Topics. Evaluating. arguments. 1 Introduction. PHI 1101, Section I (P. Rusnock) 2 Evaluating Premises. Introduction

Transcription

1 Topics ( and Critical Thinking, Chapter 2) Fall : 1 : 2 Now that we know how to identify and their parts, we move to our next question: how can we tell if an argument is a strong one? That is, how well is the conclusion supported? Two elements to be considered: Are the premises any good? Is the reasoning any good? Sometimes it s easy to tell that a premise is a good one; sometimes obvious that it is no good; and many cases where it s hard or even impossible to tell. Canada shares a border with the USA. It never gets cold in Ottawa. There is life on other planets. : 3 : 4 Similarly with reasoning The quality of premises and reasoning are independent Either Jack or Jill is a doctor. Jack isn t a doctor. So Jill is. Jill is a doctor. So she has red hair. Given a choice between two things, Jill prefers vanilla ice cream to chocolate ice cream and chocolate ice cream to fish tacos. So Jill prefers vanilla ice cream to fish tacos. The two questions are independent: reasoning can be good or bad with true premises but also with false premises. Ottawa is east of Toronto and Toronto is east of Winnipeg. So Ottawa is east of Winnipeg. Ottawa is east of Winnipeg and Montreal is east of Winnipeg. So Ottawa is east of Montreal. Ottawa is east of Moncton and Moncton is east of Halifax. So Ottawa is east of Halifax. Ottawa is east of Moncton and Ottawa is east of Halifax, so Moncton is east of Halifax. So we ask, and answer, the two questions separately. : 5 : 6

2 Truth, rational acceptability Strength of claims In the best cases, our premises are not only true but also known to be true. This is not possible in a great many instances, however, and so we often set our sights on rational acceptability. The roof of this building won t collapse during our class. This plane won t crash if I take it. I won t get food poisoning if I eat this sandwich. Etc. Rational acceptability has degrees, and our language accordingly has expressions that indicate how strongly someone is committed to a claim: It is certain that..., There can be no doubt that... It is virtually/nearly/almost certain that..., It is highly probable/likely that... It is likely/probable that... It is possible that..., It may happen that..., etc. We ask (when reading and writing): is the appropriate degree of acceptance expressed? : 7 : 8 propositions Statistical claims Many propositions state something about one or more things or individuals of a certain kind. The more individuals or things concerned, the wider the scope of the claim. 1 At least one person has walked on the moon. 2 Several people have walked on the moon. 3 Many people have walked on the moon. 4 Most people have walked on the moon. 5 Everyone has walked on the moon. The wider the scope, the stronger the claim. We ask: is the scope of the claim appropriate? Sometimes a more or less precise number is used to indicate the scope of a claim. Three out of four dentists recommend Whizzo toothpaste. Between 46 and 54% of Americans think that the President should be impeached Special cases: 0% (none) and 100% (all): universal propositions : 9 : 10 Universal claims Refutation by counterexample A universal generalization is a proposition stating that all things of a certain kind have (or lack) a certain attribute. All medical doctors have studied anatomy. All crows are black No human being is perfect. (Every human being is imperfect.) No ostrich can fly. (Every ostrich is incapable of flying.) Because they are so strong, universal claims are often hard to prove when they are true and correspondingly easy to disprove, or refute, when false. A single example of an object or individual that is P but not Q is enough to refute a claim of the form All P are Q. Everyone in this class is wearing glasses. no one in this class is wearing glasses. No professors wear glasses, etc. When we consider a universal generalization as a premise, accordingly, we ask: are there any counterexamples? (exercises, p. 30) : 11 : 12

3 Some questions to ask about sources Often, we accept a claim based on someone else s testimony they assure us that something is true, and we take them at their word. If we did not do this, our would be extremely limited. But taking another s word for something can also lead us astray. Is the source reliable? What is his or her track record for telling the truth or providing reliable? Is the source competent and in a position to know the relevant facts? Is there any reason to suspect bias? Do other reliable sources agree? How many? Do any reliable sources disagree? How many? : 13 : 14 Failures: When many sources speak... Sources who lie Sources who spread false beliefs while taking them to be true False confidence Incompetence and ignorance Anosognosia and the Dunning-Kruger effect Bias, intentional and unintentional, conscious and unconscious Are they all reliable? Are they independent? How many speak for, how many against? Is there a reasonable explanation of the disagreement? : 15 : 16 or Authorities, real or merely alleged: some questions to ask Certain people are recognized as experts or authorities in a given field: their opinions are accorded more weight because they are in a better position to know by virtue of their training, experience, etc. Universities are full of them; and we train quite a few of them. Though we can be led astray by experts, I think it fair to say that it is more common today for people to make the mistake of not trusting experts enough. Is there a field of expertise? Is the source recognized as an expert in the relevant area or field? Is there any reason to suspect incompetence, bias, or corruption? Do other experts agree? Is there a reasonable explanation for disagreements among experts in the field? If someone else appeals to experts in an argument, how have they been chosen? (Watch for cherry-picking, the selection of precisely those authorities who support a given position.) : 17 : 18

4 and commonly-held beliefs In most argumentative contexts, the participants can count on some claims being granted without supporting reasons, because everyone there accepts them as true. varies across groups There are also commonly held false beliefs; we do well to be aware of these when entering into discussion with people who hold them. We say that a set of propositions is inconsistent if it is impossible for all of them to be true together. Example: Alice is older than Bob. Bob is older than Carol. Carol is older than Alice. In this, as in many examples, we can recognize inconsistency without knowing whether the individual claims are true or false. A finding of inconsistency tells us that at least one of the claims is false, but may not indicate which one. However, an argument that relies on all the premises in an inconsistent set is in trouble. This is why consistency matters for acceptability. : 19 : 20 : First Remarks reliable principles/patterns Most often, reasoning follows repeatable patterns or forms: we can also say that it is guided by general principles. Accordingly, evaluating reasoning often boils down to determining whether these forms or principles are reliable. One of three things is true: A, B or C. A is not true and B is not true. So C must be true. There are two things of a certain kind and another two things of that kind. So altogether there are four. If P, then Q. Not Q, So not P. : 21 : 22 And some unreliable ones Showing that a principle is not trustworthy One of three things is true: A, B, or C. A is not true. So C is true. Three or four things of a certain kind have a certain property. So all things of that kind do. If P, then Q. Not P, So not Q. We can show that a principle or form of inference used in an argument is not trustworthy by giving an example of an argument in which the same principle is used, but produces obvious unacceptable results. Consider the argument: If it rains, we won t go for a bike ride. It won t rain, however. So we will go for a bike ride. If we just think about rain and bicycles, etc., it can be hard to see that the reasoning is poor. But once we identify the form of inference as: If P, then not Q. Not P. So Q we can show this form to be unreliable by pointing to an instance like this one: If we re in Winnipeg right now, then we re not in Nova Scotia. We re not in Winnipeg. So we re in Nova Scotia. : 23 : 24

5 Another example Example: Smoking can cause lung cancer. Mary died from lung cancer. So Mary must have been a smoker. Form/Principle: X can cause Y. Y occurred, So X occurred. Parallel argument: Falling off the top of the Empire State Building can cause death. Napoleon is dead. So Napoleon fell off the top of the Empire State Building. : 25