"Which Step, Specifically?"
MinutesWhen told that one action will lead through a chain of events to a dramatic outcome, lay the chain out link by link, ask what makes each link likely, and remember that the probability of the whole chain is lower than that of its weakest link.
How to do it
- 1Get the chain stated explicitly: "So A leads to B, B to C, and C to D. Is that the argument?" Much of the force of a slope argument comes from leaving the middle unstated.
- 2For each link, ask for the mechanism. What causes the move from B to C: a legal precedent, a changed incentive, a shift in attitudes, a lowered cost? "It just will" is not a mechanism.
- 3Put a rough probability on each link and multiply. Four steps that are each 70 percent likely give an overall chance of about 24 percent. A vivid, detailed chain feels more likely than a bare prediction and is in fact less likely.
- 4Look for stopping points: places on the path where a separate decision would have to be made, and where people who supported step one could and would refuse step two.
- 5Ask for a track record. Where this first step has been taken elsewhere, did the later steps follow?
- 6Turn it around on optimistic chains as well. Business plans, investment pitches, and "this one change will fix everything" arguments have the same structure with a pleasant ending.
What to say
- “Walk me through it. What specifically gets us from the second step to the third?”
- “Which of those steps do you think is most likely not to happen?”
- “Has that sequence actually played out anywhere this has been tried?”
When to use it
- •A proposal is opposed mainly on the ground of where it will supposedly lead.
- •A pitch promises a large result through a series of "and then" steps.
- •A frightening scenario is persuasive because it is detailed.
- •A causal story jumps from a first event to a distant outcome without the middle.
Counters
Evidence and how strong it is
The probabilistic point is the conjunction rule: a chain of events cannot be more probable than any one of its links, yet people reliably judge detailed scenarios as more likely than simpler ones because each added detail makes the story more representative and easier to picture (Tversky & Kahneman 1983). Walton's (1992) study of slippery-slope arguments concludes that they are not inherently fallacious; they are arguments from consequences whose strength depends on how well each step is supported, and the appropriate response is exactly to question the individual links. Volokh (2003) makes the complementary point from law: slopes are sometimes real, and they operate through identifiable mechanisms such as precedent, changes in attitudes, lowered costs of the next step, and political momentum, so asking for the mechanism separates credible warnings from rhetorical ones. No experiments test this questioning strategy as a persuasion defense, though unpacking scenarios into components is a standard debiasing recommendation in forecasting. Evidence strength: strong experimental evidence for the underlying bias; scholarly consensus on how slope arguments should be evaluated; no direct trials of the technique.
- Some slopes are real. Incremental change is how many large shifts, welcome and unwelcome, have occurred; do not treat the label "slippery slope" as a refutation. If the mechanism and the track record are there, the warning deserves weight.
- Multiplying probabilities assumes the steps are independent. Often they are not: the same cause that makes the first step likely makes the rest likely too, so the naive product can be too low.
- A low-probability chain can still matter if the end state is catastrophic. The question then becomes expected harm and the cost of precaution, not only likelihood.
- Demanding a complete mechanism for every link can become an impossible standard. Apply the same scrutiny to the chains you find congenial.
- Tversky, A., & Kahneman, D. (1983). Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment. Psychological Review, 90(4), 293-315Experiments showing that detailed, representative scenarios are judged more probable than their less detailed components, contrary to the conjunction rule.
- Walton, D. (1992). Slippery Slope Arguments. Oxford University PressThe analysis of slope arguments as sometimes reasonable, with strength depending on the support for each step, and the critical questions to ask of them.
- Volokh, E. (2003). The mechanisms of the slippery slope. Harvard Law Review, 116(4), 1026-1137An account of the specific mechanisms by which one decision can make a later one more likely, which distinguishes real slopes from rhetorical ones.