Because

  How to Analyze and Evaluate Ordinary Reasoning

  Section  13:  How we diiscover principles of reasoning

 G. Randolph Mayes

 Department of Philosophy

 Sacramento State University

13. 1  The problem of induction

 

In section 12 we learned that the connection between a reason and its conclusion is mediated by a principle. We also learned that people rarely state their principles, which means that the first step in critically examining the connection is to figure out what principle is being used.

 

Understanding the human capacity to discover principles is one of the most difficult problems in philosophy.  It is called the problem of induction.  For our purposes, the problem of induction may be understood as follows:  How can we know that one principle rather than another is operating in any particular situation?

 

This question was originally formulated, not in regard to discovering the principle of a person's reasoning, but rather in regard to discovering the principles operating in nature.  Science represents humanity's best attempt to discover the basic laws of nature, but scientists are always haunted by the possibility that what they think they have discovered to be a law of nature is actually just an accidental regularity of some sort.

 

Here is an example.  For many, many years ancient astronomers believed that planets traveled in circular orbits around the earth.  Of course,  we no longer teach that planets orbit the earth, but we also now know that the planets travel in elliptical orbits, not circular ones.  This insight was actually key to discovering that planets are in orbit around the sun.  Part of the reason this took so long to discover is that the orbits astronomers knew best, like the orbit of a ball on a string, is circular. 

 

Here is a more practical example:  If you were born and raised in a religious community of some sort, then unless you were told otherwise, you would naturally have assumed that everyone shares your religious beliefs.  But when, as an adult, you venture beyond your community, you realize that what you thought was a very trustworthy principle is, in fact, just an accident of your birth.

 

The point here is simply this:  There is no perfect recipe for determining what principle is operating in nature.  We generalize from our limited experience, and as we learn more about the world, we find that we often need to revise or discard our principles in favor of ones that are better.

 

13.2  Attributing principles to people's behavior.

 

The problem of attributing principles to the behavior of people is very similar.  Sometimes it will seem  obvious from a person's behavior that she is following a particular rule, when in fact she is following a completely different rule.

 

For example, suppose Margie always plays Reggae music when you are visiting. You might reasonably infer that Margie really likes Reggae music.  But in fact it is possible that Margie doesn't like it at all.  Rather she assumes that since your hair is always in dreadlocks that you must like Reggae.

 

Now you'll notice that in a case like this, you could simply ask your friend why she plays Reggae music so much and, if she is honest, things will be cleared up.  You'll then realize that the familiar principle

  • If person x engages in entertainment y often, then x enjoys y.

is in this case being overridden by another familiar principle, namely:

  • If guest x likes entertainment y, then y should be made available to x.

This is a distinct advantage we have over scientists who study non human nature.  Scientists can't just ask nature what principles it is using, because nature just won't say.

 

 

13.4  The problems of attributing principles to people's reasoning

 

But when it comes to studying reasoning behavior, our position is a little more like that of the natural scientist. The problem, as we noted earlier, is that we are often not in a position to ask people what principles they are using (since they aren't there to answer), and because they are not students of logic, they may simply not ever have thought about it.  So the perpetrators of reasoning may be just as mute as nature is with respect to the underlying principles.

 

Fortunately, as we noted in section 12, we are not trying to discover what principles are actually being used, but rather which principles would make the most sense of the reasoning being offered.  We already know that one requirement of this sense-making process is to attribute principles that result in deductively valid rationales.  But there will always be more than one principle (many, many, many more) that have this effect. 

 

Different principles that satisfy the requirement of deductive validity equally well can differ in two ways:

A.  The principles vary in regard to their degree of generality.

B.  The principles generalize on different things.

 

To see how these differences can arise, consider the following example:

  • Simon bought a rat terrier in order to solve his vermin problem.

This is clearly an explanation of why Simon bought a rat terrier, which we would reconstruct as follows.

 

What principle is involved?  Consider the following possibilities:

P1:  If Simon needs to get rid of vermin, then Simon buys a rat terrier.

P2:  If person x has problem y, then x buys a dog.

P3:  If person x has problem y, then person x attempts to solve y.

P4:  If person x has y, then person x does z.

P5:  If x, then y.

Of these choices, P1 and P5 are most obviously incorrect. P1 actually does not even secure the deductive connection, because it simply restates the reason and the conclusion as a conditional.  The reason and conclusion can't "plug into" the principle because there is nothing to plug into.  P5 does create a deductive connection but only by generalizing so wildly that it can apply to absolutely anything. Both principles are useless, though they are useless in different ways.  P1 is useless because it is so narrow that it fails to connect the reason to the conclusion.  P5 is useless because it is so general that it has no meaning at all.

 

P2 and P4 are also incorrect, and for similar reasons.  P2 does make a deductive connection because it generalizes Simon as a person, vermin as a problem and rat terrier as a kind of dog.  The problem with P2 is that it is stupid.  Nobody in his right mind tries to solve every problem by buying a dog.  P4 is less general than P5, but still much too general.  It applies to absolutely any reasoning in which the reason represents a person having something and the conclusion represents the person as doing something.  If we were to accept this principle, then we would be committed to seeing all of the following as equally good reasoning:

  • Simon bought a rat terrier to solve his vermin problem.

  • Betty got a can opener to give to her canary.

  • Martha has a pimple to make dreams come true.

On the list above, then, principle P3 is the only satisfying principle. It secures the deductive connection with the the familiar idea that when people have problems they often try to solve them. 

 

It is important to see, however, that P3 is by no means the only principle that could make sense here.  Here are three other principles that are in roughly the same ballpark.

P6:   If person x has an infestation of rodents, then x obtains some pet that will kill rodents.

P7:   If person x has an infestation of pests y, then x obtains some z that will eliminate y.

P8:  If person x needs to accomplish goal y, then x obtains some z that will accomplish y.

Now all of these principles are quite reasonable, and are not in any absolute sense preferable to principle P3.  P1 might be regarded as insufficiently general, since it requires the death of the rodent, but it is not unreasonable to think that death is what is intended here.  P8 might be regarded as a bit too general and slightly less reliable, since people are more likely to deal with an immediate problem than a goal.  But this is a very small point.

 

13.5  The purpose of principles revisited

At this point it can be useful to remind ourselves what principle attribution is  all about.   In the most general terms, principles are how we understand the world.  When the world makes sense to us, it's because the world is behaving according to familiar principles.  When it doesn't make sense, it's either because familiar principles have been broken, or because we are dealing with a new subject matter whose principles we don't yet understand. 

We grasp other people's reasoning in an analogous way.  When their reasoning makes sense to us, it's because it corresponds to familiar principles. When it doesn't make sense, it's because the principle eludes us.  Of course, familiarity and being reliability are two different things.  Some highly reliable principles are simply unfamiliar, and so we must learn the principle if we are to understand the reasoning.  Other very familiar principles are quite unreliable, and as responsible members of the reasoning community it can become our job to criticize their use.

 

13.6  Criteria for the attribution of principles to reasoning

 

Here is a summary of four basic criteria for using principles to connect reasons to conclusions.

  1. The principle must have the form of a conditional (if...then) statement , in which the antecedent connects to the reason and the consequent connects to the conclusion.

  2. The principle must be more general than the reason and the conclusion.

  3. The principle must not be so general that no reasonable person would assert it.

  4. When attempting to decide between principles that meet criteria 1-3 equally well, choose the principle that appears to be most reliable.

In order to meet criteria 3 and 4, it is essential to consider the principle  in isolation, rather than in connection with the reason and conclusion.  A good principle will always make sense on its own.