There are a couple of social issues in particular in which his arguments are especially strong and well supported by evidence. Welfare policies and affirmative action are two such examples, but I'm going to talk about a third one here: the criminal justice system.
My goal is to try to flesh out Ben's arguments somewhat and, since most of my readers are thoughtful liberals and leftists, see how you respond to them, because I really can't find compelling objections. (I apologize, Ben, if I distort your arguments. Feel free to correct me as you see fit.)
One of the key insights of his conservatism is his emphasis on applying psychological research on human behavior to thinking about social issues. This more behavioristic approach is anathema to many liberals--it can be very paternalistic--but even if one accepts Mill's harm principle, it is still compelling (for, we do not exist in isolation; few actions harm the self alone).
The basic idea, as I see it, is that a society should reward good behaviors and punish bad ones. Now, this can be controversial, so let's just apply it to those examples in which almost everyone would agree. For instance, we recognize that violent crimes (murder, assault, rape, etc.) should be punished. Even if there are root causes to violence which we should be addressing, this is no argument against setting up a system that discourages violent behavior.
A good psychologist knows that punishments are most effective if they are swift, certain, and severe. In our current system, this is often not the case with punishments. It's easy, particularly for first-time offenders, to get off with a light sentence or no sentence at all. It would be in our interest (and theirs!), however, to discourage first-time offenders from getting into a habit of committing crimes. If they can get away with it, that serves as an incentive to repeat criminal activities.
If we are interested in deterrence and reform, then the research would support this conclusion about the nature of effective punishment. Swift, certain, severe punishments are more effective deterrents to be sure. Insofar as they deter future acts by appropriately punished criminals, they can also have a reforming effect.
Now, learned helplessness is a possibility with excessively cruel punishments like torture--and there may be negative side effects to any punishment--but these are matters that can be decided empirically. (Since Ben knows the research far better than I, I trust his judgment about the results of experiments, if not always about their normative implications.)
At this point I want to raise a few objections, both theoretical and practical: 1) Our criminal justice system should be... well, just. Punishments should fit the crime and innocent people should not be punished. 2) The tendency to commit crimes is the result of a combination of genetics and developmental conditions, such so that it is unjust to hold people responsible for happening to be born into a bad situation. 3) There are many crimes in which the law is draconian, for instance drug-related offenses. In other words, practically speaking we find numerous violations of the first concern. 4) Racism and other biases are deeply entrenched and ineliminable in our system, so that harsher punishment is, in effect, a racist policy.
In the case of 1, I think there is a serious concern. Our criminal justice system operates with a principle of assumed innocence in which the burden is on the state to prove guilt beyond a reasonable doubt. Conservatives note that these safeguards sometimes go too far; in any case, a compelling case can be made that we err too much on one side. More innocent people are let free but so are more guilty ones. It is an empirical question whether this generates positive social utility, but the current evidence seems to be in support of the conservative stance.
This starts to work in the direction of the second objection, to which one might pose the following response. Even if we grant that individuals lack free will and, thus, responsibility for their actions, this does not mean that we shouldn't treat them as if they were in control of their actions. It can be better both for the individual and for the larger community to assign blame and punish wrongful acts.
For the individual, because a stronger punishment is more likely to change behavior (if one jail sentence is so horrific, it will discourage perps from doing things which would send them back to the slammer). For society, because it is able to isolate and contain behaviors that threaten order and stability. (Again, of course, these arguments rely on certain empirically observable conditions.)
But isn't there a tension here between responsibility and policies based on predicting behaviors? Perhaps, but there is a way to look at the role of the state in a way that is more compatible with individual autonomy. Certain actions that individuals might choose will be made to come at a higher cost, so that people will be discouraged from doing those things.
Third, we come to a practical objection. In some cases, there seems to be no significant, demonstrable harms involved; in others, no viable options for individuals except for crime. The two "long wars," on drugs and terror, are excellent examples of government going too far to restrict individual rights.
Yes, it may be the case that some of our laws are unjust, but this is more an argument against such laws rather than an argument against penalizing people who break the law. The principle that the punishment should fit the crime works both ways: in too many cases individuals who have acted in horrific ways have not been adequately penalized.
That fair punishment is difficult to implement does not constitute an argument against its value. Moreover, it does not necessarily limit the ability to challenge unjust laws. Even the act of going to prison can provides a means of protesting injustice (consider MLK's "Letter from a Birmingham Jail").
Finally, we reach the issue of racism. This is an underlying theme in all the social issues I mentioned above. An amazing thing about many otherwise intelligent liberals is that as soon as race (or gender or sexuality) enters into the picture, all self-reflective criticism stops. Their implicit argument runs like this:
1. Anything that is racist (sexist, heterosexist, etc.) is beyond the pale.This kind of argument is also used to discredit research on IQ and other controversial measures; any study which shows statistically significant group differences is automatically labeled racist. In academia, to be labeled a bigot is a death knell; even if you're tenured, no one will listen to you anymore.
2. Any policy (empirical result, standardized test, etc.) which puts blacks (other racial minorities, women, gays and lesbians, etc.) at a disadvantage is racist (etc.).
3. Thus, etc.
When you try to discuss issues like this in a public forum you get: "Oh, didn't you know that Stephen J. Gould has definitively shown that race is not even a biological category?" "Didn't you get the memo that all gender differences are socially constructed?" "Haven't you heard that there's a gay gene and sexual orientation is not a choice?" I find this kind of thoughtlessness infinitely annoying. In some cases it may happen to be correct, but let's not lie about what the evidence is, even if it hurts our case.
Thus, even if we grant that the criminal justice system is systematically racist, this objection is more an argument against current judicial practice rather than against conservative legal principles. Instead of letting blacks off on crimes for which whites are acquitted, we might just as consistently convict more white lawbreakers.
I don't know that I find all of these responses adequate, but generally speaking this mode of argumentation is one that liberals too easily dismiss, or don't even consider in the first place.
Even in a non-paternalistic state it just makes sense to discourage uncontroversially bad behavior and to encourage people to develop themselves as much as possible. This is certainly not incompatible with social reforms targeting poverty and discrimination or with human rights. Let's keep in mind that people who break the social contract thereby justly lose some of their freedoms.
So, is this case at all compelling? What do y'all think?