Eden on Arneson
About a week ago, I posted a link to a new paper by Richard Arneson, entitled Democracy is Not Intrinsically Just. John Eden writes with some interesting comments:
The basic idea that democratic arrangements are not self-justifying is quite powerful. For far too long democratic theorists have assumed, rather than argued for, one of the most important ideas underlying the democratic ethos: namely, the idea that the legitimacy and justifiability of democratic procedures has little to do with the outcomes such procedures enable. Arneson’s recent paper is an attempt to disabuse us of this prejudice, and, in the process, shed some light on why considerations of human welfare and well-being ought to play a more direct role in justifying and legitimating democracy. I’ve one main worry about the way he goes about making his case: his argument mistakenly frames the debate in terms of whether we have democratic rights, rather than in terms of whether democracy is valuable in an instrumental or noninstrumental sense. The question is not ‘Do we have a right to democratic practices?’, but rather ‘Is democracy presumptively valuable?’
As a friend kindly pointed out to me, Arneson does not necessarily believe that democracy isn’t presumptively valuable. He does hold, however, that the extent to which democratic processes are valuable isn’t as relevant as whether we have a right to a democratic say. I will argue that it is a mistake to frame the debate in these terms. My main line of argument is simple enough: If one begins, as Arneson does and as we all arguably should, with the idea that standards of justice are independent of, and sometimes chafe against, democratic ideals and practices, then it analytically follows that our all-things-considered judgments of the SUCCESS of democratic procedures will fluctuate as a function of how closely (or not) democratic processes track the demands of our conception of political justice. But this is not to be confused with the quite different, and considerably less plausible, suggestion that the VALUE of democratic norms and procedures depends exclusively on whether they are congruent with the demands of justice.
What is the Debate About?
Arneson’s argument is that democratic procedures are not in themselves just unless they tend, on balance, to produce results that we all find (roughly, in certain cases) congruent with a well-developed sense of justice. In other words, if the best intuitions provided by political morality are consistently contravened by a set of democratic procedures, then something is obviously wrong with relying exclusively on those procedures to structure political institutions and to (indirectly) regulate social life.
One might at the outset object that this argument against the intrinsic value of democracy already leaves unattended a very large set of constraints which most democratic theorists tend to think are crucial for a healthy democracy: a well-educated public, the provision of constitutional rights within a federal system, and so on. But Arneson recognizes that many non-instrumentalists would say that democracy is only valuable “given mass literacy and the presence of other cultural background conditions.” So, everyone agrees that democracy is conditionally valuable: the instrumentalists avow straightaway that outcomes are the principal vehicles of value, while the non-instrumentalists (Arneson’s foes) tend to think that once certain background conditions are met, democratic procedures 1) have an independent, non-instrumental value and 2) must therefore be implemented whenever feasible, barring special circumstances.
Let’s dwell on (1) for a moment, because it’s quite ambiguous. The claim that democratic procedures have an independent, non-instrumental value is open to two interpretations. To see this, begin with a simple question: What, if anything, is the value of democratic procedures independent of? On one view, these procedures add something crucial (but not value or utility) to the decision making procedures of any social group. The something could be legitimacy (perhaps only democratic decisions are truly legitimate), or perhaps rightness (as Rousseau seemed to have thought, on an epistemic reading of his view). On this conception, democratic procedures are independent of the utility of the outcomes they enable because they do not directly enhance outcomes. According to this picture, it is better to think of democracy as enhancing a particular decision without making the decision itself necessarily better. I’ll say more about this shortly. But before moving on to the second interpretation, let me just add that this first interpretation is entirely consistent with Arneson’s helpful observation that standards of justice are independent of the standards that define the democratic ideal. Call this the indirect enhancement theory.
On another view, democratic procedures actually directly enhance, or at least should directly enhance, the quality of the outcomes. Call this the direct enhancement theory. They do so by encouraging discussion and debate about issues of public import, which in theory helps not only bring members of the citizenry to consensus more readily and frequently, but also helps reduce the influence of partisan interests and socially noxious opinions on public decision making. Madison gave an inkling of the shape such a view might take in the Federalist Papers. Political factions could not be prevented, but their unwelcome influence on politics could be attenuated by encouraging citizens to become members of all sorts of associations. Multiple associations meant multiple identities, each of which would cancel out one another in the public forum. Contemporary thinkers like Joshua Cohen have taken the direct enhancement theory in novel, interesting directions as well. Yet the important thing to realize is that on this second interpretation, democratic procedures actually modify and enhance the outcomes themselves.
