Is the Slippery Slope a Fallacy?

Well…yes…but also…no.

Let me explain.

First, I need to address what is perhaps the largest obstacle of this discussion: what precisely is the slippery slope? I think the best way to navigate this issue is to look at the different points of view one by one and build up from that. I will begin with an excellent website: yourlogicalfallacyis.com (which you should bookmark immediately #NotSponsored). As the name of the site suggests, they characterize the slippery slope as a logical fallacy—a flaw in human reasoning; an error in the way human beings think.

From the website’s definition of a slippery slope:

You said that if we allow A to happen, then Z will eventually happen too, therefore A should not happen.

The problem with this reasoning is that it avoids engaging with the issue at hand, and instead shifts attention to extreme hypotheticals. Because no proof is presented to show that such extreme hypotheticals will in fact occur, this fallacy has the form of an appeal to emotion fallacy by leveraging fear. In effect the argument at hand is unfairly tainted by unsubstantiated conjecture.

So, according to this popular website, they say Yes: the slippery slope is a fallacy. If A is allowed to happen, then B would happen as a result, followed by C, all the way to Z, and Z is an undesirable outcome, something we wish to never happen. Therefore, A should not be allowed to happen to prevent Z from happening. The reason they say this is a fallacy because there is either little or no evidence that Z will result from A. It is simply an extreme hypothetical—and only a hypothetical—without real evidence to suggest it is probable. If they are right, it seems right to call the slippery slope a fallacy.

A Concrete Example

A common slippery slope argument I have seen in recent years goes something like this: suppose we allow same-sex marriage in America. This will result in people making similar demands for the allowance of polygamy, or polyandry, or the marriage of inanimate objects like cars, and ultimately the allowance of child marriages. We as a society should clearly never permit children to marry since this would be taken advantage of by pedophiles and other unsavory characters. Therefore, we need to keep the gate closed from the start by opposing same-sex marriage.

Is this a good argument?

Well, maybe. While it is certainly true that certain pro-pedophile/pederasty groups, such as NAMBLA, have grown bolder in recent years in their advocacy, it is very difficult to determine if this was caused by, say, the Supreme Court ruling in Obergefell v. Hodges concerning same-sex marriage. For one, members of the LGBT collective have vehemently resisted pedophiles joining their ranks persistently. For another, as we all know, correlation does not imply causation. However, what is clear is that—in fact—the so-called “extreme hypothetical” of increased demands for acceptance of pedophilia is not actually a hypothetical at all. A quick hop onto Twitter will show you how bold pedophiles have gotten in recent years: there are a great many pedophiles on the site, all putting the description “MAP” (“Minor-Attracted Person”, an insidious euphemism for “pedophile”) and/or clown emojis in their Twitter bios to advertise themselves as “proud” pedophiles. So, on the one hand, this slippery slope argument seems quite weak, but on the other hand, the conclusion appears to be true. What is going on here?

A Closer Look

On the face of it, it seems to me that the example of the slippery slope argument I have provided is incredibly weak. First, it does not address whether same-sex marriage ought to be accepted or rejected on its own merits, which is just as important—if not more important—than any possible side-effects that accepting it might have. Second, the causal chain from “permitting same-sex marriage” to “permitting pedophilia” is vague and not clearly enumerated (how many intermediary steps are there? One? Three? A billion?). Third, there is no evidence presented in the argument to support the proposition that same-sex marriage would inevitably lead to pedophilia becoming accepted in society. Even if it would, the argument gives no evidence for it, and thus the claim is not properly defended. Therefore, I feel comfortable calling this argument weak.

So…how is it that the conclusion is still true?

First, regarding inductive arguments such as the slippery slope, an argument’s being weak or strong does not tell you anything about the truth of the argument’s conclusion. An argument’s weakness or strength can only tell you the general likelihood of the conclusion being true. Well, that’s alright. Unlikely events happen all the time (someone always wins the lottery). It might also be the case that the argument itself is just poorly defended or poorly organized, meaning it could potentially be a lot stronger if the proper time and research were devoted to improving and strengthening it. On the other hand, perhaps it impossible to craft a victorious slippery slope argument — the structure of the argument itself might be inherently flawed in some way. I think this is the next place we should investigate. Let’s have a look at several different ways in which the slippery slope is articulated to see if any of them succeed as arguments.

