Argumentum ad Vericundiam
An example of the second form of appeal to authority would be to say something like this, "Well, my logic professor said that fallacies are errors in reasoning, therefore, fallacies are errors in reasoning". It is admitted by almost everyone that all human beings are fallible. We make mistakes. Even the most highly respected individual in her/his field may be wrong about some information. It may be unlikely, or even highly improbable, that so-and-so is wrong about a certain item of "knowledge". However, it is not impossible. For example, in logic, one of the most basic forms of reasoning is called modus ponens. Logic students learn this in the first course they take. Imagine you're taking a logic course from a truly eminent logician. Imagine that your teacher is hard of hearing, and has been since the first logic course he/she took in college. Imagine further that when modus ponens was being explained in your teacher's first logic course, he/she misheard the second word and has pronounced modus ponens as "modus konens". So, when you get out of your logic course, you will have learned to pronounce "ponens" with a "k" rather than a "p", and when you are corrected by one of your friends, you might well say something like the following, "Well, I should know how to pronounce modus konens because I took the logic course with Professor X, who ought to know what he/she is talking about since he/she is a highly respected logician. Who better to know these things than the logicians themselves?" The trouble, of course, is that your professor had it wrong. The point here is that one can be in error on even the simplest matter.
An example of (3) would be the following. Suppose you're visiting a friend you haven't seen for some time. You get to your friend's house and you are called upon to say the blessing at the table before dinner. You explain that you're not a religious person and would prefer not to have to say any words constituting a prayer. Your friend explains that it is a long-standing tradition in their family (a custom reaching back at least three centuries) to have the guest offer the blessing at table and that therefore you have to do it. Your friend is appealing to custom in the argument he/she gives. Admittedly this example of custom is purely local, as opposed to a more global example (involving perhaps countries and entire peoples). Nonetheless, it is an example of ad Vericundiam and the argument should be rejected from the logical point of view. This is not to say that one would not, or even should not, act in accord with the prevailing custom in this situation; it is only to say that a fallacy of reasoning has been committed. It would not be difficult to construct very hard cases of appeal to custom where the stakes are much greater (for some) than giving a blessing at table.
Argumentum ad Misericordiam
As with the appeal to the people, this fallacy appeals to one of our emotions, namely, pity. If someone attempts to persuade you to accept some argument by trying to evoke pity in you, then that person has committed the fallacy. For example, we might be able to feel sorry for the spouse of an individual who is about to be jailed for having committed some crime. The sorrow we feel may be quite legitimate, as we see that the person is likely to become destitute with the support of the jailed spouse. However, if someone tried to persuade us that the person shouldn't be jailed because the other would become destitute, then our emotion of pity is being appealed to. A good reason not to jail the person is if he/she were found innocent of the crime, or if there were relevant extenuating circumstances we could appeal to which showed that he/she acted in the best possible way which led to the crime being committed (or something like that).
As with all arguments, appealing to an emotion may win the day in terms of getting an argument to be accepted, but it is not "rational" argumentation from the point of view of logic. An example of ad Misericordiam is:
Note that in the above argument, pity is not evoked for Peterman himself, but rather for some other individuals, his sons and daughters. This does not matter. What matters is that pity is appealed to in order to get a certain conclusion accepted, that is, the conclusion that Peterman should go free. It is useful to be able to pick out what might be a good reason for letting Peterman go; say, that a credible witness is brought forth who says that Peterman was 100 miles away at the time of crime. This sort of witness would be speaking to the evidence of Peterman's guilt or innocence rather than the consequences of Peterman's imprisonment (which has nothing whatever to do with his guilt or innocence).
Argumentum ad Baculum
The next fallacy treated has to do with threatening someone with some harm, either physical or psychological, in order to get her/him to accept some conclusion. This is something of a tricky fallacy. On the one hand, we know perfectly well that if someone threatens us with harm that we think is significant, then we'll do just about anything (just about). On the other hand, just because one does a thing doesn't mean that one has accepted a certain conclusion. It is on this basis that some philosophers do not accept ad Baculum as a fallacy. However, decisive reasoning on this has not been forthcoming and every logic textbook I'm aware of, which treats the informal fallacies, includes this fallacy. Hence, we will do so.
The following is a typical example of someone threatening harm:
Two things are demanded here by the supervisor: 1) No leaks and 2) Support for the President in the form of praise and understanding. If Dave (who is one of the office workers) is scared enough about losing his job, then he will not leak and he will offer warm praise. There is a sense in which the supervisor has convinced Dave to accept certain conclusions here. The acceptance comes in the form of Dave's actions (praising) and non-actions (not leaking).
Now, if the supervisor had given the workers different reasons for no leaking and for praising the President's actions, then perhaps an acceptable argument might be shown. For example, one argument that might work is the argument showing how the President's actions help teachers at the primary levels (K-6) to obtain vital educational tools. Add to this that the President's actions allow prisons to be more productive and thereby lowering the taxes we have to pay for the upkeep of the inmates. This would all need to be laid out clearly for us to judge, but at least the arguments and reasons given for not leaking and for praising the President's actions would not rely on a threat.
