The Uninvited Interlocutor

Bitcoin donations gratefully accepted. Thanks: 1EsJU1soVbS1MkpihYjKjDpcFKGLDdMFLU

Frictionless Fiction: The Slippery Slope

with 17 comments

There are lots of good books and websites (see:  The Fallacy Files and Your Logical Fallacy Is ) on critical thinking, logic and fallacies, but unfortunately there’s still widespread ignorance of such things.  So, I think it’s always a good idea to discuss common errors in reasoning with a view to explaining why they are errors.

The “Slippery Slope” is one of the most frequently misused and misunderstood argument forms — because those who employ it believe it to be a strong argument form when it is, in fact, the exact opposite.  It’s a commonly used phrase and most who use it are attempting to show that if one does “A” then a chain of events will occur which will eventually lead to the undesirable result “D”, for example.  The imagery is quite simple: by initiating “A” one has stepped out onto the top of a steeply angled and slippery incline, so one will slide from the top of the incline all the way to the bottom without being able to stop — and at the bottom lies the undesirable outcome (“D” in our example).  Therefore, by initiating “A” one has really set a course for “D” because “A” leads to “D” and we can’t stop that journey once we’ve begun.  So, the only way to avoid ending up at “D” is to not initiate “A.”  Here’s a little graphic:

As we can see, “A” leads straight down the slope to “D” passing through the two intermediate points of “B” and “C” — we’ve slid right from the top to the bottom without being able to stop.  This would be all well and good if the chance of going from “A” to “B” is 100% and the chance of going from “B” to “C” is 100% and from “C” to “D” is 100%.  In short, if each link in the chain is a certainty. In that case we would have something known as a hypothetical syllogism or “chain argument” in logic.  The reason for the name “chain argument” should be fairly obvious: each step (“A”, “B”, “C”, etc.) can be likened to a link in a chain and as long as each link is at 100% strength, then the chain itself will be at 100% strength.  So, as long as each step is certain, then we can be certain that if we begin at “A” then it is certain we will end up at “D” — we can be certain of sliding right to the bottom of the slope.

Where the Slippery Slope quickly weakens is when there is not certainty that each link in the chain will occur — where there is only a probability that one or more links will follow on from the previous link.  Let’s take a look at another graphic:

Here we have an example of a Slippery Slope where the transition from “B” to “C” gets a bit sticky and there is only a 70% chance that getting to “B” means proceeding on to “C.”  See what that does to the final chance of arriving at “D” now?  In the previous graphic each link was 100% certain, so we could go right from “A” to “D” knowing there was a 100% chance of getting to “D” from “A.”  Now we can only be 70% certain that if we initiate “A” that it will result in “D.”  That is, 3 times out of 10 doing “A” will not lead to “D” as a result.  One might be tempted to think those are still pretty good odds, but that is simply one flaw in one link — let’s look at what happens when there are flaws in more than one link.  Here’s another graphic:

We now have a situation where each link in the chain has only a 70% chance of leading to the next link — see what that has done to the chance of ending up at “D” now?  The weaknesses are cumlative: meaning that each time there is less than certainty the diminished chance acts on the previously diminished chance quickly weakening the overall chance of ending up at the bottom of the slope.  Now we have a situation where two thirds of the time we will not end up at the bottom of the slope — two thirds of the time “D” does not occur.  Add in more links and the final chance will plummet further.  This is a fairly simple example with only four links at quite a high probability of 70% — imagine what happens to an argument with six, ten, fifty or over one hundred links — a common characteristic of most slippery slope arguments is that the starting point is far removed from the undesirable end point.

