II. Reconstructing Arguments- Preliminaries

Everything that's been said about evaluation applies most clearly to arguments that are formulated in a clear and revealing way. But little that you read is actually like that. Mostly the arguments are buried in the midst of a lot of prose, with elements unstated, connections poorly spelled out, and loads of extraneous information included. To apply our terms of evaluation usefully, we must reconstruct the argument, or put the argument into standard form.

A. The Principle of Charity

(PC)

When reconstructing an argument, always try to formulate a reconstruction that is well formed, has reasonable premises, and is undefeated. In other words, make the argument as strong as possible.

Calling this a principle of “charity” may give the wrong impression. We accept (PC) mainly because it's a waste of time to think about inferior arguments. We aren't being nice to opponents. The goal of argument analysis is to get an answer to whatever question it is that the argument is about. It's to help you decide what to think, not to see who you can refute.

B. Some Guidelines for Reconstructing Arguments

●          Make sure there really is an argument there. See text, Section II. You can’t tell whether there’s an argument by the topic. You can write about anything w/out giving an argument. A passionate statement of how much someone cares about something is not an argument about that issue. There’s an argument only when reasons are presented.

●          Use the guidelines described in the book to identify premises and conclusions. See text, Section III. Usually, it's best to start with the conclusion.

●          If the reasons seem to be the sort that necessitate the conclusion, make the reconstruction valid rather than cogent. In other words, if the only things that could make the argument fail would be things that falsify a premise, then make it valid. If other things could undermine it, make it cogent.

●          Remember that conclusions and premises can be implicit. This is one of the most troublesome aspects of reconstruction. Describe two main goals: interpreting the text, getting the best argument. The text describes three main principles that govern the addition of implicit premises - faithfulness, charity, generalization. Probably (PG) is the hardest to understand and implement.

●          The principles can conflict. I think it's more interesting to get the best argument you can out of what someone else said rather than figure out what she "really intended." Maybe that's just my taste.

●          You don't always have to come to one final interpretation. You can say that there are two (or more) possible interpretations, and go on to evaluate each.

●          When one of your reconstructions is ill-formed or the wording is confused, it's usually your fault, not the fault of the person whose argument you are reconstructing. It’s your job to get the argument into good shape.

III. Some Examples

1) pp. 120-1: #1,2

In exercise (1), only (e) clearly contains an argument. (a) and (d) may hint at arguments. There is no real argument in the other passages. In exercise (2), (a)-(c) are purely descriptive, (d) may have an argument about what Savino believes, and (e) suggests an argument for the conclusion that televising executions won’t deter potential murderers.

2) p. 128, 2d.

This one is very easy. The striking thing about it is that there is no indicator word, yet we all know what the conclusion is. How do we do this? One idea: using charity, we immediately make it a valid argument for “you will like nectarines” rather than an invalid argument for “you like peaches”. Another idea: the “will” in the nectarines sentence is actually an indicator word in this context. You might think that order helps, but the conclusion could come last as well as first. Try inserting indicator words and see if it changes the meaning.

3) Consider:

Suppose the final exam is near and a student is worried about how he’ll do in the course. The teacher says, “Well, things haven’t gone well at all so far, but if you pass the final exam, you’ll pass the course.” The student then takes the final and is sure that he did not pass it. He concludes that he will not pass the course.

There are a couple of ways to interpret the student’s argument. One way makes it a case of denying the antecedent. He therefore deserves to fail. Another way, perhaps better, is take the teacher to have meant ‘if and only if’ rather than just ‘if’. This makes the student’s argument valid. It’s not implausible to think that sometimes we do say ‘if’ when we mean, or least could justifiably say, ‘iff’. In fact, if you don’t consider this revised argument, you’ve failed to do a good job of reconstruction.

Key point: Sometimes the precise reconstruction departs from the wording of the original statement in order to make the argument valid.

4) Consider:

1. x is a red ball.
2. x is red.

What’s the pattern? You could treat ‘red ball’ as a single predicate. The resulting pattern is no good. Better to treat ‘red’ and ‘ball’ as separate predicates and take the form to be:

x is AB

     x is A 

You could also represent the premise as:

x is A and x is B

and then draw the same conclusion.

But contrast expressions like ‘fake fur’ or ‘aspiring president’ or ‘former millionaire’. You could not represent these in the same way, since the argument from the premise to conclusion that the thing is a fur, president, or millionaire would be invalid.

It is best to say that arguments like these are valid only when the premise is equivalent to “x is an A and x is a B”.

Key point: figuring out the pattern of an argument requires being sensitive to the way the terms in the argument work. You cannot do it mechanically.

5). See p. 129, #3.

1.

1. Capital punishment might lead to the death of innocent people.
2. We should not do anything that might lead to the death of innocent people.
3. We shouldn’t do (or use) capital punishment.

Notice that the conclusion is not “I oppose capital punishment.” You could rewrite (2) as”

2a. All things that might lead to the death of innocent people are things that we should not do.

