts that markets necessarily involve certain performances, namely, promissory commitments to exchange entitlements on the agreed grounds and to respect others’ title to goods on these grounds. And he recognizes that these commitments can stand in tension with other valuable normative commitments that we might wish to perform by, among other things, blocking our ability to express things we might wish to express (for example, as Barry Maguire and I have suggested, the discovery that Julian had sold inclusion in his acknowledgments would make it impossible for others to reliably uptake him as thanking those listed for their labor, even if in fact they really had done labor and he really intended to declare gratitude). Nonetheless, he contends that this remains an empirical objection since it depends on facts about the way that markets in some good might actually block our ability to take commitments to things like thanking seriously or block non-marketized version of the good.
I find the piece interesting, thoughtful, and far more careful than much of the work in this genre.
Nonetheless, I have several concerns.
The first has to do with the standard to which Julian holds expressive arguments. As I noted, he takes the value of expressive complaints to lie in the fact that they advance “necessary, intrinsic, and a priori” objections to the commodification of some good x. This, he suggests, is “the appealing features that expressive objections are supposed to have.”
I do not see why expressive arguments should be held to that standard.
Perhaps naively, I took it that the appealing feature that expressive objections were supposed to have is that they are true, not that they are so a priori. It is not obvious to me that advocates of expressive arguments have been suggesting that their objections to markets have this form. And if they have it is puzzling why this would be so important to point out. After all, as Julian himself notes, it isn’t really much of a mark against moral arguments to say that they rely on some degree of empirical assumptions. Most—perhaps all—such arguments do, under some description. One might advance perfectly non-consequentialist reasons as to why it is objectionable that I lock a person in my basement that will nonetheless rely on some degree of empirical assumptions (that a lock is expected to keep a person enclosed, that your basement does not encompass the whole possible universe). But it would hardly in itself seem important to note the existence of these empirical assumptions unless you were going to go on to contend that they were generally speaking untrue, as Julian does not seem to be committed to showing of expressive arguments.
So perhaps the real complaint is that those who object to markets on expressive grounds haven’t been providing the kind of empirical evidence necessary to justify their claims because they haven’t recognized the need. This would leave these views at best unjustified.
This is the most plausible reading of Julian’s claim. He writes of his argument that, “it should also be seen as a prompt to those on both sides of debates about markets to be clearer about how particular market practices are actually related to our other social practices and as a challenge to philosophical ambitions to evaluate “the market” in the abstract.”
But I think even this relatively weak criticism might still be weaker yet than Julian seems to want it to be. There are at least two ways in which those who raise expressive objections could take themselves to be justified in proceeding without providing much in the way of empirical evidence.
One is if the evidence that markets have the relevant features is pretty potent.
Consider the contention that you ought not shoot innocent people who pose no threat. This claim technically relies on empirical assumptions (that shooting someone can hurt them). But it would be ludicrous at best to contend such an argument was incomplete because some article defending it didn’t mention or provide evidence of this empirical fact. One isn’t required as a matter of fair argument to say things that any reasonable audience will take to be overwhelmingly true.
Proponents might say that the kind of evidence on which expressive objections rely is similarly obvious. They might, for example, contend that given what we all know about how markets operate in these here parts, markets in sex or organs are incompatible with performing respect to women or the impoverished, and so obviously so that the relevant premise does not need mention. Now maybe that isn’t true. For what it’s worth, I myself think it isn’t true. But then the complaint isn’t that they haven’t defended their empirical assumptions. It is that their empirical assumptions are mistaken. This means that Julian can’t really say the more tepid claim that what he is aiming to do in the article is to, “clarify what [expressive arguments] might involve and to emphasize the necessity of defending their empirical assumptions.” Unless he’s actually declaring their empirical assumptions to be deeply contentious, they can simply reply that ordinary norms of argument in applied ethics don’t actually demand that they do so.
Another is if we have a moral commitment to sincerity in certain expressions that states would be justified in promoting. Echoing Austin, Julian notes that failing to meet the sincerity conditions of certain performances is an abuse, one that renders the actions in question hollow. And Julian recognizes that markets in particular goods can by virtue of what he acknowledges to be their necessary features be incompatible with particular attitudes. For example, we might not be able to thank or mourn people if our relevant actions are bought or sold because the notion that we might be acting to acquire money can call into question whether we hold the relevant attitudes or intentions. Insofar as we have a responsibility to not thank or mourn or whatever in ways that are plausibly insincere, and the state is justified in preventing such abuses, then we do not need further empirical evidence than Julian himself provides. Consequently, proponents of expressive objections would not be acting wrongly in failing to provide further evidence. Now the claim that states have such a right is (I think) highly implausible. But that is an altogether different and stronger concern than Julian declares himself to be providing.
Here is a final condition in which it might seem odd to demand extensive empirical evidence—if we have a commitment to the relevant empirical facts being true such that insofar as they are not, we ought to be working to make it the case that they are. Were this the case, our objection to markets would not be contingent in the sense of “ought to be open to revision” even if they would remain contingent in the sense of “could in theory be revised and may or may not yet have been instantiated.”
How could this be so? It is ordinary to think that non-expressive moral concerns can limit the shape of our expressive practices. Consider that our expressions of gratitude are typically conventional and thus contingent in the way that Julian describes. They could be different. In principle, for example, it could be that the way we express gratitude is by stabbing people in the hand with a fork, repeatedly. But we have good reason—good moral reason—to not select a communicative convention that causes innocent people unnecessary pain. Consequently, it is not merely contingent empirical fact that our expressive conventions for gratitude do not take this form.
Julian recognizes that markets necessarily have certain features—that they, for example, involve the exchange of entitlements. But other features—like the contention that markets treat the goods exchanged as alienable or substitutable for money, he treats as only true in certain contexts, writing “none of these claims is true of all markets.” The presumption is that markets can be designed so that they don’t have these features, even if they cannot be designed so as to not involve the exchange of entitlements.
But of course gratitude conventions could be designed to involve stabbings. It’s just that they shouldn’t be designed in this way. If markets ought to be designed to have certain features and these features entail that market behavior has problematic meaning, then it follows that we are committed to making it the case that markets are expressively troubling in the relevant domain.
