Chapter Four: Streamlining

His reasons are as two grains of wheat hid in two bushels of chaff; you shall seek all day ’ere you find them.

—William Shakespeare, The Merchant of Venice

“When you come tomorrow, bring my football boots. Also, if humanly possible, Irish water spaniel. Urgent. Regards. Tuppy.”
“What do you make of that, Jeeves?”
“As I interpret the document, sir, Mr. Glossop wishes you, when you come tomorrow, to bring his football boots. Also, if humanly possible, an Irish water spaniel. He hints that the matter is urgent, and sends his regards.”
“Yes, that’s how I read it, too. . . .”

—P. G. Wodehouse, “The Ordeal of Young Tuppy”

Streamlined objects, like airplanes or torpedoes, have had nonessential features removed so that those features will not impede movement. The streamlining of arguments has the same purpose. Nonessential features of the argument are removed so that they will not get in the way of evaluation. Streamlining is one important aspect of paraphrasing. Not every passage needs to be streamlined before it can be evaluated—take, for example, the letter from Mr. Glossop in the lead quotation. But many are closer to Shakespeare’s two grains of wheat in two bushels of chaff—they are far easier to evaluate if you first discard the chaff.

4.1 Paraphrasing Only with Statements

In everyday language—as noted in Chapter 2—arguments often include expressions that are not in statement form, but that nevertheless function as statements. It is necessary to reword these sentences so that in your clarification they are in statement form.

Practical reasoning provides an important opportunity for this sort of paraphrasing. Practical reasoning is concerned with reasons for doing things. I might, for example, deliberate over reasons that bear on whether to move to a new town, or I might give you reasons for changing your major. But conclusions like Move to a new town and Change your major could not be true or false (though they could be wise or foolish, followed or ignored). They are in imperative, not declarative, form. Notice, however, that reasons offered for doing something are also reasons offered for believing something—believing, namely, that it should be done. Imperative sentences, then, should be paraphrased to statements such as I should move to a new town or You should change your major. Either of these statements can be true or false.

Consider this cheerful argument:

Be glad of life because it gives you the chance to love and to work and to play and to look up at the stars.

The explicit premise, based on the inference indicator term because, is this:

1. Life gives you the chance to love and to work and to play and to look up at the stars.

And the conclusion, apparently, is Be glad of life. But—so that we’re working with sentences that can be true or false—it should be paraphrased as the statement:

Rhetorical questions—questions that do not call for an answer, but rather assert something—also should be paraphrased as statements, since, phrased as they are in interrogative form, they cannot be true or false. Consider this little passage from Raymond Smullyan’s This Book Needs No Title:

But that doesn’t seem to bother him! As he says, “Why should I worry about dying? It’s not going to happen in my lifetime!”

The explicit premise is this:

1. Dying is not going to happen in my lifetime.

But the conclusion Why should I worry about dying? must be paraphrased, since, as a question, it cannot be true or false. A much clearer version of the conclusion is this:

Fragments—partial sentences—can also function as statements, and when they do should be paraphrased accordingly. For example, look at this tongue-in-cheek passage from Ambrose Bierce:

In his great work on Divergent Lines of Racial Evolution, the learned Professor Brayfugle argues from the prevalence of this gesture—the shrug—among Frenchmen, that they are descended from turtles and it is simply a survival of the habit of retracting the head inside the shell.

The explicit premise of this argument doesn’t even appear as a complete sentence. It is easy to find it—the inference indicator argues from makes it clear that a premise is coming next. But all that comes next is the series of prepositional phrases from the prevalence of this gesture—the shrug—among Frenchmen. So, the fragment must be converted into a statement, namely:

1. The shrug is prevalent among Frenchmen.

The conclusion, of course, is this:

It goes beyond the scope of this book to judge whether this is true or false.

Remember that the ultimate aim of reasoning is to know the truth. Premises and conclusions will not be true if they cannot be true; and they cannot be true if they are not statements.

4.2 Eliminating Unnecessary Words

4.2.1 Eliminating Discounts

Often an arguer acknowledges possible objections to the argument and explicitly sets aside those objections. That is to say, arguers discount possible objections to the argument. Discounts are best understood in contrast to premises. As we have seen, premises are statements the arguer offers as supporting the conclusion. Discounts are statements the arguer offers as not undermining the conclusion, but discounts are typically neither premises nor conclusions and should usually not be included in your clarification.