One question I have about this fallacy is this: If I do something on the basis of a threat, then even though I know that it is not "logical" to do it, isn't it at least "practical"? As a student, when the university would send me my registration forms, they would include a note saying something like, "If you do not pay the fees by 23 April, you will be barred from registration and your financial aid will be withheld." I took this language to be threatening, but even so, I made sure to pay by April 23rd. Have I accepted the fallacy? That is, have I willingly accepted a conclusion I know to be based on ad Baculum reasoning? I believe I have. Something for us to think about: Does this example show that it is sometimes better to be illogical than logical in our actions? I'm thinking that if it is just actions we're talking about, OK, but if it comes to thinking, then I'd have to answer "NO". I think it is never rational to think irrationally. So, if I commit ad Baculum, then it's all right as long as I know I'm doing it. I'm afraid this doesn't satisfy me either.
Argumentum ad Hominem
Of all the fallacies I know, this one seems to me the most obvious. Moral: One cannot refute an argument by discrediting the arguer. But that is just what ad Hominem attempts to do. It is an attack on a person -- better known as a "personal attack". There are three kinds of ad Hominems: Abusive, Circumstantial, and Tu quoque. You will be required to know these.
If someone attacks another person with abusive language to draw attention away from an argument given by the person being attacked, it is a case of ad Hominem abusive. An excellent example of this sort of ploy is when an attorney attempts to discredit a witness. Somehow, a speaker is often identified with what the speaker speaks. In logic, however, this won't work, because we go on the principle that an argument needs to stand on it's own and be evaluated on its merits, no matter who the arguer is. So, if Stephen Hawking (the physicist) argues that gravity in a Black Hole is so strong that even light cannot escape, then we would need to examine the premises of his argument to determine it's acceptability. But now, if someone were to impune Hawking himself, say, he's charged with forging his physics degree, even if the charge was correct, that would have absolutely no bearing on whether his Black Hole argument was acceptable. To abuse him by calling him a cheating, lying, borish forger has no effect on his argument. The second type of ad Hominem is Circumstantial. If we were to argue that Hawking makes the argument he does because he has something to gain if we accept his argument, then we commit ad Hominem Circumstantial. See how our making this charge attempts to take our attention away from the argument itself. But again, it's Hawking's that is important, and the argument has to stand alone, whether he gains or loses or is neutral. If the argument is acceptable, well and good --- if he gains by our acceptance of it, again, well and good.
The third type of ad Hominem is Tu quoque. This fallacy is committed when a person attacks someone and the person being attacked follows up with an attack on the original attacker. So, Penrose charges Hawking with forging his degree; then Hawking attacks Penrose by charging that he falsified his most recent labratory findings on light wave functions. The problem here is that Hawking fails to answer Penrose's charges, but rather attacks Penrose. For all we know, Penrose's charges may be correct; Hawking does nothing to refute them and certainly a personal attack on Penrose is not an adequate answer to the charges. (On the other hand, perhaps Penrose's charges are not correct. We'd have to do some digging to find out.)
The important point to remember: All ad Hominems are personal attacks and are not directed to any argument that has been made by the person being attacked. Again, Hawking and Hawking's argument are two different entities and the argument needs to stand on its own, on its own merits.
Argumentum ad Ignorantiam
This fallacy is based in the ignorance of the arguer. If we argue that Hawking's argument about Black Holes has not been proven unsound, therefore it must be sound, we've committed Ad Ignorantiam. It may be the case that the argument has not been proven unsound. What, however, does this say about the soundness of the argument? Nothing, so far as I can tell. Hence, it would be illegitimate to assert that it is sound. For soundness, we need to discover whether it is valid and whether it has true premises. Our ignorance of its unsoundness cannot inform us of its soundness.
One of the best examples I've heard of this fallacy being committed was during the McCarthy hearings in the middle of the 20th century. Recall that McCarthy (a U.S. Senator) had been accusing many people in and out of government of having communist connections. McCarthy had a person's file before him and announced before the inquiring body that the file contained no evidence that the person was not a communist. The implication drawn was that the person was a communist. Blatant Ad Ignorantiam.
Accident
This fallacy is all about applying a general rule incorrectly, or improperly, i.e., when it simply doesn't fit or in a situation where it is not applicable. For example, take the general rule that most parents try to get their children to live by: KEEP YOUR ROOM CLEAN. This is a "general" rule, not an absolute rule. If it were an absolute rule, then in the middle of a housefire, it may be thought legitimate to get the child to go in and clean her/his room. So, a kind of silly argument may be constructed by the parent:
Contrast this now with what many would say constitute absolute rules, such as The Ten Commandments. These absolute rules are never to be broken, no matter what circumstance develops. This is to say that the fallacy Accident could never be broken with regard to any of The Ten Commandments, since there is never an occasion to apply the rules incorrectly.
Hasty Generalization
This fallacy is commonly said to be the reverse of accident. In a way it is, but it also has a number of variants which may distinguish it from being such. The standard reading of Hasty Generalization is that when one reasons to a general rule from a limited number of cases, then Hasty Generalization has occurred. When you sample just a few of a group, many times your sample is considered inadequate, not large enough, although size is only one variable that must be considered in such cases. So, if you draw the conclusion that all psychology teachers are insane simply because the two you took for classes were insane, then you've committed Hasty Generalization. Or even the three you took. Or the four or five.