The crucial point to understand is — Slippery Slope arguments are inherently weak arguments.  When someone argues using a Slippery Slope they are using a very suspect methodology that should never be accepted at face value.  One always needs to look at the number of potential links in the slope and the likeliness of each link occurring. Most times when someone is using a Slippery Slope argument they are attempting to dissuade the initiation of an action by arguing the inevitability of a resultant undesirable situation — they are generally not employing a hypothetical syllogism (“chain argument”) and therefore the outcome is not certain — only probable.  That means that it’s very unlikely a given Slippery Slope situation will be inevitable — and if it is found to be, then it’s a hypothetical syllogism.  Some may, indeed, prove to be highly likely — but a Slippery Slope always needs to be carefully examined.

Often times the arguer also ignores the fact that we can frequently choose to stop our slide down the slope by exercising our will to choose as individuals and societies — we can set limits to certain actions in law and by social convention. The arguer’s aim is to present a situation where real choice is taken out of the equation in order to simplify the argument down to a forced binary choice of initiating “A” or not initiating “A.”  By presenting the argument as if there is no real choice, no braking point on the slope, the arguer seeks to strengthen the overall argument by forcing a false dichotomy — an artificially limited choice between the two alternatives of:  “A” which means not avoiding “D” or  “Not A” which means avoiding “D.  In the vast majority of cases, the situation is much more subtle and nuanced — and when someone argues in such a way to grossly simplify the situation you should be wary that it is because complexity, nuance and careful consideration hurts his or her case.  In short, the arguer is pulling a fast one in order to hide inadequacies in the overall argument.

So, next time someone says something like, “But that would put us on a slippery slope to…” understand it’s probably quite unlikely and almost definitely not a certainty.  At the very least, when you hear the phrase you should be aware that a careful examination of the case is immediately warranted.

EXAMPLE: Interestingly, just hours after I posted this there was a good example on UK Channel 4 News of an appeal to a Slippery Slope in a piece discussing gun control in the US in the wake of the Aurora shootings (at 2:55):

The arguer states that, “it [gun control] is a Slippery Slope once you start down that path taking away our freedoms.”  Imagine how many intermediate steps there could conceivably be between initiating some form of stricter gun control (e.g. simply banning large magazines as is suggested in the video) and the end point of  “taking away our freedoms” — then consider the example above with only four stages in the slope and how quickly it weakens. The arguer attempts to end the discussion before it starts by making it seem that if we take even the tiniest measure of gun control, then the state will end up “taking away freedoms” wholesale.  He also ignores the fact that society can choose to draw the line where it wishes (put the brakes on at any point on the slope) and, for example, control the availability of weaponry but not remove rights to free speech, assembly, etc.  Or, simply control the availability of certain types of weaponry and ammunition without banning all weapons outright.


Written by Uninvited Interlocutor

July 25, 2012 at 3:24 pm

17 Responses

Subscribe to comments with RSS.

  1. Reblogged this on Homo economicus' Weblog and commented:
    Thanks to Stephen Fry for sharing on Twitter. Beware slippery slope arguments.

    John Sargeant

    November 29, 2012 at 9:38 pm

  2. Great article. But a stray ‘I.e.’ In the footnote? Surely ‘e.g.’?


    November 29, 2012 at 9:47 pm

  3. Hey man. Just some schmuck from Missouri who thinks this was a great article. Good work.


    November 30, 2012 at 12:48 am

  4. Excellent break down of the slippery slope fallacy but I don’t agree with you attributing gun control worries to it. Your application of Bayesian inference only works so long as there isn’t a strong motive to reach from point A all the way down to D. Unfortunately this assumption is ludicrous. There are plenty of rather obvious political reasons why leaders as a rule do not desire a well armed populace. The only reason why the USA for example have such a liberal legal stance on gun ownership is because the constitution was written by revolutionaries who desired a strong militia. Now though, they are the ones in power and they would very much like to keep it safe from the rabble: us.

    Rational Mystic

    November 30, 2012 at 2:16 am

    • I should make clear that I highlighted an example of someone arguing against gun control who employed a slippery slope argument. That doesn’t mean all arguments against gun control are slippery slope arguments — although, quite frequently they are.