How could this be so? To see this, we need only think of other non-expressive work on the value of markets.
What is the value of markets? Why have them? One quite prominent view is that markets have value at least in part because they are effective and efficient ways to produce desirable items—everything from basic nutrition to chia pets. This is true for many reasons, because the price-mechanism provides important signals that indicate the means by which we can benefit others and ensure their access to necessary goods, because impersonal measures of interaction can reduce friction and so on. As folks like Joe Heath, Abraham Singer, Amit Ron and many others have suggested, this structure is morally valuable—arguably morally required. But markets only work to provide this value if they are designed in the right way, where the right way is suggested to require that people respond to price data, operate at arms-length in dealings, are willing to respond to relevant incentives, and suchlike rather than enacting other values like fairness or familial relations.
Now it certainly isn’t true of all markets that they are designed to have these features. But if we have a moral obligation or strong moral reason to have social institutions that play this role (because they are pro tanto good as Julian suggests is true of markets’ ability to allocate claims to things) then it is not merely contingent that markets have these features. If markets with these features thus involve certain representations or performances that are inconsistent with the proper way of relating to some particular good, then it would seem to follow that we have a non-contingent reason to not marketize that good.
Of course, as I noted earlier, Julian is certainly right to say that all this rests on empirics. For this argument to gain any traction, for example, we have to believe that markets must actually be designed with the relevant features if they are to reliably produce the desired results. But it does not follow that we need to know that much about the actual markets around us (since we are committed to markets being designed so as to have the relevant features it would follow that people are committed to their expressions being problematic in the market case.) Moreover, the kind of empirical claims at issue seem (somewhat) less controversial (or at least they could be) -bringing us back to our earlier question about exactly how much empirical evidence anti-commodification advocates are on the hook for providing and how much of a complaint against them it is that if they have not done so.
Hi all,
I also really enjoyed Julian’s article, for many of the reasons mentioned above, inter alia. And I very much agree with Brooke Brown’s construal of its central lesson: that semiotic/expressive arguments for the moral limits of markets cannot be fully undertaken from the armchair, and that their proponents need to do more to vindicate their claims, empirically.
I’d like to venture a suggestion as to how the proponent of the semiotic argument can evade the charge above. This suggestion I think is likely a controversial one, but perhaps worth considering nevertheless.
I understand (perhaps idiosyncratically) semiotic arguments to work (in such a way that they avoid the kind of collapse Julian very nicely articulates) as follows: the value of some goods are as a matter of fact incommensurable with the value of others. But they are nonetheless comparable. To borrow an example from Ruth Chang: the value of being a nurse may be incommensurable from the value of being a banker, but that does not prevent a given individual from making a rational choice as to what career they wish to pursue. In deciding who they want to be, they create the mechanism through which the two relative values can be compared. I think that markets share this function in enabling the exchange of goods and services – by putting a good (whose value is incommensurable) on the market we create the mechanism through which it can be compared with other goods that are also on the market. Putting such a good on the market then necessarily signals a kind of value choice in the sense of reflecting who ‘we’ want to be, since we could have chosen not to make that good’s value comparable via the market, or at all. Semiotic objections so understood then tell us that putting a good on the market is a poor reflection of, and does not properly reflect, who ‘we’ want to be. Now, I take Julian’s challenge to apply here since it looks like it is an empirical question as to who it is that we as a matter of fact want to be. How would the proponent of a semiotic objection then proceed? Survey the constituents of whoever the ‘we’ is? Or, and here is my controversial suggestion, might we instead think that what the proponent of the semiotic argument is doing is inviting us to imagine a picture of who we are that can then help us settle the question of who we want to be? If that’s right, then perhaps semiotic objections won’t be as decisive as their proponents may have ideally wished, but their merit is not the sort of thing that can be settled empirically. Further, they would stay true to one of the things Julian suggests is appealing about them: we could only correct the moral flaw they identify (that putting a good on the market mis-represents our values) by removing the relevant good from the market.
Very best!
Thanks to Julian for a very interesting and thought-provoking paper, and to Brookes for an excellent precis. Like Brookes, I was struck by the struck by the claim that expressive arguments are supposed to ground objections that are “intrinsic, necessary, and a priori.” I will be interested to follow the discussion on Brookes’s questions here, as I share her sense that many people who want to make expressive arguments are certainly content to see them as grounded in contingent features of the world.
But I would like to raise a question about the argument in the paper that takes the “necessary and a priori” ambition seriously. In the paper, Julian frames the dilemma as saying, “For either the expressive objection targets the contingent objectionable features of particular market participants, or it targets the way in which market participation as such is entangled with contingent social practices. This is straightforwardly the case for indication and representation, which are themselves contingent relations.” That way of framing the issue does make clear that the anticommodicationist who wants to say something “necessary and a priori” is going to need something different from indication or representation. But then shouldn’t we be looking in the kind of meaning that one finds in necessary truths? It’s been a long time since I’ve done Grice, but I take it that “indication” requires a contingent relationship. And I don’t recall what Grice would say about meaning in necessary truths. But we can understand a notion of meaning here. To say that the room is 10′ x 24′ means that the room is 240 square feet.
How could anything like that be what’s going on for the anticommodifcationist? Consider two analogies:
1. Frances Kamm characterizes rights as expressing our “inviolability.” That’s surely not a claim about indication or representation. It’s just a claim about what rights constitute.
2. Consider the sacred. Suppose that we believe that some special mountain is “sacred,” where by that we mean “belongs to God and is not ours.” This needn’t be understood as a claim about indication or representation. It’s just what it means for the mountain to be sacred.
Now, in both cases, having a market would seem to express to opposite. It would express that the we are not “inviolable” and it would express that the mountain is not “sacred” in the way just described. It would express that in a necessary and a priori way. It is, in a sense, just what “inviolable” and “sacred” mean that the thing cannot be instrumentalized. Isn’t that a kind of expressive argument that doesn’t fall prey the dilemma?
Now Julian (or Brennan & Jaworski) might reply that this is no argument at all. It’s basically just stipulative. It names a kind of noninstrumentalization (which is going to include nonmarketability) as “inviolability” or “sacred” and then QED. What we need, they might say, is an *argument* about why we should see things as inviolable or sacred. Why believe that there is anything like that in the world?
