How To Do Just About Everything
How To Do Just About Everything
Find this article at: http://www.ehow.com/how-to_4845353_avoid-common-logical-errors.html
We've all probably made easily avoidable errors of logic; therefore, everyone will want to learn how to avoid doing so in the future. There is, of course, a logical problem with this last sentence: It assumes a premise that probably isn't true--specifically, that we all care about not making logical errors. But if you do, and you want to avoid such traps, you need to hew to some basic strictures.
"Non sequitur" is Latin for "it does not follow." When an argument is a non sequitur, its conclusion does not follow from its premise(s). This term is often used to refer to particular types of arguments that clearly do not follow from their premises and never could.
For example, any argument that takes the following form is a non sequitur:
Step 1: If I am a goat, I am a hoofed creature. Step 2: I am a hoofed creature. Step 3: Therefore, I am a goat.
It is clear that this argument does not follow. Even if the premises and conclusion were all true, the conclusion is not a necessary consequence of the premises. This sort of non sequitur is also called "affirming the consequent."
Another common non sequitur is this:
Step 1: If I am in Rome, I am in Italy. Step 2: I am not in Rome. Step 3: Therefore, I am not in Italy.
The speaker could be in many other places in Italy. This sort of non sequitur is called "denying the antecedent."
"Begging the question" means assuming something that is material to the point you're trying to make. You beg the question when the truth of your conclusion is assumed by the premises that purport to prove it.
For example: "We should not adopt gun control, because unacceptable impositions on our liberty cannot be tolerated." If the premise of this argument--i.e., that "unacceptable impositions on our liberty cannot be tolerated"--is construed to be relevant to the conclusion about gun control, then the premise must assume that gun control is an unacceptable imposition. If that's the case, then of course we shouldn't adopt it; but that argument begs the question of whether gun control is a good thing.
The identity of a speaker or writer is irrelevant to the validity of his argument. All that matters is whether the premises lead to the conclusion. There are only two ways to criticize an argument: (1) One or more of the premises is or are false; or (2) The premises do not lead to the conclusion.
You may have heard people discuss, with disfavor, "ad hominem" arguments. Ad hominem is Latin for "to the man." An ad hominem argument tries to attack characteristics of a person that are irrelevant to the issue. When you evaluate an argument with reference to the speaker, you are making an ad hominem response.
People believe that the identity of the speaker is important because they think they must be watchful for bias or hypocrisy. This is incorrect in the context of evaluating arguments. It is true that bias can cause people to be dishonest, and you should be concerned about bias when you evaluate whether or not to believe a person's reports of fact. However, this is still irrelevant to an argument, because the premises are either true or untrue on their own merits. If you decide not to believe a biased speaker, it essentially means that you refuse to evaluate an argument that contains the questionable statement as a premise. It is no criticism of the argument that you are skeptical of the reliability of its premises. In order to refute the argument, you need to prove that one of the premises is false.
Hypocrisy is even less relevant to arguments. For example:
Speaker 1: It is wrong to steal.
Speaker 2: But you stole something yesterday!
Speaker 2's comment is not responsive to the remark of Speaker 1. Speaker 1 was making a statement about what is wrong, and Speaker 2 changed the subject by talking about Speaker 1's behavior. It is the same when someone presents you with an argument and you begin to discuss his or her identity. You are no longer discussing the argument; you have changed the subject.
Just look at arguments based on their merits. If you don't trust a source that is reporting as fact one of the premises, it is okay to say that you don't think you can evaluate the argument until you have verified one of the premises. What is not okay is this sort of thing:
Speaker A: Marijuana should be decriminalized because the criminal law should only forbid citizens from committing acts which harm persons or the property of others.
Speaker B: You're just saying that because you like to smoke it so much.
What is relevant to an evaluation of A's argument is whether or not his single premise, about what the criminal law should forbid, is true. Whether or not he likes to smoke the stuff doesn't enter into it. Always ignore the identity of the source when you evaluate an argument.
Anecdotal evidence proves little. If we assume the anecdotal evidence we are hearing is true, we can use it to disprove a universal claim. For example, if someone claims that all midgets are blind, you can disprove the claim by reporting that you have a cousin who is a sighted midget. In most contexts, however, anecdotal evidence is worse than useless, because it is misleading. If there are 50 people in a room, and 10 of them have stories to share about the warmth and compassion of Texans, should we take their stories as evidence of the proposition that most Texans are warm and compassionate? Of course we shouldn't: We haven't heard about even a tiny fraction of Texans. We've only heard about a portion of the behavior of Texans met by 10 people.
"Post hoc, ergo propter hoc" is Latin for "after this, therefore because of this." It refers to the logical fallacy of assuming that, because one event follows another, it was also caused by it. For example: "I thought of my brother, and 2 seconds later, he phoned me. He must have felt my aura and responded to me." Or: "The rooster crowed and the sun rose. Therefore, the rooster's sound must have caused the sun to rise."
Sometimes an event happens after another event coincidentally, and the two have nothing to do with each other. Mere proximity in time should not be taken as sufficient proof of a cause-and-effect relationship.
At other times, a number of partial causes lead to an effect, but one is selected as the sole cause. This is called "oversimplification of the cause," and it is also flawed thinking. We can't establish what is actually the case solely by referring to when things happen.
Arguing by analogy can be done in one bold, sweeping statement. For example: "Just as it was wrong to deny women the vote, it is also wrong to deny the vote to children." It can also be a lengthy, involved process, in which two speakers agree that a rule or predicate applies to one state of affairs, then argue over whether another state of affairs is enough like the first state of affairs for the same rule or predicate to apply to it.
This is a waste of time, because analogies don't prove anything. An analogy can be useful to illustrate a point or explain something to someone who is unfamiliar with a concept. For example: "You are familiar with broccoli, right? Well cauliflower is like broccoli except that it is white" Analogies can also make an argument accessible or understandable to someone who is otherwise unwilling to be open-minded about it. For example: "Those who are unwilling to accept the possibility that animals deserve moral consideration should consider that much the same thing used to be said about racial minorities."
This example, however, is not an argument. It is an attempt to get people who are simply closed-minded to look at the argument (the one the analogy is trying to support) more neutrally. Logically speaking, trying to argue about one thing by talking about something else is pointless. If you think that the line of argument that led you to believe something about X might also apply to Y, don't talk about X. Apply the reasoning to Y and see if it works.