Because

  How to Analyze and Evaluate Ordinary Reasoning

  Similarity and Difference

 G. Randolph Mayes

 Department of Philosophy

 Sacramento State University

 

Similarity and Difference

Rational inquiry is often initiated by the detection of an unusual similarity or difference.  Reasoning that depends on reasons that make explicit claims about similarity and difference is called analogical reasoning.  Two analogical arguments can be distilled from the following example.

Example 5

  • Rachel:  Dad, why does Jessie get to go out tonight?  He didn't finish his homework this week either.
  • Dad:  I know that, but Jessie happens to be getting straight A's and he didn't finish his homework because he was helping me fix the car. You just spent too much time on the phone.  And Jessie and Mike aren't just going out tonight, they're going to help with the homecoming float.
  • Rachel:  But I want to help with the float, too!
  • Dad:  Well, that's interesting.  That's not what you wanted to do before I grounded you. Your mother tells me you were going to the movies with Vinny.
  • Rachel:  It wasn't Vinny and he's not vile!  See, you're not grounding me because I didn't do my homework, you're grounding me because you thought I was going out with Vinny!

The respective principles in the above rationales are typical variations on the following two analogical principles. 

  • Principle of Similarity: If x and y are the same, then x and y should be (or are) treated the same.
  • Principle of Difference: If x and y are different, then x and y should be (or are) treated differently.

These principles sound sensible, and  when the terms "same" and "different" are interpreted very strongly (e.g., if "same" means "identical" or "sharing every single property to the same degree"; and "different" means "absolutely distinct" or "having absolutely no properties in common") they have some obviously sound applications. 

For example, if you have the job of grading apples for quality and you see two apples that seem to be identical in every respect they should get the same grade.  On the other hand, if you are just looking for an apple to eat and you are given these two apples to choose from, so that  there is no legitimate to prefer one to the other, you still aren't going to treat them totally the same, since you are going to eat only one of them.

Similarly, if your job is to separate apples from oranges, then if you see an apple and an orange you will treat them differently by putting the apple in the apple box and the orange orange box.  But, if you are hungry for an apple and an orange, then you might eat them both, thereby violating the principle of difference which instruct you to treat them differently.

In ordinary contexts the situation gets quite a bit murkier because "same" and "different" are usually used to mean "having some of the same properties" or "having some different properties."  And the problem here is that just about any two objects in the world have some similar properties and some of the same properties.  Your brain and a marshmallow are similar in that they both contain a lot of carbon and they will both fit in inside a large jack-o-lantern.  On the other hand, they are different in that marshmallows can't solve logic problems and your brain consumes more calories.

All this means that analogical reasoning is often very unreliable no matter how careful you are, but to give it the best chance of being useful we need to be sure that claims about similarity and difference are appropriate to the context of comparison.  In the example above, it's easy to see that Rachel and Dad differ, not so much about the actual similarities and differences, but about which similarities and differences are strongest and most relevant in the given context to the conclusion being drawn.  Rachel claims that the similarity between Jessie and Rachel is strongest and most relevant, and that is why Jessie and Rachel should be treated the same.  Dad claims that the differences between Jessie and Rachel are strongest and most relevant in this context, and that is why they should be treated differently.

This, of course, just raises the further question how one establishes strength and relevance in a context.  Here the matter is pretty easily resolved.  The differences between Jessie and Rachel are relevant to differential grounding because grounding in this context is ultimately for the purpose of behavior modification, and Rachel's behavior is clearly in greater need of modification than Jessie's.  The differences are stronger, because  So Dad appears to be right this time. But things might have been different.  If, for example, there had been an explicit household rule that "If child x doesn't do x's homework, then x is grounded for the following weekend," then Rachel's similarity really should have carried the day.  Even though the rule is still ultimately about behavior modification, another issue would have had even greater relevance, and that is logical consistency and respect for the rule of law. (Relevance and Logical Consistency will be treated in greater detail below.)  This discussion lead us to codify the following two errors.

Questionable Analogy (Weak Comparison)

  • Basing a conclusion on an alleged similarity between two or more things when it is not clear that the similarity in question is (a) sufficiently strong or (b) sufficiently relevant to the context to provide adequate support for conclusion.

MOI:  Explicitly identify the similarity in question and identify why you think it is insufficiently strong or insufficiently relevant to the context to warrant the conclusion.

Distinction without a difference (Weak Distinction)

  • Basing a conclusion on an alleged distinction between two or more things when it is not clear that the distinction in question is (a) sufficiently strong or (b) sufficiently relevant to the context to provide adequate support the conclusion.

MOI:  Explicitly identify the distinction in question and identify why you think it is insufficiently strong or insufficiently relevant to the context to adequately support the conclusion.

These errors are obviously very similar in some ways, but their differences are very strong and very relevant to this context!

Example 1

  • Fletch: I don't care who wins the election.  All politicians are the same. They're all liars and crooks.

Identification:  This is a weak comparison.  The similarity is relevant to the context, but not particularly strong. Even if it were true that all politicians are the same in the sense of being liars and crooks, some liars and crooks may be better or worse politicians than other liars and crooks.  For example, both Richard Nixon and Bill Clinton apparently lied to congress and broke the law.  But they were not equally good or bad presidents because of this.

