To understand that English has developed not just via new words but also through the emergence of new grammar puts in a new light a notion about language you may have heard about.

One of the most popular ideas is that a language’s grammar and the way its words pattern reflect aspects of its speakers’ culture and the way they think. Countless times I have witnessed the hush in a classroom when introducing undergraduates to this hypothesis. If one doesn’t pick this up in college, one will catch it in newspaper and magazine articles about indigenous groups, or even in bits of folk wisdom floating around. One sometimes hears that Iran is home to a uniquely vigorous homosexual subculture because its third person pronoun is the same for men and women.

This idea that grammar is thought became influential from the writings of Edward Sapir. We met him in the previous chapter venturing that English speakers came to find nuance irritating. Even that point had hints of the language-is-thought persuasion—supposedly the erosion of various aspects of English grammar was due to some psychological leaning in its speakers. But Sapir ventured only passing speculations in this vein.

It was Sapir’s student Benjamin Lee Whorf who picked up the ball and ran with it, in the 1930s, publishing several pieces on the subject which served as its foundational texts. The hypothesis is known, therefore, as the Sapir-Whorf hypothesis.

The hypothesis has also failed. Repeatedly and conclusively.9

Decade after decade, no one has turned up anything showing that grammar marches with culture and thought in the way that the Sapir-Whorf hypothesis claimed. At best, there are some shards of evidence that language affects thought patterns in subtle ways, which do not remotely approach the claims of Whorf.

Yet the Sapir-Whorf idea is cited enthusiastically in textbooks even today, and is a favorite approach to language by journalists. In 2004 a New York Times writer supposed that the language of the Kawesqar tribe in Chile has no future tense marking because, having been nomads traveling often in canoes in the past, they would usually have been so unclear on what was going to happen in the future that there was no need to ever talk about it (!). Never mind that Japanese has no future markers either, and yet the Japanese hardly seem unconcerned with the future. The point is that this Times writer would not have even floated such a notion if it weren’t for the seed planted by Whorf’s work seven decades previously. Whorf, even though he died in 1941, lent us a meme.

However, with an awareness of how languages actually come to be the way they are, we are in a position to truly understand how hopeless the Sapir-Whorf hypothesis is. The idea that our take on the world is mediated by refraction through our grammar, such that the world’s six thousand languages generate six thousand correspondent world views, is deeply appealing. It is also mistaken.

A School Is Founded

After all, is the way I think shared with all of the other Anglophones of the world—or even just all Americans—and reflected in the language I am writing in right now? Was the change from Old English grammar to Modern English grammar—not vocabulary—determined or even partially affected by the transformation of England from feudalism to industrial capitalism?

The answers to both questions would have to be yes, from the way Whorf wrote. His pièce de résistance was an observation about the language of the Hopi: that it does not mark time in any way. He argued that this made Hopi speakers think in a way completely different from us Westerners, with our persnickety obsession with past, present, and future. The Hopis, he argued, think of time as cyclical, to the extent that they even have a concept of “time” as an ongoing process in the way that we do.

Grammars do differ in what concepts they choose to mark. Spanish marks gender on nouns. Japanese does not, but it has markers showing whether a noun is a subject or object. All grammars mark some things; no grammar marks everything. Whorf’s idea was that which things a grammar happens to mark determines what its speakers perceive most readily in their daily lives:

Users of markedly different grammars are pointed by the grammars toward different types of observations and different evaluations of externally similar acts of observation, and hence are not equivalent as observers but must arrive at somewhat different views of the world.

Whorf, like many of his followers, was not quite clear as to whether he thought that grammar, once accidentally morphing into certain patterns, channeled culture, or that culture determined how grammar morphed. Presumably, there was a “dynamic” two-way relationship. But his basic point was the correspondence between grammar and thought patterns, and hence culture.

Therefore, Western scientific advances presumably correspond to our languages’ rich tense marking: “Newtonian space, time, and matter are no intuitions. They are recepts from culture and language. That is where Newton got them.” This is why, therefore, it was not Native Americans who gave the world theoretical physics.

Whorf, as it happened, was a fire insurance inspector by day, and perhaps it was partly because of this that he did not know Hopi very well. Quite simply, Hopi has as much equipment for placing events in time as any language. Here is Start sharpening your arrows; we’re going hunting:

Um angwu pay ùuhoy tsuku-toyna-ni;

you beforehand already your arrow make-a-point-will

itam maq-to-ni.

we hunt-go-will

Hopi renders the statement as something like You’ll have sharpened your arrows, then we will go hunting. And in doing so, we see two indications of pastness, and a thoroughly typical future marker. Hopi does not render the sentence as something like Sharpen your arrows, as our hunting occurs in the cycle of time. Yet Whorf’s claim about Hopi was quite explicit; i.e., that Hopi has “no words, grammatical forms, constructions, or expressions that refer directly to what we call ‘time, or to past, or future, or to enduring or lasting.”

In other words, Whorf was just wrong—and yet without the zest of his writings on Hopi, it is likely that the Sapir-Whorf hypothesis would not have caught on at all.

Of course, Whorf hedged a bit, admitting that grammar did not utterly prevent a person from being able to think about or talk about things the grammar did not explicitly mark. Rather, a grammar makes it much easier to think of and talk about some things than others:

The potential range of perception and thought is probably pretty much the same for all men. However, we would be immobilized if we tried to notice, report, and think of all possible discriminations in experience at each moment of our lives. Most of the time we rely on the discriminations to which our language is geared, on what Sapir termed “grooves of habitual expression.”

But the general tenor of Whorf’s writings reveals little interest in the potential and an ardently promulgated obsession with the habitual:

We cut nature up, organize it into concepts, and ascribe significances as we do, largely because we are parties to an agreement to organize it in this way—an agreement that holds throughout our speech community and is codified in the patterns of our language. The agreement is, of course, an implicit and unstated one, but its terms are absolutely obligatory; we cannot talk at all except by subscribing to the organization and classification of data which the agreement decrees.

The emphasis above is Whorf’s, and his writing is sometimes almost narcoticized with his fascination with the exotic inner world of the Hopi as channeled through their purportedly tenseless tongue:

It might be said that the linguistic background of Hopi thought equips it to recognize naturally that force manifests not as motion or velocity, but as cumulation or acceleration. Our linguistic background tends to hinder us in this same recognition, for having legitimately conceived force to be that which produces change, we then think of change by our linguistic metaphorical analog, motion, instead of by a pure motionless changingness concept, i.e. accumulation or acceleration.

It isn’t hard to see why so many smart people from the thirties on have thrilled to this notion, especially couched in such eloquent phrasing. Whorf was also a mesmerizing speaker, and a looker to boot. Yet because the foundational presentation was founded on sand and no one has since found any further confirmation elsewhere, it is dismaying to see how deeply the idea has permeated educated thought nevertheless.

Are We Dumb Anglo-Saxons?

Try squaring the Sapir-Whorf hypothesis with, for example, the fact that today’s English was once Old English. As we have seen, the language I am writing in was not planted in one fell swoop from on high. Modern English is the current stage of what began as a very different grammar, much like German’s. Over a millennium-and-a-half, this grammar had grammatical features from Celtic plugged into it Botox-style, while also being radically shorn of its complexities liposuction-style by adults learning it as a second language.

Now, let’s try to look at this language we speak as it is today through the eyes of the Sapir-Whorf hypothesis. We English speakers “cut nature up, organize it into concepts, and ascribe significances as we do, largely because we are parties to an agreement to organize it in this way” and “cannot talk at all except by subscribing to the organization and classification of data which the agreement decrees.” Or, as an early Whorfian put it:

The thought of the individual must run along its grooves; but these grooves, themselves, are a heritage from individuals who laid them down in an unconscious effort to express their attitude toward the world. Grammar contains in crystallized form the accumulated and accumulating experience, the Weltanschauung of a people.

The problem is that according to this logic, the differences between Old English and Modern English grammar must reflect differences between Anglo-Saxon culture and life in modern New York or London.

So first of all, we now have meaningless do. Unlike Old English speakers, we have to say I do not walk and Do I walk? Thus, presumably, since Anglo-Saxon days, English speakers have become especially alert to negativity and uncertainty, such that we have to stress verbs with do in the relevant types of sentence. Obviously that makes no sense—it’s simply that we snapped up meaningless do from Welsh and Cornish because speakers of those languages were learning English.

Now, as it happens, Whorf was not unaware that languages are always picking things up from one another, noting, “It is clear that linguistic determinism and linguistic relativity cannot be absolute in the face of the known facts of linguistic change, multilingualism, and cultural diffusion.”

But even then, note what he wrote next: “People do make new discriminations and find linguistic expressions for them, often by borrowing.” It sounds as if Whorf thought that speakers of a language took in features from another language only if those features provided a way of expressing the “new discriminations” they had fallen into.

That is, meaningless do made its way into English because English speakers for some reason had become uniquely alert to negation and questionhood and were ripe for some nearby language to give them a way to vent this alertness? But what about how in South Africa, Xhosa and Zulu picked up click sounds from Khoisan languages? Was it because they had drifted into a latent desire to make such sounds and the Khoisan languages just happened to fulfill the need? Or was it because, well, the languages spoken next door happened to have clicks in them? Clearly, drifting into “new discriminations” is not a precondition for taking a feature from another language. The precondition is, simply, proximity.

And then, what about how English got easier over the centuries? According to the Sapir-Whorf hypothesis, compared to Old English speakers, Modern English speakers are dimwits.

Since we no longer classify our nouns by gender, we are less sensitive to masculinity and femininity as abstract concepts. This implies that the fashion in late-twentieth-century academia for treating gender as a societal construct had actually penetrated the world view of Joe Barstool (Tavernstool?) as early as A.D. 1200.

We are, apparently, less alert to the fact that anger, remembrance, error, fear, and shame are things that we ourselves feel. Old English speakers “feared themselves” when they felt afraid just as they behaved themselves. These days, apparently, there is always a part of an Anglophone that supposes that, just maybe, his feelings are experienced by someone else.

And then, never mind that when we see a car coming, the fact that it is moving to here (hither) rather than already here is less vivid to us than it was vivid to an Anglo-Saxon farmer that a carriage was coming toward him rather than resting in front of him. Old English “cut nature up” in a way that rendered those farmers more aware of movement than we must be, with our one-size-fits-all here, there, and where.

We are also less clear on the difference between the immediate context of people we are talking to in the moment and people in the abstract. The Old English speaker had the pronoun man (actually pronounced “mon” in the way that we today associate with Anglophone Caribbeans) to refer to an abstract “they”—Man says that the language is getting easier—or “one”—Man speaks Old English here! But even though the typical Old English speaker spent his life in a village whereas the typical Modern English speaker has read newspapers, traveled some, and today has broadband, the Old English speaker’s grammar rendered him more cosmopolitan than we are, more aware of a world beyond his own head.

And forget our processing that when we have e-mailed something, an action has been performed while when we have left, a state has arisen in which we are gone. When using the perfect, Old English speakers used be instead of have, with a bunch of verbs that referred more to how things ended up than an event happening. Apparently to us today, “states, schmates”—everything is an action.

In the several-decades’-deep literature on the Sapir-Whorf hypothesis, I am aware of no address of what its implications would be for how a language has changed over time. This is a serious problem, because if there is anything really interesting to the hypothesis, then surely we seek meaningful correlations between the culture of Anglo-Saxons, who spoke a grammar like German’s, and the culture of Modern English speakers, who speak a language a little like Welsh and a lot like nothing else.

I venture that no scholar will see it as promising to investigate whether Modern English speakers are psychologically less alert to the nuance of daily experience than Anglo-Saxon villagers. And to the extent that Whorfians object that it is a two-way street and culture can also affect grammar, we wonder what it was about England becoming a literate, industrialized society that would have encouraged a simpler grammar.