The “Domino Effect”

If you set up a line of dominoes and knock over the first domino in the line, it will knock over the next domino, which will knock over the next domino, and so on until every domino down the line is knocked over. Apply this concept to real-world events, and you have what is called a “Domino Effect,” in which an initial event triggers a series of events that ends with some terminal event. If the terminal event is bad, you need to make sure that the initial event that started it all never happens. This is one way some people have chosen to improve the slippery slope argument.

The nice thing about the domino analogy is that every single step in the chain of events (represented by each domino in the line, where every domino is a separate event) is accounted for, unlike a slippery slope, which is by its very nature a gradient with no clearly identifiable steps. However, it poses its own challenges: now that you have identified each event in the chain, you must give a separate argument for each event. You cannot just say, “If A will happen, then B will happen, and then C, and then D, all the way to Z.” Instead, you need to demonstrate that B will likely result from A, and then demonstrate that C will likely result from B, and then demonstrate that D will likely result from C, and by then you will find that A and Z do not actually have anything to do with each other after all. It would actually be far better for your case if you could demonstrate that Z will likely result from A.

There is another problem: tumbling dominoes are easy to interrupt as long as you’re quick, or if you have prepared beforehand. All you have to do is block the line of dominoes with your hand, and whatever dominoes are behind your hand will be unaffected by the line of falling dominoes in front of your hand. In the same way, separate but related events can also be interrupted through intentional human intervention. In our earlier “from same-sex marriage to pedophilia” example, the cultural or political acceptance of pedophilia can be prevented through deliberate human intervention even with the legal recognition of same-sex marriages. Politics is downstream from culture, so as long as our culture continues to resist accepting pedophilia (which it does quite well even without social engineers meddling behind the scenes), there is no risk of a domino effect. Or, if there is a risk, it is a very small, very acceptable risk.

The Thin End of a Wedge

A woodcutter’s ax has a wedge-shaped head. The thin end of the wedge is very sharp, making it ideal for cutting wood, but the whole head is involved since the wedge shape forces the wood to split apart as the head travels downward through the log. Thus, the thin end of the wedge leads to an incremental widening of the log until the whole thing splits in half.

Is this a superior version of the slippery slope? Hardly. It is even worse than the domino analogy. The reason for this is quite simple: cutting a log is a single event. More precisely, each chop of the ax a single, isolated event that is independent of every other chop of the ax. There is no causal chain from A to Z. There is only event A: a swing and a split log. Okay, maybe you could divide the swing of the ax and the splitting of the log into two separate events, but this does not constitute a chain either.

What about the incremental widening of the log as the wedge passes through it? Surely that is where the analogy really lies. The log is widened further, and further, and further until it finally splits. Even if this were done very slowly (not with an actual swing of the ax), these increments could be observed and the log would still eventually split. Isn’t that a fair way to describe a slippery slope argument?

I’m not convinced. On the one hand, a whole wedge is a single object. The thin end and the wide end are not two separate events or a series of events. They could only be understood to be the same event. On the other hand, this idea of widening sounds more like an application of a principle instead of one event triggering another. It seems to me like this analogy is just not useful for characterizing a causal slippery slope. Maybe one could use it to describe a slippery slope resulting from a particular decision that has wide-reaching logical implications that, if followed consistently, could result in a much wider application than what was intended. However, I would not call that a slippery slope, and I don’t think most others would either. We are here to talk about chains of events, not logical implications.

A Dam Burst

When a dam breaks, catastrophe ensues. Thus, we must do whatever we can to ensure that the damn never bursts. That sounds a lot like an analogy for the slippery slope argument, doesn’t it? Not quite. A ruptured dam does not result from a step-by-step process, but rather by a failure of the dam to withstand the weight of the water pushing against it. The water is pushing against the dam all at once rather than in increments, and eventually, the dam fails due to its own intrinsic flaws. In a slippery slope argument, the idea is that we must prevent A to prevent Z, but in the dam analogy, Z is the dam bursting and A is the water. A slippery slope argument would say that to prevent the dam from bursting, we need to remove all of the water pushing against it. That’s absurd. It would make much more sense to simply repair or reinforce the dam. So, it looks like a dam burst is not a good analogy for the slippery slope argument.

It looks to me that none of these ways of looking at the slippery slope argument have done anything to rescue it from fallacious reasoning. However, we have been treating the slippery slope as an inductive argument so far. What if we reframed the argument using deductive reasoning?

