When discussing opinions with friends, I often resort to making extreme scenarios out of their opinions in an attempt to investigate the limits within which their statements hold true (to them). However, most people are not happy with extreme examples and discard them right away.

Now I'm trying to better understand whether there is something fundamentally wrong in my approach. Can somebody explain why stretching arguments in order to uncover their subjectivity is wrong? If it is not wrong, then is there a name for this technique/method? I just read the term "Reductio ad absurdum", but, in my case, I am not trying to disprove a statement, merely investigate the specifics of an established opinion.

Here is a hypothetical dialog:

  • Friend: We ought to pay (health-care) for people who get sick from smoking.
  • Me: Should we also pay for people who have accidents doing drag racing?
  • Friend: Obviously, no.
  • Me: How about (other) people who had accidents because of driving an old car with aged tires?

Here's another example:

  • Friend: If you take a photo of my face, it belongs to me because the face is mine.
  • Me: How about a realistic sketch of your face?
  • Friend: That, too, belongs to me, because it is still my face.
  • Me: What if I wrote a textual description of your face, "He has big eyes shaped like almonds, etc." Would this description also belong to you?
  • Friend: No, not this one.
  • Me: What if I wrote a very detailed and precise text?
  • Friend: Still, no. It wouldn't belong to me.
  • Me: What if I noted numeric proportions on a piece of paper (for example, "the eyes are 3 cm apart and the irises' radii are 8 mm each") and then went home and built a sketch of the face from the description and numbers. Would this new sketch belong to you?
  • Friend: ...
  • Me: What if I made a list of the local color value for 16 million points inside and around your face (using my 16 megapixels digital camera)? Would this "list" belong to me or you?
  • Friend: No, that is a photograph. That would belong to me.

Through my stretched analogies I expect my friend to either:

  1. Acknowledge that he or she subjectively draws the line on what is acceptable. From there the discussion focuses on less extreme examples onto pinpointing that tipping point/zone and debating on why the boundaries are where they are.

  2. discard my example as dissimilar and explain why it doesn't relate to his or her original statement. From there I focus on defining a better (extreme) example.

2 Answers 2


I use the same approach, and it frustrates my friends in exactly the same way =)

As Frank points out, when used as a method to test bounds, it is useful. However, there are times where it is a poor tool to use in a debate. I find it runs into the most trouble when dealing with demarcation problems. Consider your example with healthcare. I can see that you're clearly trying to close in on a hard line between what we should pay for and what we should not. However, there are a few issues:

  • The person you are talking to may not know where the line should be drawn, and may not want to set it in stone at the moment. Drawing hard lines takes a great deal of intellectual effort.
  • The solution is not always a hard line. There may be some enormously complex curve between health-care for smokers and drag racers. Trying to fit that to a simple yes/no decision for each class of person may not actually be close to the real solution.
    • In many cases, the line is one of the known demarcation problems that humans tend to not fully agree upon. Issues of right/wrong, life/death, rights/responsibilities tend to fare poorly when explored with this technique.

In my experience, I use these sorts of arguments because they aren't a big deal for me. Psychologically, I'm comfortable with drawing hard lines and seeing where they lie. I'm also comfortable ripping into a partial solution because it doesn't handle the grey areas. However, from a discussion standpoint, this is not always an ideal approach.

A classic example for me is abortion. If you argue that abortion should be illegal, I may engage in this "boundary testing" approach, questioning at what point one has an alive child. If one pushes to, say, conception, I'll start to ask what part, because conception is a dynamic process that occurs over a length of time. I'll demand a hard line in the sand. Practically speaking, even if one is willing to draw such a line, it is likely an unobservable.

However, it is known that the line between alive and dead is a tremendously difficult demarcation problem for humans. We know that the line gets murky and we expend a great deal of effort to try to avoid having to draw that line with the crispness that this bisection approach would call for.

Thus, if I go down this path, it is reasonable to assume it is because all I want to do is dismantle their idea. I want to demonstrate that, for their idea to work, they have to solve one of the great demarcation problems of all time. And you know what? People don't like that.

And it turns out that people are not 100% logical Vulcans, so forcing them to engage in logic that they don't like turns out to be a less than ideal method of exploring a debate, even if it looks good on paper.


The name for checking extreme examples may be called "boundary testing". Here is how Wikipedia defines it:

Boundary testing or boundary value analysis, is where test cases are generated using the extremes of the input domain, e.g. maximum, minimum, just inside/outside boundaries, typical values, and error values. It is similar to Equivalence Partitioning but focuses on "corner cases".


The Boundary value analysis or Boundary testing is a test design technique that is used to find the errors at boundaries of input domain rather than in the center of input.


Wikipedia, "Boundary testing" https://en.wikipedia.org/wiki/Boundary_testing

  • 1
    a very good read, thank you! The extreme examples are of course iteratively refined into lesser extreme ones, until we reach a fuzzy region of indeterminacy or a distinct tipping point. My method is indeed an iterative computational method deriving from computer science - still, I was hoping to find the term in dialectics jargon! I just read about Loki's Wager [en.wikipedia.org/wiki/Loki%27s_Wager] and seems to be somewhat related, albeit in my case the goal is not to proclaim indeterminacy, but to expose subtleties in arguments that wouldn't be evident if not poked from all sides
    – FotisK
    Commented Aug 19, 2018 at 13:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .