Arguments by Analogy

A very common form of argument in philosophy is the "Argument by Analogy". Though this kind of reasoning is probably pretty familiar, it's worth examining it in a bit of detail. This will lead to a greater understanding of how we can object to such analogical reasoning.

How Arguments by Analogy Work

Analogical reasoning is extremely common, both in philosophy as well as in more ordinary contexts. However, the reasoning involved is rarely made explicit.

Usually, people employ analogies to help establish a relatively controversial claim. They do so by trying to establish a similar claim in a less controversial case, which they claim is analogous.

So, the set-up for any argument by analogy is always the same. First, an analogy is drawn between the "controversial" case and the "uncontroversial" case. Second, a claim is made about the "uncontroversial" case. Using the picture I drew in class, the set up is as follows:

Based on this set-up, the conclusion is that, by analogy, the same thing is likely to be true in the controversial case.
To sum up, every argument by analogy can be reconstructed as follows:
1. The "controversial case" is analogous to the "uncontroversial case".
2. X is true of the "uncontroversial case"
----------------------------------
C. So, by analogy, X is true of the "controversial case" as well
Again, using our picture, the argument is as follows (with the conclusion in blue):
Now, it should be noted that analogical arguments are not deductively valid. The premises can be true without the conclusion being true. However, we ordinarily think that a good argument by analogy gives us strong reason to believe the conclusion.

In case the preceding discussion was a bit abstract, consider the following example:
One of the ways that scientists test whether various drugs will be safe and effective on humans is to observe the results of the drug on animals. Of course, the success or failure of the drug on some animal doesn't guarantee that it will have the same results on humans. (Again, arguments by analogy are not deductively valid.) But a comprehensive test on animals gives us good reason to think the effects will be similar on humans.
So, let's say that the latest drug by Drug Co. has terrible side-effects on a group of lab rats. This causes the FDA to not approve the drug. I take the FDA to be reasoning roughly as follows:

1. Humans are analogous to lab rats. (We are similar in a number of respects.)
2. Drug Co's latest drug caused terrible side-effects in the test group of lab rats.
----------------------------------
C. So, by analogy, the drug could have terrible side-effects on humans as well.

How to Object to an Argument by Analogy

Now that we've seen how arguments by analogy work, let's consider the various ways you might object to such an argument. As mentioned, arguments by analogy are not deductively valid. As such, we can't object to these arguments by simply pointing out that the premises don't guarantee the conclusion.
However, we can still object to the premises.

Remember that all arguments by analogy involve two claims:

1. The "controversial case" is analogous to the "uncontroversial case".
2. X is true of the "uncontroversial case"
So, to object to such arguments we can always question either of these claims.

First, we can question the analogy. As an intentionally bad example of analogical reasoning, consider the following:

1. Iced tea resembles whiskey.
2. Iced tea doesn't impair my ability to operate a motorized vehicle.
----------------------------------
C. So, by analogy, I can knock back a fifth of Jack Daniel's and get behind the wheel.
This is, of course, a bad argument. It is bad because the analogy between iced tea and whiskey (premise 1) is very weak. While iced tea does resemble whiskey in visual appearance, there are a number of relevant dissimilarities. They smell different, taste different, and, most crucial in the context of this argument, one contains alcohol while the other doesn't.

The general lesson here is this:

One way to object to an argument by analogy is to question the analogy. This is done by pointing out relevant disanalogies between the two cases.
Alternatively, we can always question the claim being made about the "uncontroversial case". If we have reason for doubting this claim, then regardless of how strong the analogy is, we have no reason for accepting the concluion. So:
Another way to object to an argument by analogy is to question the claim being made about the "uncontroversial case".