The disconnection between cultural development and grammar is also clear in that societies have turned upside down over time while the grammar stayed put. As psychologist Herbert Clark has put it, if the Sapir-Whorf hypothesis “has any force historically, we should find examples of beliefs failing to change over time because of the conventions that exist in the language. Such examples do not readily come to mind.”

Right—take Russia. All would agree that certain changes have occurred in prevailing beliefs in that country over the past thousand years—from brute feudalism under the tsars to Communism to glasnost to the queer blend of democracy and dictatorship of today. Yet Russian grammar during that time has always been the marvelous nightmare that it is now. Russian has changed, to be sure, but without equivalents to the Celtic adoption and the Viking disruption, and nowhere near as dramatically as English—and in no ways that could be correlated with things Peter the Great, the Romanovs, or Lenin did.

Whorfianism as Zeitgeist: Thinking People’s Street Myth

Whenever you read someone making reference to the Whorfian perspective on language, whoever the writer picked it up from has never had occasion to address just why Modern English speakers are not rather troglodytic creatures compared to Anglo-Saxon-speaking warriors.

In general, there is curiously little interest among people fascinated by the Whorfian paradigm in examining whether a given grammatical trait conditions a certain way of thinking not just in one language, but in other ones. As towering a mind as literary critic Edmund Wilson, for example, thought the reason Russians seemed unable to keep to a schedule was that Russian is a language where future tense is indicated largely via context—but then Japanese is like that, too, and the Japanese have never seemed to have any problem with schedules.

Journalist Mark Abley, engaging writer though he is, falls into this trap in his enthusiasm for Whorfianism. In French and many other Western European languages, there are two words for know: savoir means to know a fact; connaître means to know a person or to be familiar with something. Abley has it that:

My language allows me, somewhat clumsily, to get the distinction across: on the one hand, factual knowledge; on the other, acquaintanceship and understanding. But to a French speaker, that distinction is central to how the mind interacts with the world.

Really? Is Abley really so sure that the difference between knowing the capital of Nebraska and knowing a friend is more immediate to Gérard Depardieu than to Judi Dench? It’s a cute idea, yes—but does Abley actually have any grounds for supposing that it is true?

How does it sound when it’s French that has one word where English has more, and when it isn’t something as immediately evident as the European know verbs? In French, sortir means “go out,” but also covers what English would express with come out (in the earthquake, le tiroir est sorti de la commode, “the drawer came out of the dresser”), get out (someone is in a hole and says, “Sors-moi d’ici!” “Get me out of here!”), and stick out as in one’s tongue (“Sors la langue,” “Stick out your tongue”).

So—are we English speakers more attuned than French speakers to the difference between leaving home, something slipping out of place, being yanked out of a hole, and sticking out our tongues? I would venture that the answer is no. To be a reasoning representative of Homo sapiens is to understand those four processes as radically different, whether or not your language happens to have the same word for them. The same applies to how your language happens to mark knowing.

Abley also shows us that the Boro language of India has verbs with charmingly specific meanings. The implication is that to speak Boro is to be uniquely attuned to these highly particular concepts, such as:

egthu: when people getting to know one another start to establish a sense of comfort and connection

onsay: musky bodily odor, especially that emanating from the armpits, of a kind not ideal but vaguely pleasant

goblo: when a romantic pair have been estranged for a long period and decide to be together again

khonsay: to have sex for the first time with someone you are in a romantic relationship with

asusu: when a member of a couple stays always a vigilant foot or so away from the other member at a social occasion

Interesting that a culture would choose those highly particular aspects of experience to assign words to. Or is it?—I actually dissembled there. The language with words for those concepts is good old English; namely, bonding, funk, reconciliation, consummate, and hover.

Upon which now we can take a look at what the Boro words actually mean:

egthu: to create a pinching sensation in the armpit

khonsay: to pick an object up with care as it is rare or scarce

onsay: to pretend to love

goblo: to be fat (as a child or infant)

asusu: to feel unknown and uneasy in a new place

Abley’s idea is that to speak Boro is to be uniquely attuned to these concepts. However, when speakers of a language are asked what a word means, quite often they give particular uses that happen to be especially common, rather than the larger concept the word technically covers. For example, if someone asked you what consummate meant, you would likely give the sexual meaning, although you technically know that consummate means, more generally, “to bring to the highest level.” “What’s bonding?” someone asks you. You might say “When things stick together.” But you also might say “When you first feel a click with someone, like guys bonding over sports.”