Fair enough, I suppose. It does seem like this anticommodificationist owes some further explanation of inviolability or sacredness. (Though it’s an interesting argumentative burden. Why believe that there’s anything in the world that can be instrumentalized?). I think that people are likely, at this point, to turn to claims about “meaning” in the sense that concerns importance, not signification. And that’s going to present hard questions in making out the view.
But I guess that my question is why not understand (some) expressive arguments against markets in roughly this way—as making neither a claim about indication nor a claim about representation? They are making a claim that the market is, as a necessary truth, an expression of instrumentalizability, violability, or nonsacredness. There’s a big remaining question about why we should care about some things being inviolable or sacred. But it seems like the expressive argument would not, per se, face the dilemma that Julian lays out.
Let me start by expressing my sincere gratitude to PEA Soup for hosting this discussion, and to Brookes, for her characteristically sensitive and insightful reading of my paper, and also for the previous work she has done on this topic, which has been a model of clarity and rigor. Her critique reflects her deep familiarity with the argument and its vulnerabilities, and I have already learned a lot from it. Thanks also to Nico and Tony (and anyone else who posts while I’m editing this) for their thoughtful comments–I’m off to teach in a minute but hope to respond by afternoon (Eastern).
Perhaps this is as good place as any to emphasize something that might not come through as clearly as I’d like in the paper: my engagement with expressive objections comes less from thinking that they are seriously flawed than from finding them extremely interesting, though often underdeveloped. While I have framed the paper somewhat polemically, part of its work aims to set out, as clearly as I can, a plausible way of understanding expressive arguments in ethics. Now, I argue that even on this understanding, expressive arguments are less useful for anticommodificationists than they have supposed. But I think expressive arguments may be very useful elsewhere–in understanding discrimination, for example. Most of all, I would like to this sort of argument disciplined so that it can be put to its most effective use.
With that out of the way, let me take Brookes’s concerns in turn:
1. **The wrong standard?**
Brookes begins by suggesting I am holding expressive arguments to an overly strict standard. I claim that the (false) appeal of an expressive argument is that it appears intrinsic, necessary, and a priori–more crudely, that it is a knock down argument from the armchair. Brookes responds that the appeal of an expressive objection is surely that it is *true*. But I wonder if that could be all there is to its appeal. Of course we should do our best to deploy only objections that make true claims, as we should do our best to only make true assertions. But objecting is not just asserting. An objection is a self-conscious move in the game of giving and asking for reasons, which is to say that it is not just a statement of a fact but a part of a justificatory dialectic involving advocates and detractors who are willing to adjust their views and question their opponents’ views in response. That makes the idea of the objection’s force of special interest: whether it has special justification that is not prone to being easily undermined, or whether it persists over a wide range of refinements. An argument that markets in sex work are bad independently of the evidence about particular instances is naturally of greater interest to the anticommodificationist than the objection that markets for sex work in Philadelphia are bad because of unique demographic and economic factors. Both may be true, but the former are more often found in philosophical works of broad ambition. And it’s that broad ambition that I am skeptical about.
2. **Obvious Facts**
Brookes then turns to demonstrating the apparent weakness of my criticism of expressive arguments. First up is her point that some empirical assumptions about the market are obvious and don’t need mentioning, just like the empirical claim “shooting someone hurts them” assumed by the moral judgment “you ought not shoot innocent people who pose no threat”. But this analogy fails in a revealing way. The causal mechanism involved in “shooting hurts them” is well understood and hard to doubt. The causal claims involved in expressive arguments are nothing like this. Consider the paper’s case of a market for mourning services, which threatens to crowd out our ability to mourn genuinely. Of course it *conceivably* might, but whether it does or even poses a real risk of doing so depends upon the size of the market, our ability to distinguish paid and unpaid mourners, and our propensity and ability to adapt our mourning practices in response to the market. These factors are not obvious, and suggests a non-Newtonian relation between market and harm, unlike that which operates in the ballistic case.
Indeed, the history of commodification debates is full of cases where seemingly “obvious” empirical premise turned out to be seriously contested. Does paying for blood donations actually erode altruistic motivation? Titmuss predicted that paying blood donors would crowd out altruism, but empirical evidence of crowding out has been lacking or negative (see most recently Bruers’s review in *Transfusion Medicine Reviews* 2022; a landmark study showing crowding *in* is Lacetera, Macis, and Slonim in the *American Economic Journal* 2012). Predictions that commercial surrogacy would be exploitative have been complicated by research on surrogate mothers’ own experiences of the practice (e.g. Kneebone, Beilby, and Hammarberg in *Reproductive Biomedicine Online* 2022, or Yee, Hemalal, and Librach in *Women and Birth* 2020). The claim that markets in organs would undermine respect for bodily integrity rests on assumptions about social psychology that are far from obvious. Does a market in sex actually degrade the practice of sexual intimacy? Do markets in organs undermine respect for bodily integrity? These are the subject of hotly contested empirical research programs, and confident assertions of supposedly commonsense expressive concerns by philosophers do not advance them.
Let me grant that there will be limiting cases in which the empirical premises are obvious. An auction of people poses threats to the well-being of those auctioned that are well-documented and easy to predict. We don’t need a randomized controlled trial to know the results will be bad. But we also don’t need a distinctively expressive argument. The best objection to such a practice is not that it says something bad, but that (as in the shooting case) it disrespects and harms the people put on the auction block. If there is an expressive objection, it seems to simply (in Niko Kolodny’s words) “recapitulate” the thing that should really concern us.
3. **Sincerity and the State**
Brookes’s next move is ingenious: the state might have justification for promoting the sincere public performance of certain commitments (perhaps to civic participation or social solidarity); such that if a market undermines the conditions of sincere performance, the state has reason to prohibit the market. The force of this is that the state’s standing obligation to guarantee the availability of sincere performance overrides any empirical verdicts about whether markets in fact undermine performance.