It starts to get tricky after five. I'm kidding about that, but somewhere along the line it does get tricky. There are big questions about how many X's one has to observe with chaacteristic Y before one can legitimately conclude that All X's have Y. If one wishes to be literal, and unforgiving, then one might say that ALL X's must to be observed to have Y before you can make the generalization that All X's have Y. Typically, however, in ordinary life situations, in the science laboratory, and in our thinking, we don't require the observance of all of any kind to draw a general conclusion. How many protesters do you have to spray with tear gas to know that all protesters (who are not wearing masks or some other gas-defeating device) will succumb to the effects of tear gas? Not very many. Indeed, one, and not more than one. We have good reason to believe that the chemicals in tear gas will cause certain reactions in human protesters and therefore Hasty Generalization is thought not to have been committed in an example like this. This example is also instructive for showing that sometimes the same that one takes of a group does not have to be large to be "fair".
Consider, though, another "general rule" manner of expression:
False Cause
The Fallacy of False Cause is committed when one identifies as a cause of some event* an event that is not the cause of the event* referred to. Imagine someone saying that the cheerleaders should wear the blue ribbons today because the team always wins when the cheerleaders wear the blue ribbons. No one would seriously consider that wearing or not wearing blue ribbons would make any causal difference at all whether the team won or lost. Hence, False Cause.
Equivocation
The Fallacy of Equivocation occurs when an arguer uses the same word or phrase in two or more senses (that is, with two or more different meanings) in one context. So, if someone should argue that medical miracles happen every day and therefore God must care for the people who are cured by surgeons, we would naturally think that an equivocation had taken place with respect to the word "miracle". In one way, 'miracle' is used as a figure of speech expressing our wonderment over the strides medical science has taken in the last 50 years. For example, someone might say, "its a miracle that a medical team can remove all the blood from a patient for a time during an operation and the patient can still live. The other sense of 'miracle' here is the sort spoken of in the Bible, where Jesus is said to have, for example, brought a man back to life who had been dead for 3-4 days. This is the "divine intervention" sense of miracle. So, in the initial example example, 'miracle' is used in two different senses. So, Equivocation.
Petitio Principii (Begging the Question)
When one begs the question, one basically assumes to be true the very conclusion one is trying to establish. By 'assume' here, I mean that one uses the conclusion as one of the premises. Sometimes this is difficult to detect because the conclusion used as a premise is rarely couched in exactly the same terms. Consider the following example:
This is an interesting fallacy for us, because it speaks to our frame of mind and our actions in the everyday world when we're not thinking about Philosophy. Notice how we go through our lives relying on things being as they were yesterday, or an hour or a minute ago. My friends Nuu and Ed, with whom I play racquetball, are both right-handed. If tomorrow they were to come in the club and begin to play left-handed (and play as well as they play right-handed), I'd have to say that the future was definitely not like the past. Perhaps I'd question my own sanity, or my perceptions, or my memory, or whatever. And the reason I'd be so surprized is that I'm very convinced, in my normal everyday life, that the future will be like the past. But, that is to beg the question, without proof.
Another way to beg the question is to state, as fact, something which is controversial. Example:
Limited alternative
This is an easy fallacy to understand. It is committed when anyone makes an argument in which choices are given when there are clearly more choices open than are given. You go to your advisor, say, in biology, and indicate that you're thinking of changing your major. Your advisor says, "Well how about chemistry, or physics, or geology?" Admittedly this is not an argument, but you can see the point. You reply, "Well I was thinking about journalism." Clearly your advisor hadn't given you all the options. Hence, you were given a limited set of alternatives to choose from. Hence, the Limited Alternative fallacy.
On the other hand, sometimes people give arguments which contain all alternatives available. In this sort of case, the fallacy is not committed. Example:
Deontic Fallacy
An 'ought' cannot logically be derived from an 'is'. That's the standard line in moral philosophy, or ethics. That is, if each of the premises in an argument is a descriptive statement (simply describing facts about the world) and if the conclusion is a prescriptive statement (saying how one ought, or should, act) then the Deontic Fallacy has been committed. Here is an example:
I've always thought that this sort of argument has a hidden premise in it (and why the arguers always seem to fail to state it is a mystery to me). It is this:
How to Avoid Fallacious Reasoning
Realistically, there is probably no chance that we will avoid making fallacious arguments or accepting fallacious reasoning all the time. This is partly because it requires a lot of concentration sometimes to detect bad arguments. First, one has to know the fallacies; second one has to have the fallacies in mind when one is assessing an argument; third, one has to have a complete understanding of what is being said in each of the sentences in the argument. This last is a big problem sometimes, because natural languages (like English, French, German...) are rich in nuance and subtle in expression. Many times language is used in it's finest forms to "make the worse argument seem the better".
I suppose the best advice one can get on this is to be careful in assessing an argument. Know its form, know its content.