A Successful Deductive Model

There is a type of argument in logic called the “hypothetical syllogism,” which expressed as follows in logical notation:

P → Q

Q → R

∴ P → R

Again, in plain English:

If P is true, then Q is true

If Q is true, then R is true

Therefore, if P is true, then R is true

Using deductive reasoning, the connection between each term is logically necessary, which means that as long as the terms are true, they will always be connected and thus we can rationally connect A to Z through a process of logical steps. Here’s how it might look:

If A is allowed, then B is allowed

If B is allowed, then C is allowed

If C is allowed, then D is allowed…

…

…If Y is allowed, then Z is allowed

Therefore, if A is allowed, then Z is allowed

If each line in the argument is true, then it is literally impossible for the conclusion to be false, and we have a successful slippery slope argument on our hands. However, since we are switching from induction to deduction, there are a few differences from the traditional slippery slope argument that should be addressed.

1. The original inductive slippery slope pertains to causal events; the deductive version pertains to logical relationships.
2. For the deductive version to be valid, each line of the argument must be carefully analyzed both independently of the rest of the argument as well as in context with the entire argument as a whole, meaning ambiguity is absolutely forbidden.
3. The deductive version cannot tell you whether A should be allowed to happen or not. It can only tell you that Z logically follows from A.
4. The real world is not perfectly rational. The deductive version cannot tell you if either A or Z will actually happen or not. It can only tell you that Z logically follows from A.
5. Deductive arguments are far better suited to ideas and principles, rather than practical events in the real world, which is probably why the original slippery slope argument is inductive rather than deductive.

The Doctrine of Double-Effect

So, what should we do with the slippery slope argument? How are we supposed to prevent undesirable events from happening if we can’t even be sure how they will come about, or even if they will come about? I actually think that this might be the wrong way of looking at this issue. Since we are talking about what we ought to do and allowing and preventing certain events, I think we should consider this an exercise in ethics. In that case, let us take a look at Thomas Aquinas’ famous Doctrine of Double-Effect.

Aquinas observed that sometimes in the process of doing good, evil is produced as a double-effect of your good action. The example that he gave was self-defense. All human beings naturally seek to preserve their own lives, but sometimes this results in the death of another person as a double-effect of protecting yourself. If a man is charging at you with a knife, and you shoot him, you have successfully done something good: you have preserved your own life, but as a result, something terrible has happened: a man is dead.

What does this have to do with the slippery slope? A slippery slope argument would say that you ought not to protect yourself because that would result in another person dying, which is unacceptable and ought to be avoided. Aquinas would say that is ridiculous. Your intention was never to end another person’s life. The death of your attacker is a tragic byproduct of you defending yourself. It was never your objective. Your objective was merely to preserve your own life, which is very clearly good, desirable, and natural. The death of one’s attacker is actually irrelevant to the question of whether you ought to defend yourself from attackers. All that matters is whether what you are doing is good or evil based on its own merits.

Let’s go back to our same-sex marriage example. Let us set aside the politics and assume, for the sake of argument, that same-sex marriage is good and desirable. The double-effect is that pedophiles are now feeling emboldened and seeking to be included in the LGBT sphere of influence. Now, perhaps we as a society believe that same-sex marriage is worth the nasty double-effect of pedophiles being Brave Online™. In that case, we should decide to go through with allowing same-sex marriage (which is in fact what we did) regardless of any undesirable consequences, because our intention is not to recognize pedophilia as a legitimate sexual preference (it is quite illegitimate, in fact) but rather to recognize same-sex marriage as a valid civic institution. The only thing that really matters is whether or not same-sex marriage is, in fact, a good thing, but that is a discussion for another time. I am only using it here as a concrete example.

Conclusion

Based on all of this, it seems to me that the slippery slope argument is not very useful for most purposes. I think that thinking of issues in terms of slippery slopes might actually distract from the real ethical discussions that we ought to be having. Slippery slope arguments reek of consequentialism, which is the idea that an action is good or bad based on the type of outcome that it produces, regardless of what the action itself actually is. I am not a consequentialist, so I tend to find slippery slope arguments wholly unpersuasive. I think it is much more profitable to go the way of Aristotle or Aquinas and examine each action or behavior based on its own traits or characteristics and determine whether it is good or evil that way. Thinking about problems that way, we find that there is no slippery slope at all. There are just different branching paths, and all we need to do is choose the best path(s) to take.

But, of course, to do that, we would need a map…not the hypothetical doom scenarios from our slippery slope arguments.