The topic of the state’s obligations with respect to our performative repertoire is one I find interesting and hope to explore in further detail, so I am grateful to Brookes for presenting it. I hope that it is a topic my performative framework helps to illuminate–and I suspect that it does not rescue the armchair ambition of expressive arguments, for the following reasons:
First, whether the state has justification for promoting or even protecting certain performances is a substantive normative claim that would likely be open to dispute. One site of dispute is likely to be how robust the relevant practice is and whether the state’s intervention is needed or even counter-productive. This seems to me likely to raise empirical questions of exactly the same kind raised by crowding out arguments. Arguably the state protects and promotes our promissory commitments through the statutory and judicial regulation of contracts, and yet (as the philosophical and legal literatures on promising suggest) it is unclear whether contract accommodates, facilitates, or crowds out our sense of promissory obligation. Related to this, we would likely want to draw tight constraints around the state’s reasons for protecting sincere performances, which are otherwise likely to threaten other non-market practices (e.g. protecting our sincerity in promising does not seem to me a reason to prohibit pre-colonial marital practices of bridewealth). So we would want a defense of the state’s justification to pay careful attention to the experience of which activities do and do not threaten sincere performance. Second, supposing the state does turn out to have such justification, it is unclear whether the state’s actions are triggered in a case in which a market does not in fact undermine sincere performance. So here we again require an empirical premise.
4. **Commitments to Design Constraints**
Brookes suggests that we have a commitment to making certain empirical facts true. The dialectic here is: I worry that an expressive objection is contingent upon whether it makes practical sense to revise our practice so that it no longer has a negative significance; Brookes responds that even if we can revise the practice in this way we should not because of a standing commitment to the practice being such and so. Now, I worry that there is something question-begging here: whether we have a standing commitment to the practice being such and so must surely take into account various contingencies such as whether the practice is harmful in its current form, or whether it has a negative significance. So this just seems to push back the empirically-inflected question about the practical revisability of the practice.
I wasn’t sure how to map Brookes’s striking example of a practice of gratitude that involves stabbing each other with forks onto the issue, so let me work through a more familiar case. A market in kidney exchange might be thought to have negative expressive significance insofar as represents participants as not respecting bodily integrity. The revisability move can work in two ways here: one is to design markets so that their norms are closer to donation than exchange (perhaps an organ recipient must be related to the donor); another is to change attitudes about what counts as bodily integrity (perhaps by way of an advertising campaign). I take Brookes’s counter to be that the first way of going might be blocked off because there is good reason to have markets that promote efficient allocation, and requiring that a market be designed so that it makes exchange condition on blood relations would undermine the allocation mechanism. Perhaps, though I think this is again an empirical issue. Indeed, this is a place where critics of the market appear to join in an ironic alliance with the neoclassical Walrasian tradition in viewing markets through the lens of a frictionless ideal of the price mechanism. Real markets diverge from that picture in all sorts of ways, and how they perform seems to me largely an empirical matter. One reason we should not have a standing commitment to establishing markets with certain design features is that experience, and the empirically grounded economics that has largely superseded Walrasian welfare economics, show that there are a variety of ways of designing price mechanisms, and complicated issues about how they interact with other values we care about. Philosophers should absolutely be involved in this discussion about institutional design, but they have to find ways of integrating normative analysis with the relevant empirical research programs. I worry that expressive arguments have led us away from that task.
At this point everyone will surely have noted the irony that my paper is that I am talking about markets and expressive significance at exactly the level of abstraction that I want to caution against. I readily concede that, at this level of abstraction, I could easily be wrong. In any case, I find myself much edified by Brookes’s response, and must reiterate that her work here and elsewhere is a model of the care and insight that will help us further our understanding of how expressive arguments and arguments about markets work.
I am grateful for the opportunity to participate in this conversation about Julian’s excellent, important article about the expressive significance of commodification. I want to weigh in (quickly) on Julian’s arguments that the meaning of a market is unavoidably contingent. Julian contends that, insofar as the meaning of a market is revisable, it is always open to us to ask whether the benefits of the market outweigh its negative expressive outputs, in which case we should seek to revise the meaning rather than prohibit the market.
When I teach commodification to my students, I seek to generate in them the intuition that some things ought not to be bought and sold for expressive reasons. The example that seems to work best is “dwarf bowling” or “dwarf tossing”—i.e., the practice of using little people as projectiles. One feature of the example that seems relevant here is that the meaning of the practice cannot be revised without abolishing the practice itself. What is problematic about the practice is that it is meant to degrade the people who are tossed around; those who pay to watch this dubious form of entertainment would have no interest in it if instead we came to view throwing people around as a way of bestowing our utmost respect.
Perhaps the example cheats in a sense. It isn’t so much about commodification as it is about objectification. The practice would be no less troubling if little people instead offered themselves up as projectiles free of charge. So the marketization itself seems to have little, if anything, to do with the troubling expressive significance. Even still, one might think it notable that the meaning of some of the things we exchange is immune to the kind of revision the anti-anti-commodificationist would urge on us.
At any rate, I can think of other degrading practices where the degradation does consist in significant part in the monetary transaction. Take paying for sex, which Julian discusses at numerous points. It seems plausible to me that sometimes what is being offered for sale is the degradation of the prostitute, and that the exchange of money effectuates, and is meant to effectuate, the degradation. By paying money, the John communicates that he owns the prostitute, or that she is chattel, or some such. As with the dwarf bowling case, I don’t see how we could change the problematic meanings without obliterating the thing the buyer is after.
I do not believe that Julian’s discussion of Blackface (footnote 17) dispels my take on paying-to-have-sex, though I welcome his transposing the insights there to my example. If I am right, then sometimes the wrongs of commodification are not contingent on representations, or the interactions between markets and other social practices. Long live armchair philosophizing?!!
Super interesting discussion already! I too am off to run a few tasks, so writing very much on the fly without reflection, but had one immediate reaction to part of Julian’s response that I thought might be worth noting.
This relates a bit to my sometimes not feeling clear on what is at issue in these debates. Why is the example of the organ that Julian suggests above not precisely a concession that semiotic objections are correct about markets? That is to say, if we say you can’t design institutions with regards to organs like classic markets because they will have negative expressive significance instead we have to design them to be more like donation, why is that not just to concede that commodification does have problematic meaning? Precisely because of that negative meaning we have to design an institution in this realm that has different features and make it clear that this institution is NOT a market (where the market has the classic features, and has them for all the kinds of reasons that make those features good at efficiency etc). That is to say, we had to make it so that the good could be exchanged without being commodified. Of course not all exchanges are commodification — I imagine everyone in this debate concedes that gifts and donations are different (I don’t think those who oppose the sale of organs oppose organ donations, for example).