Suppose I really want you to read a certain book. I might say to you, “Look, I know that you don’t have much time for fiction nowadays and I admit that the reviews have mostly been negative. But you should read it anyway. It will give you more insight into yourself than anything else you’ve ever read.” My conclusion is You should read the book. By pointing out the poor reviews and your lack of time, I surely am not giving reasons for reading it; these would normally be reasons for not reading it.

I achieve, however, at least two important things. First, I make it less likely that you misunderstand my meaning. Otherwise you might think, “Perhaps he means some other book, since he must know that book got terrible reviews.” So discounts help in the interpretation of the argument. Second, I make it less likely that you underestimate how thoroughly I have considered my argument and how serious I am about it. Had I not mentioned the two possible objections, you might have supposed I did not know about them and that I might well give up my argument upon learning about them. As a result of my mentioning them, you are probably less likely to raise those objections yourself. Discounts help to frame the dialogue—they are conversationally useful.

But they do not normally have any bearing on the logic of the argument. Being reminded that you have little time to read and that the reviews have been bad does not provide you with a reason to read the book I recommend. So you should look to discounts to help you understand the arguer’s intent, but you should normally not include them in your clarification.

4.2.2 Discount Indicators

The best clues to the presence of discounts are discount indicators (sometimes also called adversatives, since they indicate a nonsupportive, or adversarial, relationship). In the previous argument I urged, “but you should read it anyway.” But is perhaps the most common discount indicator. It typically points backward to discount the previous clause. Other backward-pointing discount indicators are illustrated in the brief passages (not arguments) below:

I woke up with a fever. I decided, however, that I should go to work.
It rained all day. Nevertheless, we had a wonderful time.

Other discount indicators typically point forward to discount the clause that immediately follows. These include:

In spite of the great care I took, I managed to put my foot in my mouth.
Although I’m unhappy with her views on healthcare, I voted for the incumbent anyway.
Regardless of how hard you have worked, we have to fire you.

Note that in clarifying an argument, you are to eliminate not only the indicator term but also the entire discounted clause. Consider the following passage adapted from Charles Peirce:

Though those who claim to know everything there is to know about science seem to me comical, I highly respect them, for they make up the majority of those who have anything interesting to say.

The first clause—that those who think they know everything about science are comical—is discounted by the forward-pointing though. The argument comes next:

The discounted clause is not a part of the argument, so it is eliminated from the clarification.

These terms are not perfectly reliable indicators of the presence of discounts; sometimes they merely introduce a negatively stated premise—for example, We are either mortal or immortal; however, we are not going to live forever, so we are mortal. This may be clarified as follows:

Although it is introduced by the term however, the statement We are not immortal is essential to the support of the conclusion. It is not merely a “statement that does not undermine the conclusion,” as it would be were it a mere discount.^[1] Before eliminating the presumed discount, be sure the argument can function without it.

4.2.3 Eliminating Attitude Indicators, Report Indicators, and Inference Indicators

Another item to eliminate in the streamlining process is the attitude indicator, which indicates the arguer’s attitude of either belief or disbelief toward a statement. Philip Howard writes in Verbatim,

I believe that the criticism of the Oxford English Dictionary for neglecting spoken English is largely mistaken. It seems to me that a new word or new use is written down somewhere almost as soon as it is coined.

Without any paraphrasing, the conclusion of Howard’s argument would have to be expressed in this way:

I believe that the criticism of the Oxford English Dictionary for neglecting spoken English is largely mistaken.

But surely Howard’s argument is not about himself (which it would be if I were the subject of this conclusion); it is an argument about a certain criticism of the Oxford English Dictionary. The conclusion of the argument, then, should really be paraphrased thus:

The criticism of the Oxford English Dictionary for neglecting spoken English is largely mistaken.

I believe that merely expresses Howard’s attitude toward the conclusion—an attitude of belief. It seems to me that works exactly the same way with his premise. The clarified version runs as follows:

There are many other such attitude indicators, including the following:

I think that
Everybody knows that
It is clear that

The rule for paraphrasing these terms is simple: drop them in favor of simply stating what is believed, what is thought, what everybody (presumably) knows, or what is (presumably) clear.

Other terms express an attitude of disbelief. Famed biologist Ernst Mayr describes in Animal Species and Evolution how he lived with a tribe of Papuans in the mountains of New Guinea.

“These superb woodsmen had 136 names for the 137 species of birds I distinguished (confusing only two nondescript species of warblers). That Stone Age man recognizes the same entities of nature as Western university-trained scientists refutes rather decisively the claim that species are a product of the human imagination.”