This is surely a lot of what is behind the Boro verbs. After all, English has ways of expressing many of those concepts. “To create a pinching sensation in the armpit” can be expressed in English as cinch up into—and is egthu in Boro used exclusively with armpits, or was that what the consulted speaker most readily mentioned? To pick an object up with care in English can be to pluck it out. We have no verb for “to be fat as a child” but we have a noun, baby fat, which refers to exactly what the Boro word does, except not as a verb. The issue is not what part of speech people happen to express a concept in, but whether their language “feels” it. Well, on babies’ fat, English feels it, as do quite certainly all languages on earth.

In the same way, we have no verb like asusu for not feeling at home, but we have positive adjectives like acclimated and situatedI wasn’t situated yet and so I was still calling home every night. English speakers are attuned to the same mental state that Boro speakers are when they asusu.

And even where Boro really does have a word marking a fine shade of human experience that English does not—I draw a blank on an English equivalent to “pretend to love”—it still doesn’t follow that this experience is more deeply felt by them than the rest of us.

Looking at our own language is an especially effective way of truly getting this. In English, something in spot four is fourth, in spot seven is seventh, in spot eight is eighth, and so on. Only the first three numbers are distorted in a major way: first and second don’t correspond to one and two at all, and third clearly has three in there, but beaten up a bit, and what’s with the -rd? There’s no sixrd or tenrd.

Well, that’s something else weird about English and European languages. Most of the world’s languages have a special word for the first spot, like first, but then just say, as it were, “two-th,” “three-th.” So English’s second, Spanish’s segundo, and Russian’s vtoroj (when two is dva) mean that these languages channel our European language speakers’ thoughts into a heightened awareness of secondness, I suppose. That is, an English, Spanish, or Russian speaker is more sensitive to things being second than a German, a Turk, an Inuit, or an Israeli . . . Come on. We just happen to have a distinct word marking secondness; the Boro just happen to have a word for pretending to love.

Politics or Science?

Among academics and beyond, the Sapir-Whorf hypothesis has been, quite commonly, less examined than embraced. One of the reasons: what interests many about the hypothesis is less what it would imply for academic issues about human psychology than its demonstration that indigenous cultures are not “primitive,” and in fact may have some things on us.

This was an explicit mission of Sapir, and an invaluable one in itself. It is to him and like-minded thinkers of his time such as his mentor anthropologist Franz Boas that it is part of the warp and woof of modern Westerners to view other cultures as variations on being human rather than “savages.” Gone are the days when America could stampede into the Philippines as it did during the McKinley administration, casually assuming that the “natives” needed to be “civilized.”

Whorf inherited the diversity imperative from Sapir, and it permeates his writings on Hopi. To Whorf, Hopi and the world view it supposedly conditioned was not just different, but better:

Does the Hopi language show here a higher plane of thinking, a more rational analysis of situations, than our vaunted English? Of course it does. In this field and in various others, English compared to Hopi is like a bludgeon compared to a rapier.

We Westerners are obsessed with putting things into little boxes, drawing boundaries—the Hopi, however, are more in touch with higher realities:

Our objectified view of time is, however, favorable to historicity and to everything connected with the keeping of records, while the Hopi view is unfavorable thereto. The latter is too subtle, complex, and ever-developing, supplying no ready-made answer to the question of when “one” event ends and “another” begins.

The problem with this kind of thing is that too often it ends up, in essence, taking us back to the noble savage. Noble, to be sure, but in what we celebrate in them as special, savage—clothed chimpanzees, cute.

