Chapter 21: Analogical Reasoning

 

Chapter 21: Analogical Reasoning

•   In an analogy, we compare one thing with another. We might describe a person as being like a fox, a prickly rose, a robot, or a hurricane. Love is said to be a disease, a game, a drug, a heatwave, and “a smoke made with the fume of sighs” (Shakespeare's Romeo and Juliet). Less poetically, a physicist might compare an atom to the solar system—electrons revolve around a nucleus at the centre as planets go around the sun. 

•   In this section, we focus on the use of analogy in explanation and argument (literary theory distinguishes between metaphors, similes and allegories. They are all based on resemblances, and we treat them all as analogies here). The first point to note is that words such as similar and like have incomplete meanings. Saying that two things are similar has a concrete meaning only with respect to some standard of comparison. Pick any two objects, and they are bound to be similar in some way. A washing machine is like a pigeon—they both occupy space, produce waste, and are noisy. An informative analogy should make it clear how two things are similar—that is, what their common properties are. Former U.S. President Ronald Reagan once said “Government is like a baby” which on its own is rather incomprehensible. But then he added, “an alimentary canal with a big appetite at one end and no sense of responsibility at the other,” which immediately made the remark a witty and memorable one about government over-spending.

•   Identifying common properties is very important in analogical arguments, where we argue that because X is similar to Y, something that is true of X is also true of Y. Here are some examples of arguments by analogy:

•    This new pair of shoes is like my old pair. My old shoes were very comfortable. So this new pair is probably comfortable as well.

•    It is wrong to enter somebody's home uninvited. Hacking into their computers is similar. So unauthorised hacking is also wrong. 

•    You are just like my friend. He likes rock climbing. So I think you will too. 

•   These arguments all have the same form:

X is similar to Y.
X has property P.
Y also has property P. 

•   But as we said earlier, we should be explicit about how two things are actually similar. In an analogical argument, this means identifying the properties common to X and Y that explain why the conclusion is credible. Let's try this with our three arguments: 

•    This new pair of shoes is like my old pair. They are made by the same company. My old shoes were very comfortable. So this new pair is probably comfortable as well. 

•    Hacking into a computer is similar to entering a home uninvited. In both cases, we are trespassing on other people's property without permission. Since it is wrong to enter someone's home uninvited, it is also wrong to hack into other people's computers. 

•    You are just like my friend. You both enjoy outdoor physical activities. My friend likes rock climbing. So I think you will too.

•   The underlined sentences identify the common properties, and the reasoning becomes clearer. An explicit analogical argument, therefore, has this form:

X is similar to Y in that they have common properties S₁, S₂… S_n.
X has property P.
So Y also has property P. 

•   This is the most important part of analogical reasoning because it helps us understand why an analogy is supposed to work. The argument must highlight important aspects of similarity if the analogy plays any role at all in supporting the conclusion. Then you can start evaluating the argument, using the principles below. 

1.      EVALUATING ANALOGICAL ARGUMENTS

There is no mechanical method for evaluating analogical arguments, but here is a checklist of the main criteria: 

•     Truth: Are the two things really similar in the way described? Obviously, an argument is not acceptable if it has a false premise. In the third argument above, if it turns out that you do not like outdoor activities and much prefer sleeping in bed, then you are not like my friend and the argument should be rejected. 

•     Relevance: Are the shared properties relevant to the conclusion? Even when the source and target are similar, the properties they share must be relevant to the conclusion for the analogical argument to be acceptable. In other words, having shared properties increases the probability of having the inferred property. To use a concrete example, suppose we change the third argument above slightly: 

You are just like my friend since you both like to eat chocolates. 
My friend likes rock climbing. 
You will also like rock climbing.

•     This is clearly a lousy argument, even if the premises are true. A preference for chocolate does not make a person more likely to enjoy rock climbing. In other words, the common property is simply irrelevant to the inferred property. Notice that we need to use our common sense and background knowledge to determine relevance. It is not enough to focus on the argument only. For example, if scientists discover that the majority of people who like chocolates actually enjoy rock climbing, then our background knowledge has changed, and this argument would become more convincing. This shows that analogical reasoning is typically a kind of inductive reasoning. 

•     Number and diversity: Are there many shared properties of different types? The strength of an analogical argument depends not just on the relevance of the shared property. The number of shared properties makes a difference also. If both you and my friend enjoy outdoor activities, and in addition, you are both agile, with good physical strength and balance, then it is more likely that you will like rock climbing. In short, finding more relevant properties shared by the source and the target can strengthen an analogical argument.

•     Furthermore, the argument is more convincing if the relevant properties are of different kinds. For example, having strong arms is relevant to rock climbing, as having strong legs also, of course, strong fingers as well. But these are all traits of the same kind. Analogical arguments are stronger when they are based on shared relevant properties of different kinds.

•     Disanalogy: Are there significant differences between the things being compared? An analogical argument that seems strong can still be undermined if there are important dissimilarities between the source and the target. You might be agile and physically fit, and enjoy outdoor activities just like my friend, but if you are afraid of heights and my friend is not, this one critical difference makes it very unlikely that you will enjoy rock climbing. So be very careful when you evaluate an analogical argument. Even if it appears convincing and the items being compared share lots of relevant properties, it can still be refuted by a single disanalogy. 

2.      TREATING LIKE CASES ALIKE

•     Analogical arguments are important in legal and moral thinking. One reason is that we want to be consistent and fair. As Aristotleputs it, justice requires “treating like cases alike.” If a man and a woman have similar abilities and performance but the man is given a higher salary, we are inclined to think that this is unjust. Similarly, if two people committed similar crimes and their situations are analogous, it would be unfair for one to receive much heavier punishment.

•     Are there exceptions to the rule that we should treat like cases alike? Suppose I decide to donate money to help sick children instead of starving refugees. The two causes are similar in that they are both worthy of support, yet I am not wrong if I donate money to one but not the other. Or consider two students competing for just one scholarship. Both are brilliant and equally accomplished and deserving, and in the end, the scholarship went to one of them chosen by a random lottery. Should the unsuccessful candidate complain of injustice? Surely not. 

•     One thing we might say about these examples is that if we look deeper, the cases being compared are not really the same. In the first example, the difference is that I prefer helping children to refugees, and this is a morally relevant difference because I have the right to spend money as I wish. In the second example, the students are different because one of them won the lottery and the other did not. This is again a relevant difference because when resources are limited, it is legitimate to pick a recipient randomly if they are equally worthy. So the rule that we should treat like cases alike is safe after all. 

•     A critical and dedicated thinker would use analogies to reflect on the reasons behind his moral opinions. For example, animals are helpless and not very clever, just like young children. So why eat animals but not children? What is the relevant difference between them? Or consider the case of abortion. If a woman has the right to remove a tooth, why can't she have an abortion to remove a foetus (unborn baby)? Is it because a foetus has the potential to develop into a human being? But a tooth contains nerve cells that perhaps can also develop into a person given the right kind of technology. Thinking about these similarities and differences helps us understand more deeply the basis of our moral judgments. 

END OF THE PART