      Uninvited Interlocutor

      November 30, 2012 at 10:06 am

  5. Really wish I could have shown this article to my father growing up. Things just aren’t as black and white as the more strongly opinionated, aggressive thinkers see them!


    November 30, 2012 at 4:03 am

  6. A complete dismissal of the “slippery slope” is as bad as a complete adherence to it.

    Some situations where slippery slope might be relevant:

    – where the end result is very very bad indeed, but where remaining at A is “OK”, even if not perfect. You say yourself, it’s a probability at each link of the chain. 1% chance of an orwellian society? That’s too high for me.

    – where the issue at stake is one where fervent opinions come into play; where a critical mass may be reached. You assume that at each link in the chain, people will abide by careful reason. But for a taboo issue, where the populace gets angry, you may find legislation pumped 1-2-3 and all the way down the slope we go. (Or at least a little further than we like)

    – after progressing just one link in this imaginary “chain”, popular opinion and “feeling” may change. One piece of legislation moves through, and after some time the populace becomes comfortable with it (or at least learns to live with it). After some time has passed, it may be easier to push through the next piece of legislation.

    – chains? links? In reality it may be a gradient, or a continuous state space with pools of possibility and momentum playing a key part.

    Human emotion and sentiment isn’t logic. It’s more like the weather, or the stock market. Sometimes there are storms, sometimes there are unstoppable crashes. There may be – depending on the issue – some sense to a slippery slope argument, and saying it should be ignored because you can make an abstract little graph of it is just as wrong as blind adherence to it.


    November 30, 2012 at 11:32 am

    • I didn’t completely dismiss slippery slope arguments — I said they can, at times, yield a high probability. But also, that they always warrant close inspection because, by their nature, they are inherently weak.

      1% chance of an Orwellian society is a ridiculous assertion — based upon what argument? That’s just another appeal to fear based on nothing in an attempt to end the discussion before it begins. Furthermore, while slippery slopes do rely upon probability, that does not mean all slippery slopes entail some tangible probable chance to reach their proposed outcome — that is, arguing, for example, that allowing gay marriage will lead to a chain of events resulting in the sun exploding is such a non-sequitur we need not concern ourselves with the assertion.

      Also, you appear to become entangled in a contradiction when you state, “Human emotion and sentiment isn’t logic. It’s more like the weather, or the stock market.” because you are making the case that the unpredictable nature of life enhances the accuracy of a predictive model in the form of a slippery slope argument.

      While this is my particular analysis of the slippery slope argument utilizing “abstract little graphs” I’m hardly alone or out on a radical limb. This is the accepted, rational view with regard to this argument form — it’s inherently weak — so much so that it is often viewed as fallacious. Even a source as non-specialized as Wikipedia presents the topic in a similar manner — explaining the weaknesses. Take any decent course in, or read any decent book on, informal logic, reasoning or argumentation and you will find this type of analysis.

      Uninvited Interlocutor

      November 30, 2012 at 11:53 am

  7. This simplistic argument misses the point completely. Life is not like this. Once we have moved from A to B an argument to move to C becomes easier to justify because we are no longer at point A – and so it goes on down the slope. This paper is completely academic.


    November 30, 2012 at 11:54 am

    • Why does it become easier to make the next move? What’s the argument? Why haven’t all instances of situations that can be characterized as slippery slopes ended up at their final destinations?

      Furthermore, this is a circular argument — begging the question:

      Once we have moved from A to B an argument to move to C becomes easier to justify because we are no longer at point A…

      Uninvited Interlocutor

      November 30, 2012 at 11:57 am

  8. It could be argued that a slippery slope argument is never a case of ‘A will lead inevitably to D’, as in such a defined case, a decision to take step A is the decision to take steps B, C and D. So in these cases (rare), the progress from A to D is not a slippery slope, but inevitable. Therefore, when a slippery slope argument is applicable, D can only be a possible outcome, and other outcomes, or other D’s, must exist.