For our purposes, it matters little which interpretation of democracy one prefers. On either of them, Arneson’s claim that democracy is not intrinsically just is problematic – though not because he is straightforwardly wrong. Consider the first view of (1), which says that democratic procedures are of independent worth in the sense that they convey, at least in principle, legitimacy or rightness or some other property to outcomes. Now apart from the sound observation that democracy does not always usher in the right outcomes, Arneson misunderstands the nature of the claim that democratic procedures have an independent worth or value. Clearly enough, procedures that were properly democratic could certainly confer to collectively made decisions a type of public legitimacy that decisions made furtively, behind closed doors, could not. Of course, we might not agree with Rousseau that rightness is generally conferred by such procedures; after all, the idea that a minority “learns” that its view is wrong when the majority has implemented its will is certainly foolish in a number of ways. However, the idea that a public finds democratically made decisions easier to swallow on balance is one that presumptively ought to weigh in favor of democratic procedures. “Easier to swallow” doesn’t mean that we take democratic decisions themselves to be superior; rather, it means that we are more apt to ascribe legitimacy to those decisions because they were arrived at by procedures that are generally effective (assuming the appropriate cultural background conditions). Notice here that an empirical judgment (i.e., that certain procedures are generally effective in the long run) is being coupled with a normative judgment (i.e., that something of value, namely legitimacy, results from the normal operation of those procedures). It is tempting to conclude that if we believe that as an empirical matter democratic procedures are generally effective in the long run we must a fortiori be committed to the idea that any value (legitimacy of outcomes) produced by those very procedures is also justified in a broadly utilitarian way. But our judgments about the value/legitimacy of the outcomes need not be parasitic on our assessments about the general reliability of the procedures themselves.
But Arneson seems to ignore this distinction, in effect denying that we ascribe legitimacy in this way. In all fairness, he might not be denying that this is how we as a descriptive matter treat democratic decisions. Indeed, Arneson might be denying that our predisposition to treat democratic decisions in this way matters when evaluating the normative desirability of democratic schemes. But this would be doubly odd, since if I understand his view correctly i) it is unhelpful to evaluate the normative desirability of democratic schemes per se; rather, on the broadly utilitarian view Arneson advocates normative desirability can only be measured for a particular scheme under specific empirical circumstances. And apart from his own view, ii) it would be truly illiberal to dismiss the indirect enhancement theory out of hand. Of course, Arneson could argue that ascribing a presumptive value to democratic procedures stacks the deck unfairly against his utilitarian approach. In some sense this is a reasonable complaint, but then again, couldn’t Cohen or Michaelman or any of the other deliberative democrats also reasonably complain that to deny presumptive value is tantamount to a tacit, and undefended, endorsement of political utilitarianism? Hard to see who should win this round, don’t you think?
So what about the second interpretation of the value of democratic procedures? On this second view decisions themselves can be enhanced by being democratically determined: since democratic procedures at their best involve a healthy amount of deliberation and discussion, rogue and irrational views alike are more likely to be screened off, leaving only relatively plausible political arrangements as candidates for informed, engaged voters. What are we to say about this sort of challenge to Arneson?
As I intimated a paragraph ago, I do not think that Arneson has a good rebuttal for those who believe that democratic procedures do – given certain background constraints – have utility-independent value in virtue of ascribing what we might call ex post legitimacy to decisions of all kinds, popular and unpopular alike. But I do think that this second interpretation of democracy’s value is much more vulnerable than the first. For one, the picture it paints of political participation is optimistic enough to border on being quixotic. But the real worry I have about this latter interpretation is one raised, if somewhat obliquely, by Arneson himself. At one point he argues: “Rights to power over the lives of others always involve an element of stewardship. If one has such a moral right, this will be so only because one’s having the right is more conducive to the flourishing of all affected parties than any feasible alternative” (p. 10). Put another way, political power cannot be justifiably distributed unless it actually does contribute to human flourishing. Who could reasonably reject this sensible position?