Refutes rather decisively the claim that is the attitude indicator, indicating an attitude of disbelief. Others include:

I deny that
I doubt that
It is wrong to suppose that

Drop these terms in favor of the negation of the statement that is disbelieved, refuted, denied, doubted, or wrongly supposed—that is, in favor of the disbelieved statement with it is not the case that or it is false that in front of it. The statement that Ernst Mayr disbelieves—the one that he says is decisively refuted—is:

Species are a product of the human imagination.

The negation of this statement—and the conclusion of his argument—is:

It is not the case that species are a product of the human imagination.

You could have expressed the same thing more smoothly by the statement,

Species are not a product of the human imagination.

This sort of negation is usually acceptable in your paraphrase, so long as you do it with care. But there are traps to avoid. Suppose, for example, you want to write the negation of Every student will pass the course. Here you have to be wary. Not every student will pass the course is acceptable, but Every student will not pass the course can mean something altogether different—namely, that every student will fail. If you are in any doubt, it is always safe to mechanically write It is not the case that. . . .

Note that attitude indicators, whether they indicate belief or disbelief, can reflect varying degrees of confidence in one’s attitude. Some terms—call them hedges—suggest tentativeness on the arguer’s part; these include:

Perhaps
Probably
Apparently
It seems that
I have my doubts whether
I suspect that

Other terms—call them assurances—suggest confidence on the arguer’s part; these include:

Certainly
Everybody knows that
Clearly
Obviously
I categorically deny that

Phrases like I believe that can be either assurances or hedges, depending on whether we stress the I or the believe.^[2]

Often an argument is expressed as a report of someone else’s attitude, an attitude that the reporter may or may not agree with. In Fads and Fallacies in the Name of Science, for example, Martin Gardner writes:

In essence, Gall and his disciples argued that human personality consisted of a number of independent, inborn mental “faculties,” each of which was localized in a part of the brain. The larger the size of each region, the stronger the faculty. Consequently, an examination of skull bumps would reveal a person’s character.

There is an argument in this passage; but the writer is reporting it, not arguing it. The phrase Gall and his disciples argued that is the report indicator, the term that shows the argument is being reported by, but is not necessarily embraced by, the speaker or writer. It too should be eliminated from the clarification. Here is a brief clarification of the explicit portion of the argument:

Finally, inference indicators should be eliminated from your final clarification. Terms such as because and thus are very useful in distinguishing between premises and conclusions, but they are no longer useful once the argument is in standard clarifying format. The format itself clearly indicates which statements are premises and which are conclusions.

4.2.4 Eliminating Repetition

Repeating part of an argument in the same or different words is often rhetorically effective. But it has no use in your clarification of the argument. When there is such repetition, pick the best alternative—or make one up—and say it once. Pick the alternative that is clearest, most loyal, and most charitable.

In The Natural History of Nonsense, for example, Bergen Evans writes,

The civilized man has a moral obligation to be skeptical, to demand the credentials of all statements that claim to be facts, for a refusal to come to an unjustified conclusion is an element in an honest man’s religion.

His argument can be simply clarified:

This paraphrase eliminates from the conclusion the clause to demand the credentials of all statements that claim to be facts, since it seems to amount to another way of saying to be skeptical. The two clauses do not mean exactly the same thing, but nothing important seems to turn on the difference. You must rely on the context in deciding whether the statement or phrase is being repeated in different words for rhetorical effect or whether the arguer really has a different point to make. In most cases, it is easy to tell.

There is an exception to the eliminate repetition guideline. If an argument repeats—perhaps in disguised form—a premise as the conclusion, then you must repeat it in your clarification. As already mentioned in Chapter 1, these arguments usually commit the fallacy of begging the question.

4.2.5 Eliminating Wordiness

In the early 1800s a leading logic text sagely warned, “A very long discussion is one of the most effective veils of Fallacy; . . . a Fallacy which when stated barely . . . would not deceive a child, may deceive half the world if diluted in a quarto volume.”

This has been adapted to the realm of corporate annual reports to shareholders and dubbed the windbag postulate. “The chairman’s verbosity,” say the authors of an article in one business magazine, “increases in direct proportion to the severity of the company’s problems.”