Thanks to Julian, Brookes, and everyone else for this rich discussion.
Julian, I’ve got a question that is a reaction to responses (1) and (2) that you make above to Brookes–it may be tangential, but that needn’t preclude it from being useful (I hope!):
I’m wondering how you think the expressive arguments that are your target intersect with virtue-ethical characterizations of your various cases. If I have my Anscombe hat on and read Brookes’s shooting case, I think: “Well, it is obvious that it is unjust to harm innocent people: that’s just part of what ‘unjust’ means. It is also obvious that it is bad to be an unjust person, for the same reason. So, it’s bad to shoot innocent people, because shooting people almost always harms them.” The “intrinsic, necessary, a priori”-ish element comes in here via the virtue characterizations of what is and is not just and of it being bad to be an unjust person; the facts about shooting and harm are empirical elements.
With my Anscombe hat still on, I go through your follow-up cases and think: “It is equally obvious that a mourners market would be a bad social institution. A paid mourner mourns dishonestly (or if they mourn honestly, it is only by accident). Dishonesty is a vice, because it is bad to be a dishonest person. A practice that pays for dishonesty is thus vicious, and thus bad. Whether the bereaved feel better or not is beside the point.” When I get to the paying-to-donate-blood case, however, it isn’t as obvious to me what to say, because the commodification there doesn’t obviously institutionalize a vice such as, e.g., injustice or dishonesty–or at least, I can’t immediately see the relevant vice. I’m curious if you think this helps at all with thinking through expressive arguments or whether it is simply an alternative and unrelated approach.
Wow! I’m back from teaching (about global trade and inequality!) and I’m feeling humbled by the close attention the paper and discussion has received from colleagues I respect so much.
@Tony: thanks for this interesting suggestion. I agree that “who we want to be” is an empirical question, and I worry that it will be quite difficult to establish, in the way that, even after wading through legislative histories and so on, it can still be hard to say why a particular law passed, let alone why a practice (say: event contracts betting on the death of the Iranian Supreme Leader) was allowed to emerge even amidst much disagreement. I think I see what the expressivist wants to say in this context: that the fact that the market exists suggests (“says”, etc) that we agree about a value judgment we are not in fact in agreement. My worry, stated very flat-footedly, has always been that the expressive argument says this in a rather clumsy and roundabout way. Instead we (i.e. we critics) should just say: here are the reasons that we disagree with death bets. The expressive language feels like an underhand attempt to escalate an objection from something whose importance we disagree about (“death bets create moral hazard”) to something about which we should not disagree (“death bets disrespect human life”). Now, Tony’s suggestion is really interesting because it acknowledges that the expressive argument is itself performative (it is an “invitation”), and suggests that the relevant performance is something more cooperative than what I just suggested (i.e. it is not underhand escalation). I think I can concede that that is what some expressivists are up to, though I think I am still right to point out that doing so does not exactly count as a good argument and that philosophers–even philosophers who are public intellectuals!–should proceed with caution.
This is a good place to note that, biographically, Tony’s thought provoking and rigorous piece about this issue in *Analysis* was what prompted me to start writing on the topic, and I’ve tried to say what I could about it in my JESP paper. Thank you, Tony! It’s wonderful to be continuing the discussion here.
@Nico: great, this sent me back to my Grice. As I recall, Strawson and Grice (in *Phil Review*, 1956) defended the analytic/synthetic distinction against Quine by insisting upon the idea that analytic propositions can be falsified only by revising meaning–which is a way of saying that the analytic is true in virtue of meaning. I take it the upshot for the Gricean view is that “means” in “the room is 10’x24′ means that it is 240 square feet” is standing in for an entailment relation which obtains in virtue of the non-natural meanings of “10’x24′” and “240 square feet.” You can guess where I want to go with this: revisability is still on the table because markets are not necessarily part of our semiotic repertoire. Any role that they occupy is one we have conferred upon them, and thus revisable.
But I fear this sort of response is in danger of missing Nico’s point, and that it’s better to consider actual cases. So, about the mountain: I think the argument is that selling plots on a sacred mountain would say something objectionable, since “sacred” means “belonging to Us” and so selling plots means “not belonging to Us.” I have a similar worry to the one rehearsed in response to Tony’s suggestion, namely that the expressive thing is a roundabout and underhandedly escalatory way of saying something more ordinary. The ordinary thing is: selling plots violates Our common ownership of the mountain. Why not say that? Doing so illuminates several points that we can make progress on: do We own the mountain? Does a market in mountainside property violate Our common ownership? etc.
Once we are clear on what’s at stake, we might decide to (i) revise our notion of “belonging to Us” (maybe it means democratically regulated rather than non-saleable), or (ii) revise the kind of market we’re willing to accommodate (maybe Our common ownership is consistent with 99 year leasehold. Response (i) is also a way of revising the representational role of the mountain–given the analyticity of “sacred means belonging to Us,” we might be abandoning our commitment to the mountain’s being sacred. I think I can, with perfect consistency, admit that this is an expressive change that my performative framework illuminates, without conceding that the expressive objection against the market had much bite. Saying that we’ve changed the expressive significance of the mountain in order to accommodate a market in mountainside property does not really say much about whether we’ve done the right thing or not. For that we need to make non-expressive arguments about whether the mountain should be subject to common ownership or not.
@Amy: the “dwarf tossing” example is indeed powerful. Let me begin by insisting on its disanalogies with commodifying practices, in line with my discussion of Blackface. It seems to me that dwarf-throwing is a species of insult, and that insult is a kind of expressive performance. Indeed, I hope that the positive dimension of the paper is that it offers a way of understanding this sort of non-revisable expressive significance.