I don’t think anyone would. Many democratic theorists do in fact accept this view, or something very much like it. Such folks do not claim that democratic procedures are intrinsically just because we (on some more granular theory) have a right to them, given our psychology, cognitive capacities, or whatever. Such procedures are, at least to Mill and (probably) Cohen, deeply valuable to the agents who are able to participate in them because it is through these procedures that individuals develop their ability to make choices and deliberate with others on matters of common concern. It is surely in our interest to have democratic privileges, in other words. Admitting that this is so wouldn’t confer a right upon the users of these procedures. In fact, it is somewhat ironic that Mill’s quasi-egalitarian and utilitarian argument for participation in the workplace is not rights-based at all: Mill argues that it would be in everyone’s interest, worker and manager alike, if workplace decision-making were more democratic. On the Millian view, participation and deliberation are (to risk hyperbole) ‘intrinsically’ valuable, which is quite different from claiming that we do have a right to “have a say” in politics. (But even this is somewhat misleading, because Mill thought, if David Lyons of Boston University is on the right track, that political privileges and rights, including that of participation, would have to in the end be justified on utilitarian grounds.)
And here’s where the rubber hits the road: On the best available reading of Mill, having a democratic say is not simply a matter of producing utility nor is it simply a matter of providing utility-independent conditions (democratic privileges) in which human beings can flourish. Both aims – producing utility and securing democratic privileges for all citizens – need to be achieved at the same time to realize either of them. But if these aims can only be achieved together, and if utilitarianism is the challenging idea that the right is defined by the good (utility), then would it really be too far off the mark to conclude that we have a right to the conditions necessary to foster the good itself?
The answer, I think, is no. But I do not share Arneson’s rationale for holding this position. Arneson believes that if we assume that power over others has to be directly justified by utilitarian considerations (human flourishing), then it follows that there really isn’t any such thing as a right to a democratic say. Rights are, by definition, impervious to the utilitarian calculus, in effect immune to the contingencies of history and the accidents of political life. (Arneson, incidentally, also attacks arguments from publicity (only democratic decisions are sufficiently public to respect the demands of justice) and from neutrality (only democratic procedures respect the rational capacities of agents) in the course of contending that there is no ‘noninstrumental moral right to a democratic say.’)
But even if Arneson is correct to remind us that rights are by definition precluded by his utilitarian approach, it is not at all clear that the debate really concerns rights at all. Let me make this point by way of example. In his Political Economy, Mill expresses moral outrage at the nondemocratic workplace arrangements that is simply hard to capture if the only tools one can draw upon derive from rights talk. Mill certainly thought that while it was fortuitous for his theory that extending democracy to the workplace would be congruent with a commitment to utilitarianism, the fact that participation would provide opportunities for individuals to develop and express their capacities for choice in public fora was independently valuable. Put another way: While a Millian is arguably forced to argue in favor of a specific “right” or a specific “privilege” on the basis of utilitarian considerations alone, she need not hold that the value of that right or privilege is reducible to, or exclusively derived from, its conduciveness to utility. The justification and value of a privilege are two different things. And this is precisely why one can accept the idea that we do not have a right to the conditions necessary to foster the good itself (democratic privileges) without accepting the controversial notion that democratic privileges have no teeth independent of their ability to enhance human well-being.
So what follows from this? If certain forms of democratic participation do indeed lead to the social solidarity and moral development of persons that Mill envisioned, something very interesting follows: On a direct or indirect enhancement theory the tendencies of democracy to produce both a) utility and b) solidarity/moral development must be taken into account when evaluating – regardless of the normative perspective one has adopted – the desirability of the democratic scheme under consideration. In some cases, utility will be the most important factor, trumping the nondemocratic alternatives straightaway. But in other cases, we may want to defer to democratic procedures (even if they will probably not produce optimal results) precisely because legitimacy is so crucial. One need not trump up arcane examples of this principle – think for just a moment about the importance of empowering the Iraqi people to play the primary role in designing and implementing democratic institutions which they can honestly call their own.