Mark Abley, for instance, seizes upon a grammatical quirk in the Native American languages of the Algonquian family, such as Cree, Ojibwa, and the Powhatan that Pocahontas spoke. In one of them, Montagnais, the way you say You see me is:

Tshi - ua:pam - in.

you see me

But the way to say I see you is not to put I before the verb and you after. That is, reversing the example above and doing

is wrong; it is not Montagnais at all, any more than Reading book you a are is English.

In - ua:pam - tshi.

I see you

Instead, you use the You see me sentence, but stick a little syllable into it to make it mean I see you:

Tshi - ua:pam - in.

you see me

Tshi - ua:pam - it - in.

you see me = “I see you.”

So—the basic sentence is about you; only with an adjustment can you make it about I. (That is, indeed, so deliciously odd from our Anglophone perspective. Once again, languages are interesting in their grammars as well as their words.)

Abley has it that this means that Algonquian language speakers are less self-centered than Europeans, and that “to speak properly, in an Algonquian language, is to be aware of the identities and interrelationships of all the people you address.” But when we are at a Thanksgiving dinner, are we English speakers not fully aware of who is who, despite that we can put I first?

Abley marvels at the fact that Native Americans are capable of carrying on conversations among multiple participants—which is like praising a culture for cooking food or, really, being more cognitively advanced than their pets.

And in any case, just as Whorf mischaracterized Hopi, Abley leaves out that in Algonquian languages, I can indeed come first. You, if there, does have to come first, but if there is no you around and the I is interacting with a he, she, it, or they, then I has to come first. In another Algonquian language, Cree, I frighten them is:

Ni - se:kih - a - wak.

I frighten them

To say They frighten me you can’t put they first; you make they the subject by sticking in a special syllable:

Ni - se:kih - ik - wak

I frighten them = “They frighten me.”

It looks like Algonquians are just as narcissistic as we are when I am talking about them.

One episode that pointed up this fundamental commitment to ennobling The Other was the rare language-is-thought study that argued that English speakers are the more insightful ones. Alfred Bloom noted that in Chinese, one must engage in a certain amount of circumlocution to be explicit that something is hypothetical rather than real. In English we can say If you saw my sister, you would know that she was pregnant. But in Chinese, the sentence is rendered as “If you see my sister, you know she is pregnant.” For those who know Mandarin:

Rúguŏ nĭ kàn dào wo mèimei

if you see arrive I sister -
nĭ yídìng zhīdào tā huáiyùn le.

you certainly know she pregnant now

That sentence can have various meanings. One of them is neutral and not hypothetical:

“If you see my sister, you’ll know she is pregnant.”

Then there are hypothetical meanings, referring to something that has not happened or did not happen:

“If you saw my sister, you’d know she was pregnant.”

“If you had seen my sister, you’d have known she was pregnant.”

In Mandarin, context determines which meaning comes through.

Bloom did an experiment that showed Chinese speakers less alert to hypotheticality when reading stories in Chinese than English speakers reading the stories in English. On the basis of this, he supposed that since where people’s grammar is concerned, “the thought of the individual must run along its grooves” as the Whorfian I quoted above had it, Mandarin’s grooves must distract thought from the difference between reality and the hypothetical. What’s good for a perceptively challenged Modern English speaker is good for the man on the street in Beijing, right?

Apparently not: people shot at Bloom like he was a varmint. Their objections to details of his experimental procedure were reasonable, but more conclusive was their insistence that Chinese speakers process hypotheticality via context even if their grammar does not mark it as explicitly as English’s. Elsewhere, however, there is little interest in noting that, say, English speakers understand via context that knowing algebra is different from knowing the man next door, or that even if Hopi did have no tense markers, we could assume that its speakers processed that things happen before and after one another as vividly as we do. I feel reasonably confident in surmising that if Bloom’s study had shown some interpretational deficit among English speakers, no one would have batted an eye.

A speaker of American Sign Language captured the essence of how Whorfianism unintentionally demeans minority languages, mocking outsider fans of Sign. In an interview, the signer feigned “a vapid, rapt look on his face. ‘Sign language is so beautiful’, he signs, in a gushing mockery of the attitude that exoticizes sign and correspondingly reduces deaf people to the status of pets, mascots. ‘It’s just so wonderful that deaf people can communicate !’ ” Or, I would have it, “It’s just so wonderful that people who aren’t like us can think and process reality as richly as we do!”

