The Laws of Probability and Bureaucratic Style

The author is assistant to the Deputy Attorney General of the United States and a farmer teacher of English composition.

Whatever the cause, it is indisputable that bureaucrats have a deep suspicion of the way things appear. This suspicion manifests itself in the way bureaucrats write, in their reluctance to say that something either is or is not something else. Rather than take flat positions on the existence or nonexistence of facts or the identity of propositions with one another, bureaucrats prefer to remind their readers that they are dealing only in appearances, behind which realities may or may not lurk, or at most in probabilities, not certainties.

An interesting example of this phenomenon crossed my desk recently, and I offer it here not as a good or a bad example, but rather as a typical example, for it rests midway between the possible extremes. The sentence in question reads as follows: “Nevertheless, it would seem probable that reorganization would permit a significant reduction in total expenditures involved.” A close analysis of this sentence reveals that it contains a quintuple hedge, a considerable accomplishment for a sentence I know to have been written in haste, but far from a full exploitation of the possibilities for nonstatement contained in the sentence. Before exploring these possibilities, let us analyze the sentence as written.

A direct statement of the proposition would have been: “Reorganization will save money.” Watch how the insertion of the hedges dilutes the certainty of that statement. First, compromise the prediction by changing the fact predicted to a mere possibility. Thus, instead of saying that reorganization “will save” money, say that it “will permit a saving.” This is a beautiful bureaucratic hedge. Even if the saving does not actually come about, the writer can always remind us that he predicted only the possibility, not the actuality, and the possibility would have ripened into actuality had not someone else bungled the job.

The next step is to introduce doubt by changing the indicative “will permit” to the subjunctive “would permit.” Then give the statement an illusory quality by subordinating it to the clause “it seems that.” Next, hedge the “it seems” by changing it to “it would seem.” Finally, raise a question of improbability by calling it probable. The result is the quintuple hedge: “Nevertheless, it would (4) seem (3) probable (5) that reorganization would (2) permit (1) a significant reduction in total expenditures involved.” Because I am devoting my attention to just one sentence, I will not explore the implications of the “nevertheless” which opens the sentence other than to point out that it makes the whole sentence a hedge against the preceding sentence, which had said that another proposition was “likely.”

A quintuple hedge in a sixteenword sentence is a remarkable achievement on a words-per-hedge basis, but alas, the experienced, dedicated hedger has never viewed brevity as a desirable goal. Indeed, in the very sentence under examination, the writer could have lowered his hedge frequency from once every 3.2 words to once every 2 words by using the shortcut I used above and abbreviating “a significant reduction in total expenditures involved” to “a saving” or “savings.” His sentence then would read: “Nevertheless, it would seem probable that reorganization would permit savings.” The choice of the longer locution shows us that our writer, in addition to being a talented amateur hedger, adheres to the principle of Addition of Obscurity by Multiplication of Verbiage, to which we will return later.

What would a professional hedger, his heart set on deliberate nonstatement, have done with this sentence?

One ploy overlooked by our writer was the double negative. I refer not to the classical double negative, used by Chaucer for emphasis but condemned by contemporary grammarians, but the modern double negative, used by contemporary writers for de-emphasis and regarded as a hallmark of gentility. Thus, “probable” becomes “not improbable” and “significant” becomes “not insignificant.” The effect of the double negative in grammar is slightly different from its effect in arithmetic. Canceling equal numbers in arithmetic leaves what is left in the condition it would have been in if the canceled numbers had never been there. In grammar, however, the cancellation itself still appears in the sentence and has a subtle weakening effect on everything around it. Other aspects of the double negative to be explored by someone less bewildered by them than I lie in what we may call the “near-miss double negative,” exemplified by the clause “it is no more than probable,” which means, of course, the same thing as “it is no less than probable.”

Parenthetically, we should note the mysterious appearance of the word “significant” in our sentence. Should we attribute its presence to a slight residue of bona fide directness in our writer, or to a sophisticated realization that he has so successfully destroyed meaning in his sentence that he can taunt us by pretending there is still something significant in it? Either would explain his choice of “significant” over “not insignificant.”

The writer of the sentence under consideration has also overlooked the possibilities of nonstatement inherent in the multiplication of illusion. He has been content to say “it would seem” without sufficiently warning his readers that he is dealing in appearances only. The simplest thing to do is to add the word “apparently,” but greater complexities of nonmeaning can be achieved by using the pollsters’ technique — that is, by showing tendencies or trends. With a little imagination, we can change “it would seem” to “everything would apparently tend to indicate.” Some may object that substituting “everything” for “it,” a pronoun without antecedent, comes dangerously close to restoring meaning to the sentence. But just as pantheism is a disguised form of atheism, using “everything” in a context where nothing has been specified is a disguised way of saying nothing.

“Everything would apparently tend to indicate” is another example of the Addition of Obscurity by Multiplication of Verbiage. In application, this principle means that you never use a word where you can use ten, never use a word where you can use a phrase, never use a phrase where you can use a clause. A related principle is Subtraction of Clarity by Division of Verbiage. Somewhat more difficult in application, this means that you never let the parts of a grammatical element, like the subject and predicate of a clause, stand next to each other. The last clause of the preceding sentence is a good example.

One final way of destroying commitment to meaning in a sentence is to put verbs in the passive rather than the active voice. This has the twofold effect of making the sentence run against the natural flow of the English language, and of removing responsibility from any active agent in the sentence. People suffer rather than do in the passive voice. Of course, in the sentence in question, no one was doing anything anyway, but the conversion to the passive will still sap what little strength the sentence had.

What do we end up with? Something like this: “Nevertheless, it would seem, or at least everything would apparently tend to indicate, that it is not improbable that a not insignificant reduction in the total expenditures which would be involved would possibly be permitted by reorganization, at least to some extent.”

Although he follows rhetorical rules with mathematical names, the basic purpose of the writer of nonstatements is horticultural. That is, if he has enough hedge below him to cushion his fall, he’s not afraid to go out on a limb.