2.

1. Most college students are male.
2. Most males drink beer.
3. Most college students drink beer.

Notice that the conclusion is not “I know that most college students drink beer.” Notice also that the argument is invalid.

3.

1. Either Smith is the murderer or Jones is the murderer.
2. Jones is not the murderer.
3. Smith is the murderer.

No indicator words. Notice that the bit about the police dropped out. You could have added that as a premise supporting (1). This is an example of disjunctive syllogism.

4.

1. Anything that will make kids learn more will make us better off.
2. Making kids go to school 12 months a year will make them learn more.
3. Making kids go to school 12 months a year will make us better off.

Some stuff from the original omitted here.

5.

1. The president said that there will be a recession.
2. If the president said that there will be a recession, then the president believes that there will be a recession.

_____________
3. The president believes that there will be a recession.

Modus ponens. You might also use a generalization in (2) and make this another universal-particular syllogism.

************

More examples of reconstructing simple arguments

Key terms here: cheap validity, generalization, quantifier, universal generalization, linking premise, wide and narrow generalization

Generalizations are often omitted from arguments. We'll see how which one you add affects the evaluation of the argument.

Example: p. 143, #8.

Suppose we say:

1. E.Z. Marker gave an “A” to everyone of the 150 students in her class.
2. If (1), then E.Z. Marker is too lenient a grader.
3. E.Z. Marker is too lenient a grader.

This uses cheap validity. That is, it simply connects the premise to the conclusion by adding a premise saying that if the premise is true, then the conclusion is true. The resulting argument is valid. But it doesn't get out the idea behind the argument. It is better, whenever possible, to use some kind of generalization, i.e., a statement of the form “All As are Bs” or “Most As are Bs.” The “all” and “most” and other such words are called quantifiers, since they say how many or what quantity of the As are Bs. They can be imprecise, as in “Lots of As are Bs.”

Second attempt: replace (2) by “All teachers who give an “A” to all students in a class of 150 students are too lenient graders.” This is ok, but the generalization is narrower than it should be. Why limit it to classes of 150?

Third attempt: Replace (2) by universal generalization about “large classes.” It’s now ill-formed. We have to add a premise saying that a class of 150 is large.

Fourth attempt: add the needed premise. This may be ok, but it may be best to make the generalization less than universal.

Fifth attempt:

1. E.Z. Marker gave an “A” to every student in her class of 150.
2. Anyone who gives an “A” to every student in class of 150 gives an “A” to every student in a large class.
3. E.Z. Marker gave an “A” to every student in a large class.
4. Almost everyone who gives an “A” to every student in a large class is too lenient a grader.
5. E.Z. Marker is too lenient a grader.

Notice the generalization used. A wider generalization would be less plausible.

Universal generalization would be less plausible. The "I think" also suggests weakening the argument to a cogent one.

p. 143, #10

1. The Lazyboys are not well-motivated.

2. Usually, teams that are not well-motivated do not win.

3. The Lazyboys will not win.

A universal generalization would make it less plausible. And, maybe, the “I think” introduces the sort of caution that favors this reconstruction.

p. 147, 1a 

1. All democracies are places where everyone has a real say about what the government does.

2. The United States is not a place where everyone has a real say about what the government does.

3. The United States is not a democracy.

p. 148, 1c

1. All things that happen have a cause.

2. Your behavior is a thing that happens.

3. Therefore, your behavior has a cause.

4. If your behavior has a cause, then you don’t have free will.

5. You don’t have free will.

p. 148, 1k

1. Writing and speaking are just a reflection of thought.

2. If (1), then people who can’t speak and write coherently can’t think coherently.

3. People who can’t speak and write coherently can’t think coherently.

Another possibility:

1. If a person’s writing and speaking have certain characteristics, then that person’s thinking has those same characteristics.

2. If a person’s writing and speaking is incoherent, then that person’s thinking is incoherent.

Stating the pattern in this case would be difficult, given the tools developed here.

Missing Quantifiers

In an article about the upcoming baseball season a sportswriter might say:

Teams whose players have a lousy attitude, whose manager is incompetent, and whose pitchers should be used car salesmen do not win the World Series. Baltimore has ... So Baltimore will not win the World Series.

Here’s a reconstruction:

1. Teams whose players have a lousy attitude, etc. do not win.

            2. Baltimore is a team whose players have a lousy attitude, ...

            3. Baltimore will not win the W.S.

Pattern:

            1. Ls are not Ws

            2. b is an L.

            3. b is not a W

[L abbreviates the whole business. You could also have separate letters for each part in L. You could let W represent “not winning”.]

Evaluate this for form: You cannot do it! (1) is incomplete. It is a sentence with a missing quantifier. Linguists say that sentences like (1) have “bare plurals”. We talk this way all the time. It is often unproblematic. But for the purposes of doing logical analysis of arguments, we need to specify a quantifier. There is no simple rule, such as “insert ‘all’.” You just make it the best argument you can.