Maybe that message had a certain value in Whorf’s era. In the thirties, popular culture and common consensus in America were still shot through with pitiless condescension toward “natives,” “Chinks,” “jungle bunnies.” But it’s been a while. We clap when our infants don’t spill their food. We can afford to let go of clapping when exotic folks don’t, when in our times, celebrating diversity is a shibboleth of moral legitimacy among thinking First World people, and considerably, if not comprehensively, beyond.

All Homo sapiens engage in advanced mentation—yes, hallelujah. However, this doesn’t make the Cree speaker a paragon of enlightened selflessness because you comes earlier than I in his way of saying I see you, any more than our ability to explicitly get across If you’d have seen mysister, you’d have known she was pregnant makes us Anglophones wizards of truth versus falsity compared to people in China.

Does Language Channel Thought?: The Neo-Whorfians

At this point, one might ask: does language channel thought at all? It is pretty clear that people speaking non-Western languages are not walking around in psychedelic dreamscapes channeling essences of The Real unknown to us “straights” marching around in business suits. However, is it really true that grammar has nothing to do with the way we think?

Of course not. These days there is research being done in what is often called the Neo-Whorfian School. No more of the mystic, anti-Western hocus-pocus—this is serious psychological research, based on a reasoned curiosity as to whether grammar can channel thought, albeit in ways less dramatic than the straight-up Whorfians were seeking. And there are, indeed, twinkles of evidence in favor of the Sapir-Whorf hypothesis. Nothing mind-bending or kaleidoscopic—just twinkles.

In some cases, there is a chicken-and-egg issue that makes it hard to see the study as telling us that language channels culture rather than vice versa. The Guugu Yimithirr of Australia do not have terms like in front of. Instead, they refer to everything according to points on the compass—to the north of, to the south of—regardless of where they are in relation to the object. If a tree is in front of them, but in the global sense, it is to the south of them, they refer to it as south of them.

Neat—but is it that their language just happened not to have terms like in front of and behind and forced them to think of things in terms of compass points, or that their culture happened to focus on compass points and that determined how their language described position? Because the focus on compass points is so clearly a cultural peculiarity compared to most people on earth, it is hard to see from this experiment why grammar is a more likely explanation. More to the point, it isn’t really surprising that people do not have terms for something their culture does not care about—i.e., that observation would not have sparked a whole school of thought the way Whorf’s did.

In the same way, the Pirahã tribe of the Amazon have attracted media attention as a people whose language has no words for colors, numbers, or quantification (e.g., all, every), and does not even have relative or subordinate clauses—for That’s the guy who built my house one can only say That guy—he built my house. The Pirahã prove incapable of performing even elementary mathematical tasks, and the media first jumped on this, predictably, as proof of the old-fashioned language-makes-thought notion: because the Pirahã language has no numbers, the people are rendered incapable of doing math. Again, however, this puts the cart before the horse. The Pirahã are hunter-gatherers who subsist on their own. Their lives therefore afford them no reason to manipulate precise numerical concepts, such as for trade or constructing elaborate architectural monuments. In addition, the Pirahã are an unusually incurious people (no, really, they are—consult the source in the Notes on Sources), which makes them especially uninterested in fine-grained manipulation of numerals.

A natural outcome, then, would be that their language would have no words for numbers. The magical idea that language is the issue—i.e., that they would be doing algebra if only it weren’t for the mysterious happenstance that their language has no numbers—may have more visceral oomph, but little else. The Pirahã’s chronicler, Dan Everett, concurs, despite the media’s occasionally making it seem otherwise—it’s what the Pirahã are like that shapes what their language is like, not the other way around. I’m not sure how truly interesting that, in itself, is, despite how interesting the Pirahã are themselves, as well as their language—albeit for reasons other than their grammar purportedly channeling its speakers’ thought patterns.