But then doesn’t a market in dwarf tossing inherit the objection? I’m not so sure. Suppose (even though this is one of these performatively self-reflexive philosophical cases where one worries about committing the wrong by discussing it) that I am a ticket scalper and I am speculating on tickets to a particularly popular dwarf tossing contest. Am I insulting little people? I only intend to make money, and know nothing about the contest except that I can buy tickets at high prices and sell them higher. Maybe I am morally reckless but it seems to me contestable that this is insult. Amy, I imagine you think I’d be complicit in insult, and perhaps that is a kind of expressive objection–but does it follow from market participation as such, or from further contingent circumstances? What if I’m not scalping but buying up all the tickets so that I can call the whole thing off? Consider also the over-eager date who buys tickets to a contest on his partner’s advice, not knowing what he is getting himself into. He participates in the market but I don’t think he commits insult. A more pressing version is of the Hollywood writer who uses a dwarf tossing scene to sell a movie about financial excess, but any expressive concern about the movie seems pretty tightly linked to its representational form.
Part of my point is that markets are not inherently representational, whereas dwarf tossing and scriptwriting are. The reason that paying someone for sex can sometimes be degrading (as it can) is because the buyer deploys the payment of money as part of an insulting performance. (People throwing dollar bills at strippers, at least as depicted in movies, has always struck me as such.) But this is an individual performance that very much depends upon the agent’s attitudes. Participation in the market as such–which is what I take to be properly the target of the anticommodificationist–does not seem to me sufficient to count as insult even in the case of markets for sex work.
@Graham, I wonder whether this last remark, distinguishing individual behavior from institutional actions as such, is relevant to your point. We can certainly find cases of individuals participating in markets which indicate their bad character, and I take your point to be that we can often (perhaps as a matter of second nature) simply *see* that someone is behaving viciously, so we needn’t gather further evidence. I think I can concede that without admitting the rest of what you are suggesting: that therefore a social institution that allows such displays of bad character is itself bad. Why would that be so? It would be bad if it did more than allowing: facilitating or promoting or cultivating bad character seem bad traits of an institution since they further bad character rather than simply accommodating it. But all those claims about an institution are empirically contestable and require evidence. And, as I suggest in the paper, the anti-commodificationist is surely targeting the institution, rather than simply instances of bad behavior.
I hope this connects with your line of thought, as I am eager to think more about what we can say–as a matter of philosophy–about the relation between character and institution.
@Brookes: thanks for the illuminating follow up. You’re saying that by revising certain kinds of exchange away from fully commodifying norms, as in the organ case, I am in fact conceding that a fully commodifying market would be subject to the expressive objection. I think I want to simply concede this, though I’m probably in danger of making my claim even weaker than you feared! But here’s the thing. I am not trying to defend anti-anticommodificationism, and I am not even trying to defend it against expressive objections. I do think that expressive arguments need to be better articulated and qualified, in the sense that they should acknowledge their assumptions and their limited scope. If that is weak tea it is because I am largely in the business of explaining to myself how these arguments are supposed to work–again, I hope that people notice the positive proposal in the paper as well as the negative claims. So: once we have determined that markets do involve performances that in fact crowd out non-market performances given empirical conditions, and that this is bad, then I think I am satisfied. It strikes me that the kind of expressive argument made at this point is quite different–something much more definite and more compelling, even though less ambitious–than the cry that the market says something objectionable. I do also suspect that such arguments are going to be hard to establish, but that is empirical speculation on my part.
Hi and thanks to Julian for writing such an interesting paper, and to Brookes and everyone else on this thread for such stimulating discussion! At least here in Europe, this site was down from when I woke up until just recently, which is why I am chiming in late; I apologize.
Like pretty much everyone else in this discussion, my main sticking point with the paper was with the claim that expressive objections to a practice need to be necessary, intrinsic, and a priori. Like Brookes, I think this standard is both too high and also unnecessary; I just don’t know why we would want this out of an objection. But my concern goes yet deeper; it’s not just that this kind of a priori necessity is unneeded; I don’t think that I would accept any account of meaning or expressive content that could ever meet this standard for anything. I just don’t think of meanings as entites that have some kind of determinacy to them; rather, meanings, I believe, are necessarily things that are constitutively embedded in the messy, contingent world of discursive and other institutional and social practices. I can’t imagine what it would be for an expressive objection to be a priori, because this would require that we determine what is expressed by an act or practice a priori, and I don’t think *anything* has a priori determinable expressive meaning of this sort. Even the most analytic-seeming statement only is that statement in the right practical context, and we could always make up a context in which it didn’t mean what it usually does. The act of commodifying something or selling something just isn’t the same act in the context of an anarchist commune structured around mutual aid; on a street corneron the dark web; over Amazon; or in a department store in a late capitalist economy structured by multinational companies and deep inequalities within and between states. I am seriously dubious of the claim that there is some core expressive meaning to “commodification” or to “selling” or “buying” that could survive all these different contexts.
Julian – interestingly and rightly, in my view – focuses on the expressive content of the *performance* of buying/selling something. I think that this is already a rich and interesting topic. Doing things expresses presumed attitudes and values in the act of doing them, and we might find these attitudes and values objectionable. But here the idea that the expressive content could be a priori is even more mysterious to me. I get that some people are realists about determinate mental representations with semantic content that can be expressed, and I suppose I can imagine someone finding such a representation objectionable regardless of empirical context. But when we are talking about performative expression, we are talking about expressive meaing built into a public, social action. Surely, such actions are what they are only because they are entangled with other actions and the world around them, and structured by rituals and iterations that give them their public significance. Just waving one’s body around or even blowing various phonemes out one’s mouth will not constitute any kind of performance with any expressive meaning (other than maybe causal indicative meanning) at all. The performance of promising only is such a thing as it is embedded in a practice of promising, within a culture of interpersonal trust and responsibility, blah blah. The performance of buying only is such a thing as embedded in an ecology in which some things function as money; we have entrenched methods of tracking people’s debts and assets; there is some institutional set-up that determines price and holds people accountable to contracts; and so on. This all seems uncontroversial. But this is all empirical stuff. We can’t know anything about the expressive meaning of a performance unless we know what performance it is, which requires knowing all this type of social context. So, I would think, expressive objections will always be empirical!
