Read this tutorial, which defines the most common fallacies. This list narrows down some of the fallacies seen in the previous reading and is enough to get us started. We will look at a wider sample of fallacies later on in this course. For now, focus on being able to define each fallacy and identify the differences between the fallacies on the list.
Here are some examples of common fallacies:
A theory is discarded, not because of any evidence against it or lack of evidence for it, but because of the person who argues for it. For example:
The truth of a claim is established only on the basis of lack of evidence against it. A simple obvious example of this fallacy is to argue that unicorns exist because there is no evidence against such a claim. At first sight, it seems many theories we call scientific involve this fallacy. For example, the first law of thermodynamics holds because so far there has not been any negative instance that would serve as evidence against it.
But notice, as in cases such as this one, there is evidence for the law, namely positive examples. Notice also that this fallacy does not apply to situations where there are only two rival claims and one has already been falsified, then we may justly establish the truth of the other even if we cannot find evidence for or against it.
When offering an argument, pity is appealed to. Usually this happens when people argue for special treatment on the basis of their need.
For example, a student argues that the teacher should let them pass the examination because he/she needs it in order to graduate. Of course, pity might be a relevant consideration in certain conditions, as in contexts involving charity.
The truth of a claim is established only on the basis of its popularity and familiarity. This is the fallacy committed by many commercials. Surely you have heard of commercials implying that we should buy a certain product because it has made to the top of a sales rank, or because the brand is the city's "favorite."
Inferring that P is true solely because Q is true, and it is also true that if P is true, Q is true.
The problem with this type of reasoning is that it ignores the possibility that other conditions apart from P might lead to Q. For example, if there is a traffic jam, a colleague may be late for work. But if we argue from his being late to there being a traffic jam, we are guilty of this fallacy – the colleague may be late due to a faulty alarm clock.
Of course, if we have evidence showing that P is the only or most likely condition that leads to Q, then we can infer that P is likely to be true without committing a fallacy.
In arguing for a claim, the claim itself is already assumed in the premise.
For example, "God exists because this is what the Bible says, and the Bible is reliable because it is the word of God."
A question is posed in such a way that a person, no matter what answer they give to the question, will inevitably commit themselves to some other claim, which should not be presupposed in the context in question.
A common tactic is to ask a yes-no question that tricks people to agree to something they never intended to say. For example, if you are asked, "are you still as self-centered as you used to be?" no matter whether you answer "yes" or "no," you are bound to admit you were self-centered in the past. Of course, the same question would not count as a fallacy if the presupposition of the question is accepted in the conversational context.
The whole is assumed to have the same properties as its parts. Anne may be humorous and fun-loving and an excellent person to invite to the party. The same may be true of Ben, Chris, and David, when considered individually. But it does not follow that it will be a good idea to invite all of them to the party. Perhaps they hate each other and the party will be ruined.
Inferring that Q is false just because if P is true, Q is also true, but P is false.
This fallacy is similar to the fallacy of affirming the consequent. Again the problem is that some alternative explanation or cause might be overlooked. Although P is false, some other condition might be sufficient to make Q true.
For example, if there is a traffic jam, a colleague may be late for work. But it is not right to argue in the light of a smooth traffic that the colleague will not be late. Again, the colleague's alarm clock may have stopped working.
The parts of a whole is assumed to have the same properties of the whole. It is possible that, on a whole, a company is effective, while some of its departments are not. It would be inappropriate to assume they are all effective.
Putting forward an argument where a word changes meaning without having it pointed out.
For example, some philosophers argue that all acts are selfish. Even if you strive to serve others, you are still acting selfishly because your act is just to satisfy your desire to serve others. But surely the word "selfish" means differently in the premise and the conclusion when we say a person is selfish we usually mean that he does not strive to serve others. To say that a person is selfish because he is doing something he wants, even when what he wants is to help others, is to use the term "selfish" with a different meaning.
Presenting a limited set of alternatives when there are others that are worth considering in the context. For example, "Every person is either my enemy or my friend. If they are my enemy I should hate them. If they are my friend I should love them. So I should either love them or hate them." Obviously, the conclusion is too extreme because most people are neither your enemy nor your friend.
Assumption is made to take some independent statistics as dependent. The untrained mind tends to think that, e.g. if a fair coin is tossed five times and the results are all heads, then the next toss will more likely be a tail. It will not be, however. If the coin is fair, the result for each toss is completely independent of the others. Notice the fallacy hinges on the fact that the final result is not known. Had the final result been known already, the statistics would have been dependent.
Thinking that because X derives from Y, and Y has a certain property, X must have the same property also.
For example, "His father is a criminal, so he must also be up to no good."
A conclusion is drawn which does not follow from the premise. This is not a specific fallacy but a general term for a bad argument. So a lot of the examples above and below can be said to be non sequitur.
Inferring that X must be the cause of Y just because X is followed by Y.
For example, having visited a graveyard, I fell ill and infer that graveyards are spooky places that cause illnesses. Of course, this inference is not warranted since this might just be a coincidence. However, a lot of superstitious beliefs commit this fallacy.
Within an argument some irrelevant issue is raised which diverts attention from the main subject. The function of the red herring is sometimes to help express a strong, biased opinion. The red herring (the irrelevant issue) serves to increase the force of the argument in a very misleading manner.
For example, in a debate as to whether God exists, someone might argue that believing in God gives peace and meaning to many people's lives. This would be an example of a red herring since whether religions can have a positive effect on people is irrelevant to the question of the existence of God. The good psychological effect of a belief is not a reason for thinking that the belief is true.
Arguing that if an opponent were to accept some claim C1, then he or she has to accept some other closely related claim C2, which in turn commits the opponent to a still further claim C3, eventually leading to the conclusion that the opponent is committed to something absurd or obviously unacceptable.
This style of argumentation constitutes a fallacy only when it is inappropriate to think if one were to accept the initial claim, one must accept all the other claims.
For example, "The government should not prohibit drugs. Otherwise the government should also ban alcohol or cigarettes. And then fatty food and junk food would have to be regulated too. The next thing you know, the government would force us to brush our teeth and do exercises everyday."
Attacking an opponent by attributing to them an implausible position that is easily defeated when this is not actually the opponent's position.
For example: when many people argue for more democracy in Hong Kong, a typical reply is to say that this is not warranted because it is wrong to think that democracy is the solution to all of Hong Kong's problems, or to say that one should not blindly accept democracy. But those who support democracy never suggest that democracy can solve all problems (e.g. pollution), and they might also agree that blindly accepting something is rarely correct, whether it is democracy or not. Those criticisms attack implausible "strawman" positions and do not address the real arguments for democracy.
Where there is contradicting evidence, only confirming evidence is presented.
Source: Joe Lau and Jonathan Chan, https://philosophy.hku.hk/think/fallacy/list.php
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License.