Usually the diluting material should have never been there in the first place. When this is the case, says philosopher Max Black, “The abuse involved is a kind of converse of the emperor’s clothes—too many clothes and no emperor.” We can thank famed sociologist Thorstein Veblen for unintentionally providing this marvelous example in The Theory of the Leisure Class:

In an increasing proportion as time goes on, the anthropomorphic cult, with its code of devout observances, suffers a progressive disintegration through the stress of economic exigencies and the decay of the system of status. As this disintegration proceeds, there come to be associated and blended with the devout attitude certain other motives and impulses that are not always of an anthropomorphic origin, nor traceable to the habit of personal subservience. Not all of these subsidiary impulses that blend with the bait of devoutness in the later devotional life are altogether congruous with the devout attitude or with the anthropomorphic apprehension of sequence of phenomena. Their origin being not the same, their action upon the scheme of devout life is also not in the same direction. In many ways they traverse the underlying norm of subservience of vicarious life to which the code of devout observances and the ecclesiastical and sacerdotal institutions are to be traced as their substantial basis.

H.L. Mencken offers up a nifty paraphrase of this passage (probably sacrificing some loyalty for the sake of humor) remarking that Veblen’s real point is this:

Many people go to church not because they are afraid of the devil but because they enjoy the music and like to look at stained glass, the potted lilies, and the reverend pastor.

Mencken adds this, for good measure:

This highly profound and highly original observation might have been made on a postage stamp, thereby saving a good deal of wasted paper.

Eliminate wordiness, save a tree.

Extra words do not necessarily mean bad writing. Note this gem by Robert Herrick:

Come, let us go while we are in our prime,
And take the harmless folly of the time.
We shall grow old apace, and die
Before we know our liberty.
Our life is short, and our days run
As fast away as does the sun;
And as a vapor or a drop of rain
Once lost, can ne’er be found again;
So when or you or I are made
A fable, song, or fleeting shade,
All love, all liking, all delight
Lies drowned with us in endless night.
Then while time serves, and we are but decaying,
Come, my Corinna, come, let’s go a-Maying.

There is an argument of sorts in this verse, directed to Corinna. Herrick surely could have put it much more succinctly. Its most straightforward paraphrase is something like this:

Of course, Herrick’s goal is art. Ours is a certain sort of clarity—the sort that will better allow us to tell whether statements are true or false and whether they are logically connected to one another. The result is not always pretty.

4.3 Neutralizing Slanted Language

Statements convey information. Depending on the words chosen for the job, they can also influence the emotional reaction to that information. When Henry Louis Gates applied to Yale in 1969, he wrote, “My grandfather was colored, my father was Negro, and I am black.” Twenty-five years later, in the preface of his Colored People: A Memoir, he addresses his two daughters:

In your lifetimes, I suspect, you will go from being African Americans, to “people of color,” to being, once again, “colored people.” (The linguistic trend toward condensation is strong.) I don’t mind any of the names myself. But I have to confess that I like “colored” best, maybe because when I hear the word, I hear it in my mother’s voice and in the sepia tones of my childhood.

There is no substantial difference in the information that each of the terms communicates; the major difference is in the emotional reaction it is likely to arouse. A skillful writer or speaker chooses language with this in mind—dressing approved statements in positive language and disapproved statements in negative.

Influencing emotions is an important and legitimate use of language. But it can sometimes cause trouble in arguments by illegitimately slanting the argument—that is, by unjustifiably aiming it toward the audience’s emotions rather than the audience’s reason.

Janet Novack, for example, argues in a Forbes article against including self-esteem studies as a part of the elementary school curriculum. She describes such studies as among the “new politically correct fads” that are pushing out more traditional subjects because of the need to satisfy “do-gooders who want their touchy-feely subject taught.”

This description certainly communicates information: liberals favor self-esteem studies (they are “politically correct”); there is much current interest in them (they’re “fads”); they are promoted by those with a concern for society (the “do-gooders”); and they emphasize relationships (they are “touchy-feely”). But this information by itself doesn’t seem to be enough to establish that self-esteem studies have no place in elementary schools.

Such a negative conclusion is nevertheless hard to resist, due to the negative emotions aroused by Novack’s choice of terms. Her terms suggest—though probably do not logically imply—the following about self-esteem studies: they are based on dogmatism (they are “politically correct”); they will not stand the test of time (they are “fads”); they are promoted by those who do not consider the broader consequences of their actions (the “do-gooders”); and they are mindless (they are “touchy-feely”).

Not only do these terms sway the emotions, but also no argument is given for the legitimacy of a negative emotional reaction, and success in winning us over to the conclusion depends, at least in part, on that reaction. Thus it counts as slanted language. Because such language can impede honest critical reflection about the information, it is best to eliminate it from your clarification of the argument.

