The Mechanisms of the Slippery Slope

Eugene Volokh*

(forthcoming 116 Harv. L. Rev. ___ (2003); Sept. 16, 2002 draft)

I. Introduction............................................................................................................................................................... 2

II. Cost-Lowering Slippery Slopes and Other Multi-Peaked Preferences Slippery Slopes................................. 11

A. Cost-lowering slippery slopes....................................................................................................................... 11

B. Cost-lowering slippery slopes as multi-peaked preferences slippery slopes........................................ 18

C. More multi-peaked preferences: “Enforcement need” slippery slopes.................................................. 20

D. Equality slippery slopes and administration cost slippery slopes.......................................................... 24

E. Multi-peaked preferences and unconstitutional intermediate positions................................................ 37

F. The hidden slippery slope risk and unexpected outcomes exposing multi-peaked preferences........ 38

G. The hidden slippery slope risk and the ad hominem heuristic................................................................. 40

III. Attitude-Altering Slippery Slopes.......................................................................................................................... 41

A. Legislative-legislative and judicial-legislative attitude-altering slippery slopes: The is-ought heuristic, and the normative power of the actual.......................................................................................................................................................... 41

B. Legislative-judicial attitude-altering slippery slopes: “Legislative establishment of policy”............ 45

C. Just what will people infer from past decisions?........................................................................................ 50

D. Judicial-judicial attitude-altering slippery slopes and the extension of precedent............................... 59

E. The attitude-altering slippery slope and extremeness aversion behavioral effects.............................. 61

F. The erroneous evaluation slippery slope.................................................................................................... 62

G. Are attitude-altering slippery slopes good or bad?................................................................................... 64

IV. Small Change Tolerance Slippery Slopes.............................................................................................................. 65

A. Small change apathy, small change deference, and rational apathy....................................................... 66

B. Small change tolerance and the desire to avoid seeming extremist or petty.......................................... 69

C. Judicial-judicial small change tolerance slippery slopes and the extension of precedent................... 71

V. Political Power Slippery Slopes............................................................................................................................... 73

A. Examples........................................................................................................................................................... 73

B. Types of political power slippery slopes..................................................................................................... 76

VI. Political Momentum Slippery Slopes..................................................................................................................... 78

A. Political momentum and effects on legislators, contributors, activists, and voters.............................. 79

B. Reacting to the possibility of slippage—the slippery slope inefficiency and the ad hominem heuristic 83

VII. Implications and Avenues for Future Research................................................................................................... 84

A. Considering Slippery Slope Mechanisms in Decisionmaking and Argument Design.......................... 85

B. Thinking About the Role of Ideological Advocacy Groups.................................................................... 86

C. Fighting the Slippery Slope Inefficiency..................................................................................................... 88

D. Slippery Slopes and Precedent...................................................................................................................... 89

E. Empirical Research: Econometric, Historical, and Psychological............................................................. 89

F. When (If Ever) Should We Avoid Slippery Slope Reasoning?............................................................... 90

VIII. Conclusion................................................................................................................................................................. 91

“In other countries [than the American colonies], the people . . . judge of an ill principle in government only by an actual grievance; here they anticipate the evil, and judge of the pressure of the grievance by the badness of the principle. They augur misgovernment at a distance and snuff the approach of tyranny in every tainted breeze.”

— Edmund Burke, On Moving His Resolutions for Conciliation with the Colonies, Speech to Parliament, Mar. 22, 1775.

I. Introduction

You are a legislator, a voter, a judge, a commentator, or an advocacy group leader. You need to decide whether to endorse decision A, for instance a partial-birth abortion ban, a limited school choice program, or gun registration.

You think A, on its own, might be a fairly good idea, or at least not a very bad one. But you’re afraid that A might eventually lead other legislators, voters, or judges to support B, which you strongly oppose—for instance, broader abortion restrictions, an across-the-board school choice program, or a total gun ban.

What does it make sense for you to do, given your opposition to B, and given your awareness that others in society might not share your views? Should you heed James Madison’s admonition that “it is proper to take alarm at the first experiment on our liberties,”[1] and firmly oppose something that you might have otherwise supported were it not for your concern about the slippery slope? Or should you accept the immediate benefits of A, and trust that even after A is enacted, B will be avoided?

Slippery slopes are, I will argue, a real cause for concern, as legal thinkers such as Madison, Jackson, Brennan, Harlan, and Black have recognized.[2] And these arguments comport at least partly with our own experience: We can all identify situations where a first step A has led to a later step B that might not have happened without A, though we may disagree about exactly which situations exhibit this quality.[3] A may not logically require B—but for political and psychological reasons, it can help bring B about.[4]

But, as legal thinkers such as Lincoln, Holmes, and Frankfurter have recognized, slippery slope arguments are of limited utility.[5] We accept, because we must, some speech restrictions. We accept some searches and seizures. We accept police departments, though creating such a department may lead to arming it, which may lead to some officers being willing to shoot innocent civilians, which may eventually lead to a police state (and all this has happened with the police in some places). Yes, each first step involves risk, but it is a risk that we need to run.