Arneson could, of course, retort that we think it’s important to let the Iraqis develop their own flavor of democracy because this, as the old saw goes, will produce more utility in the long run. All I can say to this – and I certainly should say more – is that such a response unjustifiably treats all reasons, all considerations which stir our passions and motivate us to action, as fundamentally of the same coin. This makes a caricature of deliberation about matters moral. The character of moral reasoning is not, at least from the first person, consistent with the claim that all value can be regarded as an input to a utilitarian calculation. Many philosophers have made this case before with much more verve and acuity, so I won’t rehearse their arguments at length. Suffice it to say that we value democratic procedures not for what they do for us, but rather for what they say about us.
It is true that we want to see ourselves in a certain way, as free and equal citizens able to help make decisions that will affect the lives of all of us. Perhaps this desire is itself problematic. Maybe it is in our best interests to defer to the opinions of experts far more often than we are currently disposed. Yet if Arneson does mean to say that democratic procedures do not have any special status, or, to say the same thing differently, that democratic privileges have no value other than the utility they produce (where utility is an extremely expansive, nebulous notion), then he has misled us. Of course, I would admit that it is possible that both the indirect and direct enhancement theories might not (independently or jointly) justify the claim that democracy is independently valuable. But to make either argument – either that legitimacy isn’t a by-product of democracy (pace indirect enhancement theory) or that decisions aren’t enhanced through being the result of democratic deliberation (pace direct enhancement theory) – much more needs to be said.
posted by Lawrence Solum 7:24 AM
Lynn on Dawkins & Functionalist Explanation
T.J. Lynn writes in response to my trio of posts ( Natural Goodness: From Facts to Values, Metaethical Prejudice: More Remarks on Ethical Naturalism & Naturalistic Ethics):
I think you're wrong--at least in part-- about Matt Evan's post on
First, you say "Naturalists, like Dawkins, believe that the sex drive
does have a purpose, and that purpose or function is reproduction.
Indeed, Dawkins' most famous work, The Selfish Gene is an explication of
the nature of that purpose--which Dawkins believes is best understood as
explained by individual genes, as opposed to whole genomes, organisms,
or species. Evolutionary biologists rely on functionalist explanations
as their bread and butter: they couldn't do without them."
I don't think that's right. Here's what Dawkins says in his book (at
least, this is an internet quote of his book--which I acknowledge is not
nearly the same thing): Now, natural selection favours replicators that
are good at building survival machines, genes that are skilled in the
art of controlling embryonic development. In this, the replicators are
no more conscious or purposeful than they ever were. The same old
processes of automatic selection between rival molecules by reason of
their longevity, fecundity, and copying-fidelity, still go on as blindly
and as inevitably as they did in the far-off days. Genes have no
foresight. They do not plan ahead. Genes just are, some genes more so
than others, and that is all there is to it. (p. 24)
This passage suggests that Matt has it right. There is no purpose to
the project, only automatic processes with inevitable outcomes.
Second, and more importantly, I think you're a bit hard on Matt and his
metaethical views. I do agree that he both overstates and misstates his
case. But it seems to me that he gets a fair amount right. For
example, it isn't uncommon for those who oppose naturalism to
concentrate chiefly on opposing ethical noncognitivists. Charles
Larmore, to choose but one handy example (his Morals of Modernity is on
my desk) chooses to attack philosophical naturalism by attacking the
'anticognitivists.' (He does, to be sure, criticize what he calls
cognitive naturalism, but it isn't the focus.) There is a rather
casual--but it seems to me generally accurate--association of naturalism
and noncognitivism. And that association is common even--or perhaps
especially--among naturalists. Matt isn't wrong to make that
association (even if he thinks you're the one making the association!).
And his point against that position is quite a bit stronger.