More interesting are cases where culture cannot possibly be the issue. In German, the word for key is masculine (der Schlüssel). If you give the key a personal name, Germans tend to have an easier time recalling it if the name is masculine; they more readily associate the key with a picture of a man than a woman, and describe it with words like hard, heavy, jagged, metal, serrated, and useful. In Spanish, the word for key is feminine (la llave), and Spanish speakers are more comfortable with keys’ having female names, associating them with pictures of women, and they tend to describe them as golden, intricate, little, lovely, shiny and tiny. Maybe it is relevant that when I once asked a Dutch person what keys were like (key in Dutch is the masculine sleutel), she said, with a distinct animation, “I imagine a big, giant key with decorations on it like the kind that would open a castle!!” I don’t know why she pictured that precisely, but I take the liberty of assuming that the key she imagined was a fella.

So speakers of languages with gender, deep down inside, have a sense that objects are boys and girls. It is also documented that a Spanish speaker, if asked to imagine a table (la mesa) as a talking cartoon character, is likely to imagine the table’s voice being high and sweet because in their language table is feminine.

However, in real life it is very, very rare that we go about imagining inanimate objects talking at all. In general, speakers of languages that assign gender to nouns do not on an everyday basis see inanimate objects as sexed “men” and “women.” The gender class of objects is something lying deep in their psyches, which we can tease out with careful experiments. However, it has nothing to do with the immediacy of daily experience. For anyone who has been close to a speaker of a language with gender, think about it—do they give any evidence of thinking of chairs and toothbrushes as “God’s creatures,” with a sex and the traits traditionally associated with it?

Unless you have known some truly unusual people (likely fond of dropping acid), you will agree that the gender that their native grammar happens to assign to inanimate objects does not color their world view. If it does, it is in a way that would seem significant only to an academic psychologist plumbing exquisitely fine-grained niceties, hard to classify as a “world view” that would interest even the educated layman.

Studies like this show that language does have some glimmers of effect on thought. But they do not support the more dramatic implications that suck the air out of a room when the textbook version of Sapir-Whorf is brought up. They bring to mind University of California linguist Paul Kay’s pithy observation about the whole business: “If anthropologists had not assumed that the people they went out to study have ‘world views’, would they have found them?” Neo-Whorfians reveal the truth: perceptual distinctions of a subtle, slight, and subconscious nature, not “world views.”

There is nothing “cultural” about imagining dulcet-toned tables if forced to by someone supervising an experiment you signed up for, with said experiment being the only time in your life when you will ever imagine a table with the power of speech. Paul Kay and Willett Kempton, who have done research in the Neo-Whorfian tradition, readily acknowledge the simple truth: “If the differences in world view,” they write, “are to be interesting, they must be sizable. Minuscule differences are dull.”

And yet, mainstream sources continue to flag the Sapir-Whorf hypothesis in its grand old rendition as a going concern. “The significance of Whorf’s hypothesis lies less in its possible truth, and more in its continuing ability to generate thought and discussion on a problem which is central to the whole anthropological project,” Alan Barnard and Jonathan Spencer have it in the Encyclopedia of Social and Cultural Anthropology. Or, as a widely used college textbook casually recites: “The Sapir-Whorf hypothesis therefore might suggest that English speakers can’t help paying more attention to differences between males and females than do the Palaung and less than do French or Spanish speakers.”

But again, what is the value of an investigation classified as ever “might suggesting,” which inherently includes a hypothesis that Modern English is a philistine grammar, numb to the details that Old English channeled its speakers into noticing every day, such as the personal nature of emotion, or whether things are approaching us or sitting still?

The idea that the world’s six thousand languages condition six thousand different pairs of cultural glasses simply does not hold water. The truly enlightened position is that, by and large, all humans, be they Australian Aborigines, Japanese urbanites, Kalahari hunter-gatherers, Cree Indians, Serbs, Greeks, Turks, Uzbeks, Amazonians, or Manhattanites in analysis, experience life via the mental equipment shared by all members of our species. No one is “primitive,” but just as important, no one is privileged over others with a primal connection to The Real.

If you find an error or have any questions, please email us at Thank you!