    For me this means that whenever a slippery slope argument is made, it must be questioned, as it is never inevitable.

    It can also be argued that a slippery slope argument is always a negative argument. If D was seen as a positive outcome, it would never be presented as a slippery slope. Not just another reason for questioning a slippery slope, more of a given. It is for me, as I always question slippery slope arguments, and I always start from the position that the argued outcome is wrong, or at the very least, unlikely.

    It is my conclusion on such arguments that bother me the most. The D presented in such arguments is usually emotive, and designed to appeal on a superficial, sound bite, level as a way of preventing A. This is the power of the slippery slope argument, as it gives the anti A’s a powerful, but specious, sound bite argument that will be listened to by a lot of people. Because of this effect, it becomes more important not only to always question these arguments, but to question them in the same forum that they were first presented.

    John T Shallow

    November 30, 2012 at 3:16 pm

    • Some very good points here on the always negative nature of the resultant outcome of slippery slopes (where are the slippery slopes to positivity/happiness?) and the overly emotive characterization of the end result (“it’s a road to an Orwellian society!”) — the slippery slope leverages the fear of emotively charged negative outcomes in order to convince or dissuade.

      Uninvited Interlocutor

      November 30, 2012 at 11:52 pm

  9. Perhaps when all points on the slope from A to D have a clear causal link, the slippery slope argument becomes more valid. Obviously there is no clear causal link between legalising gay marriage and causing the sun to explode so the slippery slope argument is false there. However, banning smoking indoors in a public place does have a causal link to making it illegal to smoke indoors in a rented apartment or separating children from smoker parents (which is literally what many anti-smoking lobbyist groups in California are trying to do and sadly succeeding to a large degree). This obviously does not automatically make this line of argument valid. you still have probability to contend with, but the probability of these events have an intelligent agent behind them: the asshole trying to take your liberties away from you.

    Rational Mystic

    November 30, 2012 at 7:19 pm

  10. Not having any real logic training I found this very interesting. It seems to me that most people who reffer to the argument may in fact have a more ‘foot in the door’ or ‘camel’s nose’ objection in mind. That is they feel that some agency may find D desirable and by convincing them that A is reasonable and necesary may hope to do the same with B, C and D over time as they adjust and get used to the change. For example American gun enthusiasts (with whom I disagree by the way) would be correct to fear that the gun control lobby has more in mind in the long term than regulating magazine length. With regard to the British press; the legislation recomended by Lord Justice Levinson would actually enshrine the freedom of the press rather than chip away at it or send it sliding down the slope.

    Dan Hales

    November 30, 2012 at 11:13 pm

    • “Camel’s nose,” “thin end of the wedge,” or “foot in the door” are all just synonyms for the same argument form. I think you’re right that many people see them as somehow different from what I’ve described, but they’re just different names for the same thing. The issue always lies with the likelihood of progressing beyond the initial step or beyond what’s deemed reasonable/effective/warranted, etc. There will certainly be some cases where the fear of progression is justified, but the longer the slope (i.e., the greater the number of steps involved in order to get to the final undesirable end) the less likely that fear is warranted.

      The takeaway message should be: if your sole argument is a slippery slope and the starting point is removed from the end point by more than a few steps, then in most cases your position is not going to be overly well supported. In such a case, it’s best to find another argument to support your case — and if you can’t find one, then it might be wise to reconsider your position. If you can find one, then it’s probably best to leave the slippery slope behind and just go with the better argument. In both cases, it works out that it is best to drop the slippery slope argument.

      Uninvited Interlocutor

      November 30, 2012 at 11:30 pm

  11. Thanks for shedding light on the slippery slope. Seems it’s not that slippery after all! The slippery aspect is a mirage for the most part, and it seems to vanish upon closer inspection.

    Tamer (@onetamer)

    December 3, 2012 at 9:39 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: