Chapter Four: The Goal and Structure of Reasoning
Reasoning is a goal-driven activity. The goal of reasoning is to provide rational support for claims, statements, or actions. Take, for example, the following medical reasoning:
Since (A) the chemotherapy has no chance of working in an advanced case of leukemia like Bill’s, (B) we should not even attempt it.
Statement A provides a reason for statement B. But in reasoning we seek more than just to give reasons. We want to do that well, by giving reasons that actually succeed in supporting the claim they seek to support. Determining whether reasoning is successful requires an understanding of the basic structure of reasoning. In what follows, we will present some of the basic structure of reasoning.
Because reasoning is the activity of providing support for some statement or action, all reasoning requires at least two items: the support-giver and the support-receiver. The support-giver is called a reason, evidence, or a premise. The support-receiver is called a conclusion. The combination of both is called an argument. More complicated arguments can contain more than one premise and even more than one conclusion.
How do you know if an author is presenting reasoning? How can you, as an author, make it clear to your reader that you are providing rational support for a statement? Good writers of dialectical essays will make their reasoning explicit. To make reasoning explicit, it is crucial to use standard premise and conclusion indicators.
Premise indicators: since, because, for, reason for. Note that a sentence leading up to, and supporting, a conclusion is a premise even if it does not have an explicit premise indicator. We know it is a premise from its relation to an explicit conclusion.
Conclusion indicators: so, therefore, thus, consequently, it follows that
There are innumerable ways to indicate that one presents a reason or a conclusion, and this list is not exhaustive. A good writer of arguments will find some way to make his or her reasoning explicit. The main mistake to avoid is putting the burden on the reader of figuring out the reasoning of your essay. One common way to fail to make reasoning explicit is by presentation of a list of sentences, with no premise or conclusion indicators to show which sentences are premises and which are conclusions. Consider the following:
(A) John is a good basketball player. (B) He runs well. (C) He shoots well.
In this list, which are the premises? Which are the conclusions? Is there only one? There is no way to know. A might be a premise and B and C both conclusions, or B and C might both be premises and C a conclusion. The context of an essay might make it clear to the reader, but the best way to make it clear is to use the indicators. If I am attempting to support A with the evidence from B and C, I indicate so. Here are some examples:
-Since John runs well and shoots well, he is a good basketball player.
-John is a good basketball player, for he runs well and shoots well.
-John runs well and shoots well. Therefore, he is a good basketball player.
It is often helpful to diagram any argument you are analyzing/presenting in order to clarify what the argument is. We could diagram the example argument presented above as follows:
Premise: The chemotherapy has no chance of working in an advanced case of leukemia like Bill’s.
Conclusion: We should not even attempt it.
In presenting such an argument, the presenter is committed to claiming that the premise is true, or at least probably true, and that the conclusion follows logically from the premise or is very likely to be true because of the truth of the premise. Good reasoning seeks to satisfy these conditions.
Testing and Criticizing Reasoning
These conditions of good reasoning also give rise to a strategy for testing and criticizing reasoning. We can test someone’s reasoning by determining the premise(s) and the conclusion(s), determining whether the premises are true or probably true, and by determining whether the conclusion follows logically from the premise or is very likely to be true if the premise is true.
We can criticize someone’s reasoning in the following two ways: (1) If it turns out that a premise is false or probably false, we can reject the argument as weak. A false premise does not give us a genuine reason for accepting its conclusion. (2) Should the conclusion not be likely to be true if the premise is true, then the premise provides no real support for the conclusion.
Please note the following logical terminology:
- Should it turn out that a premise in an argument is false or probably false, we can reject the argument as unsound. An argument is sound if and only if all of its premises are true and it is valid.
- Should the conclusion not follow logically from the premises, then the premises provide no real support for the conclusion and the argument is invalid. An argument is valid if and only if the conclusion must be true if the premises are true.
Consider the argument diagrammed above regarding leukemia. Suppose that a medical researcher points out that an article published last month in the New England Journal of Medicine showed a 50% success rate of treating advanced leukemia with a new chemical. That evidence would show that premise A is false. On this basis, we could criticize the argument as weak.
Or suppose that even though the premise is really true, another article shows that even in cases of advanced stages of cancer, patients fare much better while being treated due to the hope and comfort they feel at getting treatment, even though the success rate is negligible. The improved end of life situation might provide a reason for treating them even though physiologically the treatment does not work. That study might show that even though a patient has no chance of recovering through the chemotherapy, it ought to be administered anyway if he or she wants it. In this case, the argument is weak because the conclusion—that we ought not to attempt the treatment—is not likely to be true because the premise is true.
Fallacies: Typical Failures of Reasoning
Logicians describe a set of standard ways in which a premise can fail to support a conclusion. It is helpful to be aware of some of these standard fallacies in order to help you spot fallacies in other people’s reasoning and to avoid them in your own.
Fallacies of Relevance
An ad hominem argument is literally an argument aimed at the man who makes a claim, not at the claim itself. Such an argument attempts to show that a statement is false, but it fails to provide a reason for thinking a claim is false. The fact that someone makes a claim does not by itself make it likely that the claim is false. Here are some examples:
-The government statement that we will get out of Bosnia in a year is obviously false since Clinton made it.
-The claim that God is dead is false since Nietzsche said it and he went crazy.
Neither of these premises gives direct support for their conclusions.
An argument ad populum, which is literally an appeal to the views of the masses, is another standard fallacy of relevance. The fact that most people believe something is not itself evidence that what they believe is true. So, the following arguments would contain fallacies:
-Since everyone believes that life has meaning, it must have meaning.
-The skepticism view that we do not know that there is a world of physical objects must be false, for everyone not engaged in philosophy believes that he/she knows that the external world of physical objects exists.
Since it is possible that what everyone believes is false, this sort of argument is unsuccessful.
An argumentative appeal to tradition to support a claim or action presents another fallacy of relevance. The fact that tradition holds some claim to be true or some action to be good does not by itself show that the claim is true or the action good. In the tradition of Roman Catholicism, homosexual acts are deemed immoral, but the fact that the tradition holds this view, does not make it correct. So the following arguments contain fallacies:
-Abortion must be wrong since the Roman Catholic tradition takes it to be wrong.
-It must be wrong to make it difficult to get an abortion, for the liberal democratic tradition has for many years steadfastly opposed laws designed to limit access to abortions.
Both of these arguments appeal to tradition to support a moral view, but both fail by introducing considerations not directly relevant to the truth of the conclusions they aim to support.
An argumentative appeal to authority attempts to support a claim or action by indicating that some authority believes the claim or embraces the action. But the fact that an authority accepts a claim or embraces an action does not directly show that the claim is true or the action good.
-I must have wanted to have sex with my mother because Freud says all males want to.
-Dr. Milton is an authority on Socrates, so my view of Socrates, which is different from his, must be false.
A red herring is a mode of reasoning in which an irrelevant side-issue gets introduced to divert attention from the central issue. The image refers to a practice of attempting to cause hunting dogs from an opposing team to get diverted from the scent of the fox by dragging a red herring, a smelly fish, across the hunt path. The dogs would get diverted onto the path of the smellier scent. A writer introduces a red herring in the following exchange between John and Bill:
John: The Russians were correct to point out that NATO’s bombing of Yugoslavia violates international law and so is as criminal as the acts that the bombing was supposed to counter.
Bill: Those Russians haven’t been able to get their act together since the fall of the Soviet Union. Drinking vodka in Russia is up in the past five years.
Bill's points are clearly red herrings. To respond to John's point, Bill needs to argue that the bombing does not contravene international law, or if it does, there is some higher moral law that makes violating international law okay in this case. Instead, Bill introduces true - but irrelevant - side issues designed to divert attention from the main point.
Fallacies of Relevant, but Insufficient Evidence
The fallacy of hasty generalization is committed when someone reasons from a single case, or even several cases, to a general conclusion insufficiently supported by the small array of examples given. For example, consider this following reasoning:
-Tom, who is in a fraternity, gets drunk every night, so all fraternity members get drunk every night.
This sort of reasoning is not, however, improved by adding a few more instances. Consider a modified instance of this fallacy:
Tom, Dick, and Harry, who are in a fraternity, get drunk every night, so all fraternity members get drunk every night.
Another type of relevant but insufficient evidence arises when one constructs in an argument a false dilemma. A false dilemma is proposed when two alternatives are assumed. One is ruled out, and the other consequently endorsed when there are more alternatives than just two. Consider the following example:
-Either John is a liar or he is honest. He is obviously not a liar, so he must be honest.
The problem with this reasoning is that since there are other ways to be dishonest, one cannot conclude that John is honest, just by ruling out that he is a liar. For example, one can be a deceiver without being a liar. In some of Clinton's public announcements about his affair with Monica, he made statements that were intentionally misleading and so deceptive, but did not, at least in some of those statements directly lie.
Reasoning can also fail by treating a complex question as something simple. For example, if one jumps from the medical fact that a treatment has little chance of working to the conclusion that we ought not to pursue it, we are introducing a true premise that does not support its conclusion because there are additional factors involved in the decision about how best to proceed. A patient may want to pursue long-shot treatments in particular cases to try to stay alive long enough to see her grandchild. Treatment decisions are complex questions, and a simple, single premise reasoning about how best to proceed is, therefore, fallacious.
Another fallacious form of reasoning that introduces some relevant evidence insufficient to support its conclusion is the fallacy Post Hoc, ergo Propter Hoc: literally after this and because of this. Suppose you have a conversation with your History professor. Afterwards you receive a lower than usual grade on your next essay. You might reason that your lower grade came about because of something you said in that conversation. But just because the low grade arises after the conversation does not show that it arises because of the conversation. More evidence is needed to show that.
Arguments employing analogies are suggestive and important but often too weak and questionable to support the evidentiary burden they bear: such fallacies are called questionable analogies. There is no magic formula for determining whether an analogy is questionable. Typically one looks for differences that invalidate the appeal which makes the analogy attractive. Consider the following example:
-A doctor tells a patient which treatment he is going to pursue without seeking permission of the patient. He treats the patient like a child. Therefore, since the patient is a competent adult, he treats the patient inappropriately.
Is it fair to see the treatment of the patient and child as analogous? There are some similarities, but are there crucial differences? For example, the doctor has no control of his patient. So the patient could refuse the treatment and seek medical treatment elsewhere. Can the child seek parental guidance elsewhere? Because of these differences, the analogy appears weak.
A person commits a fallacy of equivocation when he/she uses a word in one sense in a premise to prove a conclusion using the word in a different sense. Consider the following argument:
-Sally's God is fashion design, so God exists and should be worshipped according to the dictates of biblical scripture.
‘God' in the first sense means "passion" or "ultimate concern." But the existence of God in the first sense shows nothing about the existence of God in the second sense.
Fallacies of Dialectical Bad Faith
In dialectic, as we have described it, defense of a thesis is indirect and so involves defending it from some line of argument, either a possible criticism or a response to a possible criticism that would be in conflict with it. The defense is no stronger than the challenge made of the thesis and the response given to that challenge. Good faith efforts at defending a thesis require good faith efforts at presenting substantial challenges and responses. Weak challenges and responses, or failure to take a challenge seriously, undermines the effectiveness of the argument. In what follows we describe two such fallacies.
A person begs the question when he or she asserts as true what has been called into question by some challenge. Consider the structure of a dialectical essay. Step Three presents a possible criticism of some claim in Step Two. One can either ignore the challenge or respond to it. Ignoring the challenge, even in subtle ways, can result in begging the question - asserting as correct what has been called into question. The way to avoid begging the question is to respond in detail to the possible criticism, by refuting some premise in it. Consider the following exchange:
John: There is a God.
Bill: There cannot be a God since (a) there are evil human events and (b) no God would allow such events.
John: But since there is a God, your reasoning must be faulty.
The problem with this reasoning is that John does not even attempt a critique either of Bill's premises or of his inference from those premises. Instead John continues to assert his claim, which has been called into question by Bill's criticism. Asserting this claim without first attempting to refute the premises or the inference in Bill's reasoning is begging the question. While often parading as reasoning, (note the premise indicator, ‘since') begging the question is a refusal to construct a reasoned response to a challenge. In that way this fallacy undercuts the very project of dialectical reasoning.
An argument presented against one side of an issue, which is weaker than other possible challenges, is said to be a straw man argument. The image is of a boxer punching a defenseless straw man, not a real one who can fight back. It is crucial to the dialectical enterprise to present a possible criticism or response that is as compelling as possible - even if you are going to argue against it. It is clear why strong arguments from the side you oppose are important. If I can show that my thesis can be defended against fairly compelling objections, then I have shown something of substance. If, instead, I show that my thesis can be defended against relatively weak objections, I have not shown much about the strength of my thesis. Just as I show no real skill as a boxer to knock down a straw man, I show no real merit for my thesis to knock down a straw man argument in its defense. There is no easy method for determining whether an argument is a straw man. That tag depends on being able to think of stronger possible criticism.
One easy way to generate a straw man argument is to construct a possible criticism that obviously contains any the fallacies that we have described in this section. Then expose the fallacy. If there are other more compelling criticisms you could have investigated, then you have knocked down a straw man.