A Conspiracy Theory of Connotations

Discussions of censorship often operate from the assumption that the main motivation of censors is the suppression of dissent. For that reason, critiques of censorship often attack the idea of suppression: Censorship is often counterproductive and only makes samizdat material more popular. And if an idea is systematically censored, we can never really be sure that it’s wrong, since we’ll never see a full and honest accounting of the evidence for and against it. These are good arguments against suppression, and there are plenty more where they come from.

However, the goal of suppression does not explain a lot of contemporary censorship, which aims to punish innocuous statements alleged to carry some sort of pernicious hidden message capable of changing the way people think and behave. In such instances, the censorious impulse appears to be paired with a clownishly ridiculous idea of how language and society work—a kind of conspiracy theory of connotations. Three examples of this bizarre approach have made the news in recent weeks.

According to a report in the Telegraph, Puffin books have made “hundreds of changes” to the work of late children’s author Roald Dahl, apparently with the cooperation of the Dahl estate. These include removing every instance of the word “fat,” changing “female” to “woman,” describing the Oompa-Loompas in Charlie and the Chocolate Factory as “small” rather than “tiny” (and always as “small people” rather than “small men”), changing “boys and girls” to “children,” having a character read Jane Austen rather than Rudyard Kipling, and removing words like “crazy” and “mad.” The goal here is not to suppress a point of view but to remove any mention of certain kinds of characteristics. This seems to be partly out of the desire to avoid causing unnecessary offense to parents. But the revisions also seem to have been motivated by concern that young readers might infer that these characteristics are important in some way, and build undesirable worldviews from that inference.

Last month, Stanford University was widely mocked and criticized after a guide appeared on its Elimination of Harmful Language Initiative (EHLI) website. The guide—since removed but originally intended for use by the university’s IT department—stated:

The goal of the Elimination of Harmful Language Initiative is to eliminate* many forms of harmful language, including racist, violent, and biased (e.g., disability bias, ethnic bias, ethnic slurs, gender bias, implicit bias, sexual bias) language in Stanford websites and code.

The footnote denoted by the asterisk read:

* We understand that it may not be possible to eliminate all harmful language on our sites and in our code due to costs, resources, or other reasons. “Eliminate” is a goal to strive for even if it can’t be achieved.

The 13-page guide listed over 150 allegedly harmful words and terms, and in each case provided a suggested alternative. Words like “guru,” “white paper,” and “seminal” were now to be avoided. Using “guru” apparently “negates its original value” as “a sign of respect” in “the Buddhist and Hindu traditions.” The term “white paper” apparently “assigns value connotations based on color (white = good), an act which is subconsciously racialized.” Using “seminal” apparently “reinforce[s] male-dominated language.”

The explanations offered for these changes imply an absurd theory of language. The notion that simply hearing the phrase “white paper” can lead someone to “racialize” something must have seemed plausible to whoever wrote Stanford’s EHLI guide, but it’s not at all clear what it means to “racialize” something or how this process is effected. The absurdity is mitigated a bit (I suppose) by the qualifying adverb “subconsciously,” but this also serves to emphasize the strangeness of the underlying theory. It implies that words work in some mysterious and unseen way, and that we are powerless before the influence they exert as they colonize our minds. This is the essence of conspiratorial thinking.

Similarly, it’s already a stretch to accept that “seminal” constitutes “male-dominated language,” but what does it mean to “reinforce” it (or, in academese, to “reinscribe” it)? Talk of “reinforcement” suggests that male-dominated language is a consequential factor in determining relations between the sexes, and that every time someone uses such language, they are somehow helping to prop up an archaic (and implicitly oppressive) system. But I have no idea what such reinforcement might actually entail. The only way to make any meaningful cause-and-effect sense of this is to accept that the word “seminal” has the power to convince anyone who hears it that men are better than women. In the final analysis, it requires us to believe that we are tools of words rather than vice versa.

A few days after the Stanford fiasco, the Associated Press Stylebook Twitter account posted a tweet in a similar vein which announced:

We recommend avoiding general and often dehumanizing “the” labels such as the poor, the mentally ill, the French, the disabled, the college-educated. Instead, use wording such as people with mental illnesses. And use these descriptions only when clearly relevant.

