Criticising an Argument


It is crucial to understand the approaches to criticising philosophical positions which philosophers employ—both to use these tools to criticise others’ work, and to anticipate criticisms of your own work. Here, we’ll explore three common approaches to criticising philosophical positions—reductio ad absurdum, counterexamples, and dilemmas. For advice on criticising philosophical arguments see the page on argumentative fallacies.







A common way of arguing against a philosophical position is to provide a counterexample to it. Counterexamples are best employed when the view to be criticised makes a generalisation—for instance, a statement like “All dogs have four legs”. To provide a counterexample to a generalisation, simply give an exception to the rule—a single case where the generalisation is false. In this case, point out a three-legged dog.

The key to a good counterexample is to make it clear, simple and plausible. For instance, pointing out a three-legged dog is a much better counterexample to “All dogs have four legs” than inventing a scenario in which a mad scientist genetically enhances a dog to have six legs. The mad scientist case would be a counterexample, but is less simple and less plausible than the well-known case of three-legged dogs (It’s also less clear—it’s not obvious that the genetically enhanced dog is a dog as we know it).

Formulating counterexamples is an area that allows you to unleash your creativity and inventiveness as a philosopher. Creating your own clear, simple and plausible counterexample to a thesis is a great way to add originality to your paper. So is making a well-known counterexample clearer and simpler (though be careful to make sure it still works as a counterexample!)

There are two broad types of counterexample—propositional and argumentative counterexamples.

Propositional counterexamples are exceptions to statements—like our three-legged dog.

Argumentative counterexamples show that a particular form of argument is invalid by giving an argument with the same logical form which is obviously invalid. These can be very helpful in pointing out logical fallacies.

For instance, suppose I argue:

If Plato is an idealist, then Berkeley is the inheritor of Plato.

Berkeley is the inheritor of Plato

Therefore, Plato is an idealist.

You may find it hard to criticise this argument on the face of it, especially if you’re unfamiliar with Plato, Berkeley, and idealism. But you don’t have to know anything about them to show that this is fallacious. You can simply provide another argument of the same logical form (PàQ, Q, therefore P) which is obviously invalid, for example:

If unicorns exist, then the moon orbits the earth.

The moon orbits the earth.

Therefore, unicorns exist.


What to avoid:

There are several pitfalls to avoid when creating your own counterexamples.

  1. Just because an argument for a position is fallacious or invalid, does not entail that the position itself is wrong. If you discredit an argument for X, be careful not to claim this means X is false! You will need a separate argument for that. It’s often prudent to make the thesis of your paper more nuanced—for instance, instead of arguing “Realism is false”, argue “The No Miracles Argument for Realism fails to show that Realism is true.” This mistake is sometimes called the Argument from Fallacy or the Fallacy Fallacy.
  2. Always apply the Principle of Charity to any argument or position you criticise. If a position can be easily rephrased or reformulated to avoid your counterexample, you should amend the position yourself. A strong counterexample refutes the thesis at its strongest. Otherwise, you will be accused of defeating a straw man.
  3. Do not use analogies as counterexamples. Analogical arguments can be powerful, but are not appropriate counterexamples. For instance, if I argue:

    Abortion is morally acceptable because a woman has an absolute right to determine what happens with her reproductive system.

    Then the analogous argument:

    Punching people in the face is morally acceptable because a woman has an absolute right to determine what happens with her hand.

    is not a good counterexample. While the two are analogous argumentative forms, I am not committed to the claim ‘a woman has an absolute right to determine what happens with her hand’ just because I committed to the claim ‘A woman has an absolute right to determine what happens with her reproductive system’, so the counterexample fails.


How to respond to counterexamples

Suppose someone formulates a counterexample to a thesis you want to defend. There are several options open to you.

  1. Show that the counterexample is not a true counterexample.
  2. Modify your thesis to accommodate the counterexample.

There are several ways to modify your thesis. One is to weaken the scope of your claim. For instance, suppose I claim “Killing is always immoral”. You provide the counterexample of killing hostile soldiers during war, so I weaken my scope to: “Killing civilians is always immoral”. This may seem like a retreat, but is often a good thing—it allows you to refine your thesis and state more precisely what you mean.

Another way is to move from a universal generalisation to a weaker claim—for instance “Killing is usually immoral”. This can be a very problematic move, as claims about what often or usually happens are (usually!) defended only by empirical evidence.

