Here’s a simple model of the slippery slope. You have to adopt a position on an issue and defend your position to yourself and your critics. The spectrum of positions ranges from the left-most extreme to the right-most extreme and you have to decide whether to take one of these extreme positions or some moderate point in the interior.
Defending a moderate position is a delicate balancing act. It’s a very special set of utility functions which attain their maximum right at that point, and you need to convince your critics that the right utility function happens to be one of those. Any slight perturbation of a utility function in that set will push you to the left or right so your critics have an easy task. And once you’ve lost the first battle your credibility is damaged.
The easiest positions to defend are the extreme ones. At an extreme position you have a binding constraint. To defend your extremist position it is enough to say that you are such an extremist that you would like to move even farther to the right if that were possible. The set of utility functions that have an optimum somewhere to the right of the right boundary is a large set. You can perturb such a utility function and the extremist position will still be optimal.
The same logic explains why a few special interior positions can be robust to the slippery slope. Think of a kinked budget constraint. A large set of utility functions achieve their optimum at a kink.
10 comments
Comments feed for this article
March 27, 2012 at 11:11 pm
twicker
Curiously, for this argument, middle positions actually tend to be the most robust to attacks. By, “robust to attacks,” I mean that people who are moderates tend to remain moderates – they may move from moderate liberal to moderate conservative, depending on the issue, but they don’t readily go to extremes. Moderate positions tend to consist of a series of different middle grounds, none of which are entirely dependent on the others for consistency (e.g., abortions should be legal and rare). Thus, even if you can persuade a person that an extreme position would really be better (e.g., don’t worry about “rare,” or that “rare” really would be best achieved by outlawing abortions), the arguments that led to the moderate position functionally “absorb” the attack, and the moderate position stands (“Ok, so we add more constraints – but the legality must remain in order to prevent back-alley abortions, which would be worse”).
Extreme positions, however, tend to be extremely consistent – so consistent that an attack on one chink in the armor can bring the entire edifice crashing down. Continuing with the abortion example, making abortions illegal in 100% of the cases immediately faces strong attacks from people who would rather keep it legal in cases of rape, incest, and murder. A truly extreme example was seen in Mississippi’s whopping defeat of the personhood amendment, which crashed the moment people discussed the possibility, not even the surety, that it would cause problems for birth control, fertility treatments, etc. On the opposite side, if you go to the extreme of the pro-choice side and remove all restrictions, it becomes ridiculously easy to come up with cases where there *should* be restrictions, given the severe psychological impact it often has.
One of the more interesting findings in areas outside economics is that people on extremes relatively often swing from one extreme to another, given that, once you can question one aspect of the bedrock faith, the “bedrock” looks more like quicksand. In an attempt to replicate that absolute black-and-white faith, these people, who often receive comfort from that ease of categorization, go scampering to some other extreme (not necessarily the “opposite pole;” there are plenty of poles around, including the various forms of anarchy, authoritarian movements on right and left, etc.). Moderates, on the other hand, never had that kind of black-and-white view, so having to add more nuance comes fairly easily and is accepted, no faith shaken.
For a physical metaphor, it’s reminiscent of Civil War fortifications: the masonry forts that had been popular got literally pulverized (they were strong – but very brittle and, thus, relatively easy to breach by direct application of cannonshot in one location), while the earthen forts absorbed the cannonshot and, while a bit worse for the wear, survived far harsher attacks.
March 28, 2012 at 12:12 am
Lones Smith
I always use the slippery slope as a metaphor when I teach supermodular games. Typically this epithet is tossed out at someone who allows some of an undesirable activity. People understand that once some is allowed, that if the best response function is increasing — i.e. the touchstone of supermodular games — then then next period sees some more, ad infinitum.
March 28, 2012 at 11:19 am
Assorted links — Marginal Revolution
[…] 2. The slippery slope. […]
March 28, 2012 at 12:24 pm
Orthodox
The best slippery slope arguments are those where “bright red lines” are involved. Once the initial breaking of the taboo/custom/law is made, then there’s nothing to stop society from moving farther and farther.
Looking at the 20th Century, it’s clear that the moderates and extremists kept shifting along the slippery slope and many of the slippery slope arguments proved incredibly accurate, especially relating to the sexual revolution and government power. There’s no way that gay marriage won’t lead to legalized polygamy, for example, unless you believe that a 50-year trend will suddenly stop dead in its tracks, while federal laws are literally a non-stop slippery slope, take your pick (such as the income tax amendment will only be used to tax the rich!). There are many suckers (moderates) who buy into the “that will never happen” argument, which is why it pays to be as extreme as possible, since moderates always look for the middle of two extremes and do not come up with original ideas.
March 28, 2012 at 1:30 pm
tlwest
Orthodox, I think what you think a weakness is considered a strength by many moderates. Certainly polygamy seems unconscionable to me today. But 30 years from now, perhaps society will believe that it’s acceptable. I am not arrogant enough to believe that I am the ultimate arbiter of good/evil for all society for all time. If society wishes to override the my current wishes in 20 years, then it will *and it should*.
I suspect most moderates don’t believe society should be held in bondage by its laws, forcing it into shapes that that no longer represent the will of the people. If the will of the people (with obviously some +/- on either side) has changed in 20 years, I think most moderates feel it is the responsibility of government to change the laws to accommodate that will.
Obviously sometimes the law will lead and sometimes it will trail general social shifts.
March 28, 2012 at 1:15 pm
James Hansen
I have two tiny practical observations to make. Within organizations like companies or newspaper offices, it appears to me that there is a strong tendency for moderation to be related to actual responsibility – the inverse being that those with little or no responsibility tend to hold more radical opinions in one way or the other. Of course, organizations may well select for moderation in identifying their hierarchies, so it is to difficult to assert a casual relationship. The second and somewhat related point is that people who are paid – or at least professionally encouraged – to express a particular point of view, whatever it is, very often come to hold it quite sincerely. It is – almost – amusing to watch corporate press officers for instance slowly evolve into true believers. The same appears to hold for most forms of professional advocacy.
March 28, 2012 at 5:39 pm
mobile
If there is uncertainty about what is being optimized, as there will be for any deliberator with enough humility, an extreme solution has a risk of being a very bad solution if the realized welfare function is much different from what you thought it would be. A moderate solution might not be close to optimum for most of the possible welfare functions, but it could be more robust to errors in the objective function coefficients.
In spaces where people will stake out quite different extreme positions, that is evidence for uncertainty about what is to be optimized.
March 28, 2012 at 6:11 pm
Alan Keffer
Empty talk about nothing. To the aphorism “those who can do, those who can’t teach” should be added “those who haven’t a clue talk.”
March 28, 2012 at 9:09 pm
jeff
I love you Alan Keffer
March 30, 2012 at 9:17 am
Mario Rizzo
What is a moderate position and what is an extreme position is not some exogenous “fact.” It is determined, in part, by the participants in the debate. For example, the purveyors of “libertarian paternalism” have tried to sell their ideas as some sort of mean between the extremes. I have thought long and hard about these issues. The interested reader should look at my articles in the UCLA Law Review (“The Camel’s Nose is in the Tent”) and the Arizona Law Review (“Little Brother is Watching You”). The latter is specifically about new paternalism.
Click to access Rizzo.pdf
http://works.bepress.com/mario_rizzo/30/
I also recommend the article by Eugene Volokh in the Harvard Law Review, “The Mechanisms of the Slippery Slope.”
Click to access slipperyshorter.pdf