This makes sense given the sorts of test cases Julian wants to talk about. Is it objectionable to commodify sex? Surely that depends on how sex works within a cultural context, richly understood. Most people don’t object to breeding cattle for money (some do, including me, but for very different reasons than the ones on the table here), so it’s not the mere idea of sex that’s doing the work – it’s sex within specific contexts and with specific meanings. The objectionability of commodifying sex among a group of friends of roughly equal power who are playing a completely consensual gambling game of sex poker, with a buy-in of 10€ and a limit of 20€ (if you find that objectionable) is dramatically qualitatively and quantitatively different from the objectionability of paying a pimp 5000€ for a night with a sex-trafficked 18-year-old who can’t speak the local language and will only be paid in drugs. And I would insist that this is so even if no one is harmed in either particular case. It seems unimaginable to me that anyone would think that there may be some common a priori meaning here to object to in both these cases. Moreover, if there were some such thing, that would not be what was morally interesting. The morally interesting issues around buying and selling sex, including the expressive issues, are all issues that turn on just this sort of empirical detail. But again, remember, I am also arguing more fundamentally that there is and could not be such a thing as a priori knowledge about what commodification of sex expresses, because commodification is only the act that it is, and sex is only the act that it is, in virtue of the contingencies of the social and material context within which people are moving their bodies around.
I think that maybe there is an elision pervading this discussion that might help explain Julian’s original intuition. There is an important difference between saying that you need empirical evidence of *particular incidents of harm* from an act in order to obect to it, and saying that you need empirical evidence in order to tell whether it has harmful expressive content. In responding to Brookes, Julian writes, “An argument that markets in sex work are bad independently of the evidence about particular instances is naturally of greater interest to the anticommodificationist than the objection that markets for sex work in Philadelphia are bad because of unique demographic and economic factors.” Ok, but the first half of this sentence says something weaker than what he seemed to argue for in the paper. I find it very plausible that I can deem an instance of sex work (expressively) bad even if I don’t have any evidence that *this particular* instance harmed anyone; indeed I suspect this is common. I can accept this while at the same time thinking that I can only determine its expressive meaning and object to it if I know a bunch about the context in which it happened – maybe not the details of philadelphia demographics (although this would add expressive nuance), but a bunch of stuff about how the socially rich practices of sex, sex work, underregulated markets, etc. work in countries roughly like the US. All of that is empirical knowledge. I suggest that maybe Julian got captured by the correct and interesting realization that those giving expressive objections want those objections to hold even when we have no empirical evidence that the *particular* act with the supposed expressive content did harm, and he inflated that into the idea that those giving expressive objections want them to hold *without any empirical context or evidence at all.” I think normally people giving such objections want the first but not the second. (Consider all the debates about 15-20 years ago over the expressive objection to prenatal testing: many bioethicists argued that even if a particular test harmed no disabled person, the expressive meaning of normalizing these tests would make life in general worse for disabled people and those that loved them, in the context of ableist culture.) And I think that Julian is right that they can’t have the first, but I don’t think he has good reasons for being dismissive of or disappointed by the second.
@Quill: thanks for this rich response, characteristically grounded in sensitivity to our tangle of actual practices. I’ve learned a lot from your work on our communicative practices, and indeed I hope to inhabit your pragmatism about meaning. So I hope to show that our positions are closer than might appear, without (I hope) appearing too evasive.
Here’s where I think we are in agreement: you write that “meanings … are necessarily things that are constitutively embedded in the messy, contingent world of discursive and other institutional and social practices.” I certainly agree that claims about expressive significance are grounded in our contingent practices, and my background assumption is that “the market” is another set of messy contingent practices, such that the interaction between these practices is unpredictable. I take your vivid illustration of the distinction between the contexts of sex-poker and sex trafficking to be a sign that we agree that significance is a matter of social ecology. One wrinkle here is that I do think there is something promisingly non-contingent about performances: once we have fixed on a contingent performative practice, then someone’s performance has a non-contingent (because normative) relation with the attitude in that performance’s sincerity condition. If I make a false promise using the conventions of our promising practice at t_0, I cannot rely on a revision to that practice at t_1 to say that I’m not liable for my insincerity. And that is true of a false promise made in the context of exchange too. But this is a mere wrinkle, because while my insincere promise in the course of market exchange might (despite my insincerity) appear to commit me to some commodifying norm of exchange, my claim is that the expressive performances constitutive of markets have a messy relationship with the other practices that affect their significance.
So what do we disagree about? You say you find the demand that expressive arguments be intrinsic, necessary, and a priori mysterious. I think I understand why that demand is appealing, though I think it cannot be satisfied. But insofar as you disagree with that, you must find my view of anticommodificationists uncharitable. Perhaps I’ve erred in that way (though I think not: When Sandel argues that a terrorism prediction market expresses a dehumanizing attitude, I take him to be making a claim that holds regardless of social context); still, I hope at least that in doing so I’ve performed the service of clarifying how these arguments must be working.
But in the course of your comment you add a very interesting error theory of my supposedly uncharitable view of anticommodificationists: that I’ve inflated the true claim that anticommodificationists think they do not need to establish evidence of harms in particular instances of a market practice into a false claim that anticommodificationists think that they do not need to establish any empirical context at all. I think at this point you mean to say (I wasn’t sure how to parse the final sentence) that while I am right they can’t have the second, strong claim; I would be wrong to say that they cannot have the first, weak claim. The way I put the weak claim in the paper is: anticommodificationist want to say that the generic act of participating in a market is objectionable independently of what is going on with a particular participant. And while I think you’re right that expressive anticommodificationists want to make this weak claim and *can* make it, it does not seem to me a completely obvious point. Sometimes anticommodificationists make expressive objections that sound a lot like the indication relation (e.g. that market participation indicates an objectionable attitude), and I don’t think this can be established without evidence regarding an individual participant’s attitudes, or at least the probability that they have the objectionable attitude. Indeed, I think it is a substantive claim that has not always been clearly articulated that an action must be a performance, in the sense I characterize, in order to be able to say why it has expressive significance independently of what is contingently going on with the agent. As I’ve said earlier in the thread, I hope people take up this positive claim of the paper as much as my negative claim about the force of anticommodificationist arguments–though I readily admit that this will not seem particularly novel in light of your own work on performative significance.