I do agree with your more general point on the unnecessary belief in a
right/left split on the issue.
posted by Lawrence Solum 10:16 AM
Technical Work on Functionalist Explanation and Naturalism in Ethics
In response to my posts on Metaethical Prejudice: More Remarks on Ethical Naturalism & Naturalistic Ethics, the author of the Technical Work blog writes:
I have to rise to a point you made today about naturalism in ethics. You said:
"Naturalists, like Dawkins, believe that the sex drive does have a purpose,
and that purpose or function is reproduction. Indeed, Dawkins' most famous
work, The Selfish Gene is an explication of the nature of that
purpose--which Dawkins believes is best understood as explained by
individual genes, as opposed to whole genomes, organisms, or species.
Evolutionary biologists rely on functionalist explanations as their bread
and butter: they couldn't do without them."
Many evolutionary biologists certainly rely on functional explanations
(nitpick: "functional", not functionalist; "functionalism" means something
different to social scientists), but I don't know of any biologists who hold
that function involves purpose. The whole topic of function is at the center
of a rather complex series of debates right now. I've listed several
overlapping collections of papers and a few articles below. A lot of the
definitional discussion is aimed at getting rid of the idea of purpose--
even when they put the word in their titles.
In short, purpose may be necessary to a naturalist ethics a la Foot, but
it's not needed (and many see it as confusing and/or irrelevant) to
understanding the ways organisms and species work. So we're back to the same
old problem of how to get from is the "is" to the "ought". Postulating
purpose in nature (with or without the assistance of an intelligent
designer) is, I think, a petitio principii.
What amazes me in philosophical discussions of the subject (e.g., the one in
progress on Crooked Timber right now) is the way in which all of social
organization-- families, courts, police, jails, churches, gossip networks,
etc et etc-- just get blotted out of the discussion, as if they had nothing
to do with defining, transmitting, transforming or enforcing ethics. It's an
extraordinary sort of blindness, much like the philosophy of science went
through between WWII and the late 1970s. Perhaps ethics will catch up soon.
Meanwhile, thanks for your blog-- I enjoy it immensely.
Here are some references for function in biology:
Allen, C., M. Bekoff and G. V. Lauder. (Eds.) 1998. Nature's Purposes:
Analyses of Function and Design in Biology. Cambridge, MA: MIT Press.
Amundson, R. and George V. Lauder. 1994. ³Function without purpose: The uses
of causal role function in evolutionary biology.² Biology & Philosophy 9:
Buller, D. J. (Ed.) 1999. Function, Selection, and Design. Albany,NY: SUNY
Davies, P. S. 2001. Norms of Nature: Naturalism and the Nature of Functions.
Cambridge, MA: MIT Press.
Wimsatt, William C. 1972. ³Teleology and the logical structure of function
statements.² Studies in History & Philosophy of Science 3: 180.
posted by Lawrence Solum 10:03 AM
Bass on Naturalistic Ethics
In response to my posts on Metaethical Prejudice: More Remarks on Ethical Naturalism & Naturalistic Ethics, Matthew Bass writes that my post made him
think about a Discovery Channel program on the mating habits of sea lions that I saw last month.
In a sentence of two, sea lion mating is rough stuff. First, the males fight, and the winners win the opportunity to rape the females. This, assumedly, is good for the species because only the strongest genes get passed on. I can understand, therefore, at least the frame of the argument that one can conclude that this is how sea lions "ought" to behave (in a naturalistically ethical kind of way.)
But what struck me was not this general rule but the solitary exception. Down the beach from the ruckus of the mating season, were two solitary monogamous sea lions. As the show explained it, these sea lions were not part of any minority sea lion mating pattern -- they were in fact the only two monogamous sea lions ever encountered. It was the same two, and the met and mated year after year. This was an advantage for the male, who didn't have to fight his way through, and for the female, who was not attacked from the strongest male. It seemed likely, however, that their offspring would not fare ver well genetically when their turn to mate came around. The fact that this mating pattern was not seen more often gave further evidence that it was an evolutionary dead end. Essentially, the sea lions "ought not" to be mating this way.
Where a layman such as myself loses the concept of naturalistic ethics is that this line leads to the conclusion that the monogamous sea lions are acting unethically, or immorally because they are doing what the naturalists tell us they "ought not" do.
Naturalistic ethics -- I must conclude -- may exist, but they must necessarily differ from "real" human ethics.
posted by Lawrence Solum 8:49 AM