A third approach is to exclude the specific counterexample by regarding it as a special case in some way. For instance, I could change my thesis to: “Killing is always immoral, unless in the context of warfare”. It is important to provide clear reasons why the counterexample is exceptional. If you don’t provide good reasons, you might be accused of modifying the thesis purely to evade refutation—sometimes called ad hoc modification. A particular form of fallacious thesis-modifying is called the no true Scotsman fallacy—this is when the thesis is subtly changed to deliberately exclude counterexamples. For instance:

“No Scotsman ever moves to London.”

On being given an example of a Scotsman who has moved to London, changing the thesis to:

“No true Scotsman ever moves to London.”

Here the intention is that ‘true Scotsmen’ are those Scotsmen who don’t move to London, making the thesis trivially true.




Reductio ad Absurdum


Reductio ad absurdum translates to ‘reduction to absurdity’. It is a powerful philosophical tool, used to show that a position must be false, because absurd consequences would follow if it were true. It uses the logical form modus tollens:

P -> Q

Not Q

Therefore, Not P

This is a logically valid argument.

When constructing a reductio argument, the crucial step is deducing a consequence of the position which is sufficiently absurd as to be unacceptable. The best reductio arguments will deduce a contradiction from the thesis. However, reductio ad absurdum arguments can also be constructed by deducing a simply unpalatable consequence from the thesis. For instance:

If determinism is true, then no actions are morally right or wrong.

Actions are morally right and wrong

Therefore, Determinism is false.


Responding to Reductio arguments

There are two ways to respond to a reductio:

  1. Bite the bullet. ‘Biting the bullet’ means accepting the unpalatable consequence. For instance, in the determinism example, a determinist could simply accept that there is no morality.
  2. Refute the argument—show that the absurd consequence does not actually follow from the position. In the example above, this would mean showing that actions can actually be moral or immoral despite determinism.






Dilemmas are complex forms of reductio argument, where there are two (and only two) possible options for a defendant of a position, and both lead to unpalatable consequences. Note that this differs from the colloquial sense of ‘dilemma’ that you might encounter in, say, ethics—an ethical dilemma is a situation in which it is difficult to decide which course of action is the right one. Argumentative dilemmas are distinct from these tricky problems.

A dilemma has the logical form:

P -> (Q or R)

Q -> S and R -> T

(Not S) and (Not T)

Therefore, Not P

Again, this is a valid inference.

For instance, consider the following argument:

If determinism is true then either we are not morally responsible for our actions or we are morally responsible for actions we could not have avoided doing. (P -> (Q or R))

If we are not morally responsible for our actions then all punishment and reward is unfounded, and if we are morally responsible for actions we could not have avoided doing then morality is just a matter of luck. ((Q -> S) and (R -> T))

It is not the case that all punishment and reward is unfounded, and it is not the case that morality is just a matter of luck (Not S and Not T)

Therefore, determinism is false. (Not P)

In a dilemma, we call the P -> (Q or R) premise the disjunctive premise (as it contains an ‘or’ phrase) and the (Q -> S and R -> T) premise the conjunctive premise (as it contains an ‘and’ phrase).

In order for a dilemma to be sound, it must be the case that

  1. The absurd consequences (S and T) do follow from the two options (Q and R)
  2. The two options (Q and R) are the only options available for a defender of P.


How to respond to a dilemma

There are three ways out of a dilemma:

  1. Bite the bullet. As with a reductio argument, if the absurd consequences are not contradictions, you have the option to bite the bullet and accept one of the absurd consequences, perhaps arguing that it is not as absurd as it might seem. For instance, a defender of determinism might accept that morality is a matter of luck.
  2. Go between the horns. The two options in a dilemma are called the “horns of the dilemma”. Going between the horns means showing that there are in fact other options available—in other words, showing that the disjunctive premise is false. If you successfully go between the horns, you have shown that the argument was a false dichotomy.
  3. Grasp a horn. Grasping one of the horns of the dilemma means showing that one of the absurd consequences which is supposed to follow from a horn of the dilemma does not in fact follow. So, for instance, the proponent of determinism could grasp the first horn of my dilemma by showing that despite no one being morally responsible for their acts, punishment and reward systems are morally justified.