It is often difficult and sometimes impossible to find a completely neutral way of communicating information. We saw, for example, that those with a concern for society is more neutral than do-gooders. And so it is, but the phrase those with a concern for society seems to have a slant to it—but now in the positive direction. In such cases, my best advice is to prefer the more neutral language even if it is not perfectly neutral.

Sometimes slanted language is blatant. An Associated Press item reports that, in a poll regarding the American presence in Lebanon,

A razor-thin 48%–47% plurality thought that “the Marines should be withdrawn.” But in the same poll, by a 55%–39% margin, they agreed with the contention that “after being occupied for a long time by Syria, Israel, and the PLO, Lebanon needs help from the U.S., including the presence of U.S. Marines to help Lebanon re-establish control of its own country.”

The two claims in quotation marks are substantially the same. But at least 7 percent of those polled had a different answer when the terse statement about withdrawal of troops was replaced by more sympathetic—and more slanted—words about Lebanon’s need for help.

But in other cases the slant can be remarkably subtle. Consider the following scenario described by Kevin McKean in Discover:

Threatened by a superior enemy force, the general faces a dilemma. His intelligence officers say his soldiers will be caught in an ambush in which 600 of them will die unless he leads them to safety by one of two available routes. If he takes the first route, 200 soldiers will be saved. If he takes the second, there is a one-third chance that 600 soldiers will be saved and a two-thirds chance that none will be saved. Which route should he take?

Researchers have found that most people favor the first route, reasoning that it’s better to save those lives that can be saved than to gamble when the odds favor even higher losses. But what about the following scenario?

The general again has to choose between two escape routes. But this time his aides tell him that if he takes the first, 400 soldiers will die. If he takes the second, there is a one-third chance that no soldiers will die, and a two-thirds chance that 600 soldiers will die. Which route should he take?

In this case, most people favor the second route. The first route involves the certain death of 400 men. At least with the second route there is a one-third chance that no one will be killed. And even if the general loses this gamble, his casualties will be only 50 percent higher.

As you may have noticed, these are not two different scenarios but two descriptions of the same scenario. Researchers find that even when people realize this, they often continue to recommend contradictory actions for the general. It just sounds smart to do something that will certainly save 200 men and stupid to do something that will certainly sacrifice 400—even if the two, in the end, amount to the same thing.

4.4 Summary of Chapter Four

Streamlining is a very important part of the clarification process. Be sure to paraphrase as statements all nonstatements that function as part of the argument. These include imperatives when they serve as the conclusion to practical reasoning, rhetorical questions, and fragments. You should also strive to eliminate all unnecessary words. These include discounts, attitude indicators, report indicators, inference indicators, repetition, and any sort of nonessential wordiness. Finally, you should neutralize slanted language.

4.5 Guidelines for Chapter Four

• Include only statements in your clarification. If a premise or conclusion is expressed as a partial sentence or a nondeclarative sentence, paraphrase it as a statement.
• Exclude discounts from your clarification of the argument.
• Use discount indicator terms to identify discounts, making sure it is a discount rather than a negatively stated premise or conclusion.
• Eliminate attitude indicators, which are indicators of belief or of disbelief; insert a negation at the beginning of disbelieved statements. Also eliminate report indicators and inference indicators.
• Eliminate repetition of statements and phrases that do not add to the argument; use whatever expression best captures the arguer’s meaning.
• Eliminate wordiness that does not contribute to the argument’s clarity.
• When slanted language has not been argued for, and when there is more neutral wording available, eliminate slanted language in favor of more emotionally neutral language.

4.6 Glossary for Chapter Four

Attitude indicator—indicates the arguer’s attitude of either belief or disbelief toward a statement. I believe that and I deny that are examples.

Discount indicator—indicates a nonsupportive, or adversarial, relationship between statements rather than a supportive one. But and despite are examples. Sometimes also called adversative.

Discount—a statement the arguer offers as not undermining the conclusion. Discounts are typically neither premises nor conclusions and should usually not be included in your clarification.

Report indicator—shows that the argument is being reported by, but is not necessarily embraced by, the speaker or writer. So-and-so argues that is an example.

Slanting—unjustifiably pointing a premise or conclusion toward the emotions rather than the reason of the audience.

Streamlining—removal of nonessential features of the argument so they will not get in the way of the evaluation process. It is one important aspect of the paraphrasing procedure.