That tweet—also deleted in its original form after a great deal of derision and pushback—probably represents the nadir of this perspective on language. The AP appears to believe that the definite article—one of the most common words in written and spoken English—has the power to “dehumanize” people, but a wholly synonymous alternative formulation does not. Apart from the prima facie absurdity of this claim, using the definite article was—until five minutes ago—usually deemed to be politically correct rather than politically incorrect, at least when the word “community” was appended—“the LGBT community,” “the African-American community,” “the disabled community,” and so on.

“Wokeness” is an inchoate phenomenon. It may be that, as critics of its pejorative use allege, it is too difficult to define to be of any analytical use. But surely this backwards understanding of language, not to mention the strident and apparently arbitrary way in which it is policed, is central to what bothers people about contemporary progressive activism. The notion that our minds are somehow vulnerable to subliminal psychic attack by malevolent political forces seems to do violence to common sense and experience.

Nevertheless, this view has recently been adopted by a number of academic philosophers who maintain that much of politics is conducted using code words, usually described as “dog whistles” or “fig leaves.” Last year, the feminist philosopher Kate Manne wrote the following in the New York Times:

We praise arguments for being muscular and compact and criticize prose for being flabby, flowery and, implicitly, feminine. When it comes to our metaphysics—our pictures of the world—we pride ourselves on a taste for austerity, or as W.V.O. Quine put it, ‘desert landscapes.’ And what is the fat body in the popular imagination but excess, lavishness, redundancy?

Critics of this view pointed out that a bad argument might be described as “thin” (and indeed a good argument might be described as “meaty”), but there are more problems here than the specific examples. Manne’s view is that someone will hear “muscular argument,” infer that being “muscular” is good, and therefore become fatphobic—or more fatphobic—without any intervening reflection. We are, in other words, simply puppets manipulated by other people’s linguistic choices.

Some academics even want politics to be conducted by explicitly redefining words to fit political preconceptions. One of the most prominent redefiners is recent MacArthur Genius Grant recipient Ibram X. Kendi, who has devoted considerable energy to popularising a more expansive definition of the word “racism.” If the conspiracy theory of language fails to convince us that words are capable of harming people, potential harm can still be alleged because the meanings of words that describe harm have become excessively broad.

Against this strategy stands a conviction that the world comes first, and then our feelings about it, and that our language only follows after. The dispute about language is therefore a dispute about the direction of causation. The new censoriousness is informed by the idea that our problems begin with language, which determines how people feel about things, which in turn determines behavior and outcomes.

A characteristic example of this tendency is the introduction of euphemisms to replace negatively-charged language. But this can result in what Steven Pinker has called a “euphemism treadmill,” as formerly neutral terms acquire negative connotations and have to be replaced until they become tainted and have to be replaced in their turn, and so on and so forth. In a 1994 New York Times article, Pinker wrote that this process “shows that concepts, not words, are in charge: give a concept a new name, and the name becomes colored by the concept; the concept does not become freshened by the name.”

Similarly, when we broaden a name to include more concepts, its connotation changes. An accusation of racism, for example, has become—as many commentators warned it might—“watered down” or “diluted” by the attempts to redefine it. The counterproductive result is that an accusation of racism has been devalued by expansive and promiscuous misuse and now carries less force than it once did.

The theory exemplified by the Stanford IT and AP Stylebook instructions is magical thinking in the most fundamental sense. It holds that we can change the world merely by speaking incantations. If this were true, it would certainly make life easier. It is much simpler to enforce speech codes than it is to grapple with the intractable problems and trade-offs involved in the ordinary business of politics.

But the backlash against language policing, and the fluidity of language itself, show that political life is never that simple. There have always been progressive bureaucrats who evince an illiberal desire to stamp out disagreement, but the Great Awokening is also a story about how entire swaths of the intelligentsia—despairing of, or simply too self-important for, everyday political action—have instead placed their faith in sorcery.