1. Why so much stress on validity when it does not guarantee truth among the premises or in the conclusion?
2. When identifying unstated premises, what kinds of assumptions, and how many of them, can one put into another person's mouth?

1. Why so much stress on validity when it does not guarantee truth among the premises or in the conclusion?

This question is natural and well motivated. The emphasis on validity can feel like a false promise, when validity delivers no truth. But our answer is just as natural: Logic almost never tells us whether a given sentence is true or false. (The exceptions are such sentences as "It is either raining or not raining," always true in any weather, because it is a tautology, and "It is raining and it is not raining," a contradiction and so always false, again regardless of the weather. But these claims convey no information.)

And why does logic not pronounce on the truth or falsity of a claim? Because it was not designed to. Logic exists in order to let us know whether a group of premises works to produce or support the argument's conclusion. Certainly validity is not everything, because a valid argument with at least one false premise may have no value at all. But without validity a deductive argument cannot be worth anything.

Perhaps the greatest appeal of validity follows from the fact that it does not come in degrees. An argument is either valid or not; and we can use strict, clear, foolproof rules to tell which to call it. Nor can you improve on the internal structure of a valid argument: As far as the relationship between conclusion and premises goes, validity is the most that an argument can aspire to, and the most it can achieve.

Very often people try to express the reliability of valid arguments with special language. "If the premises are all true, the conclusion can't conceivably be false"--or "absolutely must be true," or "has to follow." Such language seeks to capture the sense one gets from a valid argument, that there is no room for doubt about the step between premises and conclusion. "All humans have lungs. All animals with lungs have hearts. Therefore, all humans have hearts." Take the premises and you have taken the conclusion, too.

The language of conclusions absolutely proceeding from premises may help you understand validity. If it does not, forget it. It suffices to say of valid arguments that their conclusion is true whenever all their premises are true. True is true; talk of guaranteed or infallible or certain truth hammers home the point that valid arguments don't fail, but it adds no information. Just know that when you have accepted all the premises of a valid argument, you will have no grounds for rejecting its conclusion.

2. When identifying unstated premises, what kinds of assumptions, and how many of them, can one put into another person's mouth?

Every day we make assumptions about other people's assumptions. You assume that the waiter who approaches your table with pen and pad assumes you will order food. The friend who gossips about a third party assumes you won't tell that third party, and you know it.

Life without the making of such assumptions belongs to those unsocialized types who need every little thing spelled out.

To a great extent, identifying unstated premises works like these ordinary interpretations of other people. Anyone who makes an argument wants it to be a good argument, and must be assuming whatever plausible claims make the argument good. We justify our act of supplying those assumptions with the principle of charity, according to which we interpret another person's claims so as to make them as true as possible, as coherent as possible, and--when the claims come packaged together in an argument--as valid or strong as possible.

The principle of charity goes too far only when the desire to make another person's argument valid (or strong) leads us to attribute assumptions to the argument that the person who's making it could not want to hold, or could not reasonably be expected to believe. Someone says: "I'm thirsty, so I must have had salt in my lunch. Because whenever I'm not this thirsty, I've left the salt out of my food." We may lay out this argument as follows:

If I'm not thirsty, my food has been left unsalted.

I'm thirsty.

Therefore, my food was salted.

This is an invalid argument, resting on a common error called the fallacy of denying the antecedent (see Chapter 10). You can make it valid by adding this premise: "When not being thirsty follows from not adding salt to food, then being thirsty follows from adding salt." Not a perfectly plausible claim, since we get thirsty for other reasons; but nor is it a howling error.

Do we attribute this assumption to the speaker? In all likelihood, the person is not assuming any such complex claim about nutrition, only making a little logical faux pas. The principle of charity takes us too far if we let it commit other people to all the premises that would fix their arguments.

In practice, being human, we are more likely to make other people's arguments too bad than too good. But it will not hurt to remember that the search for unstated premises needs to be tempered not only with a recognition of which assumptions are plausible in themselves, but also with a recognition of which ones are plausible as the speaker's or writer's thoughts.