This need makes many people impatient with slippery slope arguments.[6] The slippery slope argument, the flip response goes, is the claim that “we ought not make a sound decision today, for fear of having to draw a sound distinction tomorrow.”[7] To critics of slippery slope arguments, the arguments themselves sound like a slippery slope: If you accepted this slippery slope argument, then you’d end up accepting the next one and then the next one until you eventually slip down the slope to rejecting all government power (or all change from the status quo), and thus “break down every useful institution of man.”[8] Exactly why, they ask, would accepting, say, a restriction on “ideas we hate” “sooner or later” lead to restrictions on “ideas we cherish”?[9] If the legal system is willing to protect the ideas we cherish today, why wouldn’t it still protect them tomorrow, even if we ban some other ideas in the meanwhile? And of course, even if one thinks slippery slopes are possible, what about cases where the slope seems slippery both ways—where both alternative decisions seem capable of leading to bad consequences in the future?[10]

My aim here is to analyze how we can sensibly evaluate the risk of slippery slopes, a topic that has been surprisingly underinvestigated.[11] I think the most useful definition of slippery slopes is a broad one, which covers all situations where decision A, which you might find appealing, ends up materially increasing the probability that others will bring about decision B, which you oppose.[12] If you are faced with the pragmatic question “Does it make sense for me to support A, given that it might lead others to support B?,” it shouldn’t much matter to you whether A would lead to B through logical mechanisms or psychological ones, through judicial ones or legislative ones, or through a sequence of short steps or one sharp change. Nor should it matter to you whether or not A and B are on a continuum where B is in some sense more of A, a condition that would in any event be hard to define precisely.[13]

The question is whether A might lead to some harmful decisions in the future, through whatever mechanisms. To answer this question, we need to think—without any artificial limitations—about the entire range of possible ways that A can change the conditions (whether those conditions are public attitudes, political alignments, costs and benefits, or what have you) under which others will consider B.

The slippery slope is a familiar label for many of the most common examples of this phenomenon: When someone says “I oppose partial-birth abortion bans because they might lead to broader abortion restrictions,” or “I oppose gun registration because it might lead to gun prohibition,” the common reaction is “That’s a slippery slope argument.” But whatever one calls these arguments, the important point is that the observer is asking the question “Does it make sense for me to support A, given that it might lead others to support B?,” which breaks down into “How much do I like A?,” “How much do I dislike B?,” and, the focus of this article, “How likely is A to lead others to support B?”[14] And this last question in turn requires us to ask “What are the mechanisms through which A can lead others to support B?”

It is these real-world mechanisms on which I will focus.[15] Slippery slopes, camel’s noses, thin ends of wedges, floodgates, and acorns are metaphors, not analytical tools. My goal is to describe the real-world paths that the metaphors represent—to provide a framework for analyzing and evaluating slippery slope risks by focusing on the concrete means through which A might possibly cause B.

Specifically, I want to make the following claims, which are closely related but which are worth highlighting separately:

1. Though the metaphor of the slippery slope suggests that there’s one fundamental mechanism through which the slippage happens, there are actually many different ways that decision A can make decision B more likely. Many of these ways have little to do with the mechanisms that people often think of when they hear the phrase “slippery slope”: development by analogy, by decision A changing people’s moral or empirical assumptions about B, or by people becoming “desensitized” to decision B.[16]

To illustrate this briefly, consider the claim that gun registration (A) might lead to gun confiscation (B).[17] Setting aside whether we think this slippery slope is likely—and whether it might actually be desirable—it turns out that the slope might happen through many different mechanisms, or combinations of mechanisms:

• Registration may change people’s attitudes about the propriety of confiscation, by making them view gun possession not as a right but as a privilege that the government grants and therefore may deny.

• Registration may be seen as a small enough change that people will reasonably ignore it (“I’m too busy to worry about little things like this”), but when aggregated with a sequence of other small changes, registration can ultimately lead to confiscation, or something close to it.

• The enactment of registration requirements can create political momentum in favor of gun control supporters, thus making it easier for them to persuade legislators to enact confiscation.

• Non-gun-owners are more likely than gun owners to support confiscation.[18] If registration is onerous enough, over time it may discourage some people from buying guns, thus diminish the fraction of the public that owns guns, diminish the political power of the gun-owner voting bloc, and increase the likelihood that confiscation will be politically feasible.

• Registration may lower the cost of confiscation—since the government would know which people’s houses to search if the residents don’t turn in their guns voluntarily—and thus make confiscation more appealing.

• Registration may trigger the operation of other legal rules that make confiscation easier and thus more cost-effective: When guns aren’t registered, confiscation would be largely unenforceable, since house-to-house searches to find guns would violate the Fourth Amendment; but if guns are registered some years before confiscation is enacted, the registration database might provide probable cause to search the houses of all registered gun owners.[19]

I think that in the registration-to-confiscation scenario, only the latter two mechanisms are fairly plausible; in other scenarios, others may be more plausible. But the important point is that being aware of all these mechanisms can help us as citizens and policymakers think through all the possible implications of some decision A—and can help us as advocates make more concrete and effective arguments for why A would (or would not) lead to B. And even if you are skeptical of one kind of slippery slope claim, you may find that the others are worth considering.

2. As the above example illustrates, slippery slopes are not limited to judicial-judicial ones, where one judicial decision leads to another through the force of traditional judicial precedent. They can also be legislative-legislative, where one legislative decision leads to another (Madison’s concern in his famous Remonstrance Against Religious Assessments[20]), judicial-legislative, and legislative-judicial.

3. Slippery slopes may occur even when a principled distinction could be drawn between decisions A and B. The question isn’t “Can we draw the line between A and B?,”