Analogies The Final Installment

In the past two posts, I've explored the benefits and the dangers of using comparative language to express an idea or an argument. Referencing a book by Robert J. Gula, I distinguished between the poetic purposes of an author and the logical purposes of an author.

A writer or reader or audience member of any kind must consider how poetic and logical purposes are different.

Since the purposes of poetry are different from the purposes of logic, our standards in assessing the effectiveness of each will also be different. A poetic comparison, if it succeeds at all, will succeed on poetic terms; a logical comparison, if it succeeds at all, will succeed on logical terms.

As you've read in my two previous posts, the terms that the laws of logic impose on us are strict.

In this post, we'll look at the last example that Gula gives to illustrate a fallacious chain of analogical reasoning, and then we'll look at the recommendations he offers by way of summary.

In Nonsense Gula writes, "Analogies are abused when a person uses the terms of one element to predict  the terms of another element" (144).  To show this definition in action, let's look at the following exchange:

P: "Who are voting for?"

Q: "Probably Smith.  I'm an environmentalist, so I think the government should be putting a lot more money in making renewable energy a practical way for the United States to generate electricity and transport its citizens to and from wherever it is their going.

P: "If the environment is important to you, then you shouldn't vote for Smith."

Q: "Why not?"

P: "Well, because take a look at the current administration's record on environmental issues. I guess it hasn't had any photo ops where high ranking cabinet members actually shoot polar bears from helicopters, but needless to say, it hasn't exactly been friendly to the tree and the flower.

Q: "But that's irrelevant. What this administration has done is one thing. What Smith will do is an altogether different thing."

P: "No. Not when one considers all the similarities between the current president and Smith. For example, the president's father made his fortune in the oil business, and so did Smith's. The president attended Exeter, then Princeton, then The University of Chicago School of Law. Guess what Smith's academic career looks like? You got it, Exeter, Princeton, Hyde Park. After he graduated from law school, the president was awarded a federal clerkship before going into politics, just like Smith. I think it's clear from all these similarities that since the president said one thing during his campaign about the environment but ended up implementing a different policy, we can conclude that Smith will do the same thing."

If you can forgive my long winded example, what is your assessment of it?

P tries to argue by analogy. P connects several similarities between the current president and Smith. P then goes further by drawing a conclusion about Smith's future behavior. Such an analogical reasoning chain would be fallacious.

At the end of his discussion of analogies, Gula (144) recommends answering the following questions when evaluating the cogency of an analogical argument:

1. "Have all the properties of X (the object of analogy) and all the properties of Y (the other object of the analogy) been cited?"

2. "How many of these properties are similar?"

3. "How many of these similarities are relevant?"

4. "To what extent are some of these similarities actually not as similar as they have been made out to be?"

5. "To what extent is X different from Y?"

Analogies are everywhere. We use them all the time. When we are being asked to assent to a speaker's conclusion based on analogy, we should be aware that arguing from an analogy has a high standard to satisfy before the argument's conclusion can be accepted.

Analogies Part I

In Nonsense, Robert J. Gula spends a good amount of time discussing and evaluating the use of comparison both as a poetic feature of our language and as a logical feature of our language. I think there is enough material to justify at least two blog posts on the topic. This is the first. 

Gula begins chapter twelve by recognizing the benefits of having the linguistic and physical mechanisms capable of generating an almost infinite number of comparisons and contrasts: "Devices of comparison and contrast can aid us in expressing ourselves. They add vividness and richness to utterances, they are sometimes a way of making concrete an abstract idea, and they sometimes provide a means by which we can clarify a complex idea" (141).

Gula then uses his poetic knowledge to illustrate his point. He refers to an evocative comparison made by John Donne, the sixteenth century English poet. 

"[W]hen the poet John Donne compares two people in love to the legs of a drawing compass, his comparison allows him to concentrate a complicated thought into just a few lines [. . .]" (142).

In one regard, the legs of a drawing compass are individualized; they are detached from one another. This detachment allows the drawing compass to perform its intended function. Yet the legs of a drawing compass are not completely detached from one another. They share a common hinge. 

The image of a drawing compass's legs provides us with an concrete image of separation and unity. When two people are in love, they are still two individual people. One can fly to Miami; the other can fly to Helsinki. True, they are not constrained by something as tangible as a pair of handcuffs, yet  they do seem to be constrained in a more abstract way. The thoughts of each person for example may repeatedly cycle back to pictures and articulations of the beloved, thus constraining each's ability to effectively perform any task that requires an act of higher cognition. Each may be constrained emotionally by the absence of one's beloved. 

Thus, the balancing of separation and unity is nicely captured by the concrete image of a drawing compass. Donne's analogy is a creative achievement to be enjoyed.

Some creative achievements may not have the poetic excellence or the literary endurance of a lyric by John Donne. For those of you who are interested in the stock market, an event recently occurred--Apple Computer released its first quarter earnings results on Wednesday--that gave people a chance to create some charming colloquialisms. For example, I heard one person say that "Apple is sitting on a mountain of cash." 
I guess an image is helpful when you try and understand how big a pile of money 137 billion dollars actually is. 

I heard another person say, after the earnings report came out and the stock was down about ten percent, "Apple is a melting ice cube." 

This expression is, in my opinion, better than the previous example because it's less cliche, and it captures in a familiar image the bearish case against Apple Computer. Again, it may not have the longevity of a Donne poem, but comparing a stock that was once the darling of Wall Street to a melting ice cube is a analogy that I appreciate; it is a creative achievement that helps people understand a complex phenomenon. 

However, like other creative achievements that are properly enjoyed, analogies can be used immorally. 

According to Gula, "It is one thing to use an analogy to help convey an idea. It is another thing to argue by analogy or to use an analogy to form an inference or judgment. [. . .] Analogies are abused  when they try to claim similarity and to establish identity under the guise of merely suggesting that similarity" (142).

Gula then provides four examples to clarify the sorts of situations in which analogical reasoning is abused. I will explore the first one here and continue exploring the rest in next week's post. 

Suppose you are discussing with a colleague  your favorite chili recipe. "My grandfather's chili is the best I've ever had."
"Yeah," your colleague replies with a dismissive smirk. "That's because you've never had my recipe. For reasons of nostalgia and elderly respect, I know my chili can never surpass your grandfather's." Your colleague, smiling, then looks you in eye and says, "Parity, I think, would be impossible for you to dispute." 
"I doubt it."
"What's in your grandfather's chili? Tomato sauce? Chili Powder? Some cayenne? Garlic? Salt, pepper? Right?" The colleague nods, shrugs his shoulders. 
"Right. Well I put all that stuff in my chili so your grandfather's chili and mine would taste the same. Equally delicious."

Here we have a speaker claiming that because some object (a pot of his chili) contains several of the same ingredients that another object contains (a pot of grandpa's chili), they are identical in taste. But it seems like a pot of chili is too complex an object to be generalized in this fashion. Sure, the two recipes share some features, share some ingredients, but the conclusion that they will taste the same because of their shared ingredients would be fallacious. The similarity between the two pots of chili would need to be more comprehensively established before such a conclusion would be justified.