Example 2

  • Amanda:  Ash, I don't know how you can keep talking on the cell phone while you're driving when you know how dangerous it is.  It kills more people than drunk driving. You're basically a murder waiting to happen.
  • Ashley:  I know, but it really isn't that dangerous.  The reason it kills that many more people is because there's a lot more cell phone users than drunk drivers.  Every one who drives rather than walks is a murder waiting to happen if you want to think that way. Besides, I use a hands free cell phone.
  • Amanda:  You still shouldn't do it.  That study we read said that there's no difference between hand-held sets and hand free sets.
  • Ashley:  Well, that can't be right.  It's just common sense that a hands free is safer. After all, you've got both hands on the wheel.  I know that I am very much in control.

Identification

This is a weak distinction.  Ashley makes a distinction between hands free cell phones and hand held cell phones with respect to safety.  This is a relevant difference, but it is weak since there is evidence against it, and the evidence she gives in support of her view is just her own subjective feeling of safety.

It's worth noting that Ashley's first point is actually quite legitimate.  It's not nearly as dangerous for a single individual to talk on the cell phone while driving as it is to be intoxicated while driving, and that seems to be the relevant context for individual decision making. But Ashley  does makes a questionable distinction between the two kinds of cell phones.  She is rejecting the data that show that there is no measurable difference between the two types of cell phone as it relates to traffic accidents simply because she feels it's wrong. 

Example 3

  • Klaus:  You know, I used to really think that being gay was totally perverted and wrong.   But I'm majoring in biology now, and I'm  getting like this whole different perspective.  I've been reading how common homosexual behavior is in all sorts of other species.  So I'm really starting to see that homosexuality is just a totally natural phenomenon, and that we shouldn't be so judgmental about it.


Identification

Weak Comparison.  Klaus compares homosexuality in humans to homosexuality in other animals, arguing that since homosexuality isn't wrong in animals, it isn't wrong in people either.  This relies on the assumption that human homosexual behavior is like the homosexual behavior of other species, as the rational below demonstrates:

The two behaviors, while physically similar, are not similar in a way that is relevant to drawing moral conclusions.  This is because the behavior of non human animals is not subject to moral evaluation.
   
Example 4

  • Fayanne:  You know, I just voted and rather than feeling good about it, I almost wanted to cry.  There were these people in line who obviously had no idea what they were doing.  The whole time I'm thinking, my vote counts the same as these idiots?
  • Gabriela:  Oh God, totally. I don't even get why people like that should have the right to vote.  I think there should be some kind of test you have to pass, like our driving test.  Bad voters do just as much harm as bad drivers, right? Why shouldn't they have to get a license to vote, too?

Identification:  Weak Comparison.  Gabriela argues that people should not have the right to vote, rather that they should first have to demonstrate competency by passing a test. Her reason for this is that ignorant voting is like bad driving with respect to the degree of harm that can be done, and a test is required of all prospective drivers.  This is a rather weak comparison because (a) a bad driver can do much more harm than a bad voter and (b) the effectiveness and fairness of a driver competency test is more easily established than the effectiveness and fairness of a voter competency test.

After you examine a bunch of comparisons and distinctions critically it's easy to start assuming that most comparisons and criticisms are weak, which of course is not correct.  The next two examples provide an antidote to this tendency.

Example 5

  • Seymour:  Don't you think it's funny the way people are all down on kids playing computers and video games because we should be outside playing, but then when they hear about some kid who has his nose in a book all the time and they're all  "Oh, that's so wonderful. I wish my Seymour would read more." 

Identification

This actually seems to be a strong comparison.  Seymour argues that reading and playing computer and video games are similar insofar as they are both sedentary activities.  It's possible that reading has other benefits that video games doesn't, but that's not relevant to his rather limited point. 

Example 6

  • Lars:  Olaf, you know what you said about Gustav saying he was going after my girl Gretchen?  I went to kick his ass, and he said he doesn't even know Gretchen.
  • Olaf:  So, what, he's calling me a liar? 
  • Lars:   He said it wasn't true. He said he has his eye on Hilda.
  • Olaf:  He's calling me a liar.  Damn, now, I have to go kick his ass.
  • Lars:  OK. 

Identification:  This is not a weak distinction.  There is a very important difference between lying and saying something that isn't true, which is highly relevant to the context.  Lying consists in willful deceit, but one can say something false by accident.

Finally, it's important to realize that whether  a comparison or distinction is weak or strong is often a matter that can not easily be decided, especially by people with limited knowledge of a subject.  In fact, analogies are quite often used to make people feel like they understand things when they really don't. 

Example 7

Senator Thrift:  Most of my colleagues in the Senate want the U.S. to borrow what will amount to trillions of dollars to bail out major financial institutions and manufacturers.  These are corporations who largely created their own problems by running their businesses in irresponsible ways.  Every responsible working person knows that when times are tough you do not borrow money to finance your spendthrift ways.  What you do is quit spending so damn much money.  You tighten your belt, learn to live with less, work harder, and pay off your debts.  Why should the government be any different?

Identification

This sounds like a compelling comparison, but the truth is that the government is not like an individual household in this regard.   Government has longer time horizons, a much greater ability to bear risk, and an almost unlimited access to capital. More important, one of the legitimate functions of the government is to stabilize the economy so that individual households can function.  So while it's true that the government has to pay attention to a lot of the same financial issues that individual citizens do, it will sometimes be rational for the government do things that would be irrational for an individual.