To conclude, we appear to agree that viable expressive arguments, which have the virtue of abstracting from participants’ attitudes and other evidence about individual transactions, nonetheless rest on empirical claims about context. I suspect we will diverge on what counts as good evidence, but I hope we can also agree that anticommodificationists could do more to articulate their empirical assumptions–to paint a richer picture of the markets they are describing. Perhaps we can only get clear on that requirement through the practice of writing about particular markets–and I do look forward to exploring that.
I’m not sure when comments close, but I wanted to be sure to thank everyone again for engaging in this discussion with me, and for paying such generous and thoughtful attention to the paper. Thanks especially to Brookes for the detailed and sensitive comments, and to Kerah and the rest of the PEA Soup team for hosting (and platform maintenance). I’d be happy to continue discussion by email with anyone who comes across this after the comments close!
Julian, I am also not sure when comments close! But thanks for your rich response! I will just quickly say that, first, it does seem like we mostly agree on things. But second, I am not convinced by your claim that “One wrinkle here is that I do think there is something promisingly non-contingent about performances: once we have fixed on a contingent performative practice, then someone’s performance has a non-contingent (because normative) relation with the attitude in that performance’s sincerity condition”. I’ve spent a lot of time arguing that the force (and with it the expressive content) of performances is temporally dynamic and unsettled, and partly fixed by uptake, which can retroactively shift or determine the orginal act. See, for example, my example of the marriage proposal which may be a joke or a serious proposal depending on uptake, which I used in my now-kinda-old “Performative force, convention, and discursive injustice” paper but have repurposed since then. So, I think even this connection is contingent and depends on empirical facts.
Sorry my previous last sentence was really contorted. I said, “And I think that Julian is right that they can’t have the first, but I don’t think he has good reasons for being dismissive of or disappointed by the second.” – I meant, expressive objectors are right that expressive objections don’t depend on showing contingent harm in particular cases, but they should not feel that their objections are weaker or disappointing if they depend on other sorts of contextual empirical knowledge.
But yes, lots of agreement here, and I appreciate your paper and this wonderful discussion!
Hi. This is a wonderful discussion of a fabulous paper, and I’m only sorry that, for personal reasons, I had to join the party so late.
By now Julian has replied helpfully to some of the reactions I shared with Brookes Brown. But one of my remaining thoughts echoes Nico’s remark about the mountain, and is illustrated by Amy’s example of objectionable “tossing” practices.
As someone who’s used expressive arguments in my work, I have to report that for me, at least, they often do not involve any of the forms of expression that make up Julian’s dilemma. When I target an act for committing expressive harm – like consistently ignoring one employee’s (or student’s) contributions as opposed to all the others, say — I do not mean it involves any social practice of representation, revisable or otherwise, or indicates the presence of a problematic attitude, or engages in a performance whose sincerity condition is problematic.
Instead, I use “expresses” in the sense of “treats someone as though.” That is, I use “φ-ing expresses P” in the sense of “φ-ing treats S as though P.” And, in the spirit of what Nico was aiming for, I think the relation between many behaviors and how they treat people “expressively” is primitive and necessary, bypassing communicative practices or commitments. Not always, but often enough.
So, if you talk to me and I don’t really listen, but instead keep looking around the room at other people, or if you make plans with me and I always flake or come late, my actions commit an expressive harm only in the sense that they treat you as though you count for less. The actions themselves simply have that implication, and without any appeal to what these behaviors represent by some convention or other, or what attitude they suggest I have. If, walking briskly, I accidentally stomp on your foot and fail to stop and acknowledge you in some sense, I treat you as though it’s perfectly fine to injure you. Like it’s nothing to fuss about. Or at least this is how it seems to me with such behaviors.
The paradigm is a slap in the face. We call disrespectful actions “a slap in the face” because the behavior is indisputably disrespectful (I don’t mean the light playful kind you sometimes see in certain cultures).
So if commodification treats people as though they are less sacred or inviolable, or as objects (in Amy’s examples), then they would express this problematic content regardless of the representative or performative practices they implicate, if any.
This is not an objection to the argument Julian makes, which I think handles the various horns of the dilemma he presents quite compellingly. And, I should add, most proponents of expressive arguments that I have read use it in ways captured by that dilemma. But I’m pushing this “treating as” interpretation as one more possibility to consider, which may lie beyond the paper’s framework. If so, what do you make of it?
Thank for the response, Quill, and thanks for the comment, Jeff. I’ve only now found time to return to this discussion, though everyone has surely moved on.
@Jeff: I do find the “treats as if” relation a tempting way to think about a certain kind of expressive significance. But on further thought I find it an elusive way of talking.
One way to take it is as a somewhat inflated way of talking about treating people badly. If I look around the restaurant while my wife is trying to tell me something important, I do not just treat her as if she were less important, I treat her as less important–certainly a relational injury. We’re free to call this an expressive injury, but I’m not sure what special work “expressive” is doing.
Other cases are more ambiguous. Suppose I am in a rush and step on your toe, not turning back to acknowledge you when you yell. This sequence is compatible with several possibilities. One is that it expresses my disregard in the sense that it is evidence of it—this is expression as indication. Another is that the sequence involves performances i.e. actions that are constituted by my commitment to a norm of disregard. I am happy to admit both of these expressive objections to our moral vocabulary, though I think they are unhelpful to the anticommodificationist for the reasons the paper gives.
A third possibility is that neither of these is true, but that my action strikes you or others as the kind of thing that would ordinarily indicate or perform disregard. I am worried about this sort of objection. I might be rushing to the emergency room, or I might be hard of hearing. It’s unclear why I should be held guilty by association with those people whose actions do indicate or perform disregard.
Now, I actually think that my account of performances is a way of articulating what is appealing about the idea that a certain action can express a proposition by treating another as if some norm held. But it is more narrowly applicable to just those cases in which a person has engaged in a distinctively performative act. I might involuntarily move my hands when the President’s calvacade motors part and this may or may not be evidence of my attitudes; but if I make a rude gesture–if I perform an insult–then, even though I have a detached or ironic attitude, I perform a category of action that is constituted by the fact its felicitous usage requires sincere contempt. That makes it reasonable to say that I treat the President as if I hold him in contempt–regardless of what my actual feeling is. I think we agree that this is a useful thing to be able to say. My further thought is that the only things we do like this in markets is exchange promises and respect property rights, and that there are no further moral implications we can draw from that fact alone.