I’ve been holding off writing a post about Benjamin Lee Whorf thinking that I would wait until I had time to do the topic justice; but the announcement (more here) of a new article in Science has caused a flurry of posts around the internet, and I felt that it was an opportune time to make a simple point: Whorf never said that language determines thought.

It would be interesting to examine why people feel the need to recast Whorf’s argument in such essentialist terms. I think it is for one of two reasons: (a) Some people like to argue that culture doesn’t matter, and Whorf seems like a suitable straw man for an equally essentialist argument. Or, (b) they want to make an essentialist argument about culture (how words are untranslatable, or why some cultures are superior to others, etc.) and they think that Whorf will give justification to their argument.

What most people never do is actually readwhat Whorf wrote. It is abundantly clear that the unnamed author of this Science Daily press release (published by Columbia University Teachers College) did not read Whorf. They start the article with the following statement:

During the late 1930s, amateur linguist Benjamin Lee Whorf posed the theory that language can determine the nature and content of thought. But are there concepts in one culture that people of another culture simply cannot understand because their language has no words for it?

Lets put aside for now the derisive remark that Whorf was an “amateur,” and ask if this accurately reflects his theory. We can learn a lot by looking at how this statement differs from that attributed to Dr. Peter Gordon, the author of the article which prompted this press release:

“Whorf says that language divides the world into different categories,” Gordon said. “Whether one language chooses to distinguish one thing versus another affects how an individual perceives reality.”

Now, I haven’t read Dr. Gordon’s paper, but this much closer to Whorf’s actual position than that of the press release author. The press release version of Whorf’s theory states that language “determines” thought, while Dr. Gordon’s version simply states that linguistic “categories“”affect” our “perception” of reality. Morover, he does not say that concepts are “untranslatable”.

I suppose it isn’t as exciting a theory when put this way, it sounds almost … obvious. After all, if linguistic categories didn’t have any affect at all on our perceptions, then language wouldn’t matter at all. But Gordon’s version is still a little off. Whorf wasn’t really talking about our “perception of reality.” Whorf’s point was that some cultural differences in behavior where linked to conceptual differences arising from linguistic analogies. That is to say, it wasn’t so much that we are locked into thinking about the world a certain way because of our language, but we have a tendency to do so – and this tendency has an effect on our cultural behavior.

In order to make his point clear, Whorf drew examples from his experience working as “a fire prevention engineer (inspector) for the Hartford Fire Insurance Company.” He discussed how

the cue to a certain line of behavior is often given by the analogies of the linguistic formula in which the situation is spoken of, and by which to some degree it is analyzed, classified, and allotted its place in) that world which is “to a large extent unconsciously built up on the language habits of the group.” And we always assume that the linguistic, analysis made by our group reflects reality better than it does.

He gives the example of workers smoking by “empty gasoline drums” (emphasis added):

Thus, around a storage of what are called “gasoline drums,” behavior will tend to a certain type, that is, great care will be exercised; while around a storage of what are called “empty gasoline drums” it will tend to be different-careless, with little repression of smoking or of tossing cigarette stubs about. Yet the “empty” drums are perhaps the more dangerous, since they contain explosive vapor. Physically the situation is hazardous, but the linguistic analysis according to regular analogy must employ the word “empty,” which inevitably suggests lack of hazard. The word ‘empty’ is used in two linguistic patterns: (1) as a virtual synonym for “null and void, negative, inert,” (2) applied in analysis of physical situations without regard to, e.g., vapor, liquid vestiges, or stray rubbish, in the container. The situation is named in one pattern (2) and the name is then “acted out” or “lived up to” in another (1), this being a general formula for the linguistic conditioning of behavior into hazardous forms.

That is to say, as a result of the analogy we make between “empty” and “null and void” we behave as if “empty gasoline drums” are safer than full ones, even though the reverse is the case.

Now, the bulk of Whorf’s argument is not about specific words, but around the grammatical categories of Hopi language. From his “amateur” research into Hopi he discovered that they grammatically classify time differently than does the English language. This is because for us time is a countable “individual noun,” like books, CDs, and cars. That is as opposed to “mass nouns” which are not immediately countable, but need to be first broken up into units. For example, we can’t say “three waters” but must say “three cups of water.” Water is an example of a mass noun. Now in Hopi time is not an individual noun (like books), but nor is it a mass noun (like water). That is because Hopi has no mass nouns.

Since nouns are individual already, they are not individualized by either type-bodies or names of containers, if there is no special need to emphasize shape or container. The noun itself implies a suitable type-body or container.

As a result their way of dealing with cyclical time is different from in English:

So for the phase nouns we have made a formless item, “time.” We have made it by using “a time,” i.e. an occasion or a phase, in the pattern of a mass noun, just as from “a summer” we make “summer” in the pattern of a mass noun. Thus with our binomial formula we can say and think “a moment of time, a second of time, a year of time.” Let me again point out that the pattern is simply that of “a bottle of milk” or “a piece of cheese.” Thus we are assisted to imagine that “a summer” actually contains or consists of such-and-such a quantity of “time.”

In Hopi however all phase terms, like “summer, morning,” etc., are not nouns but a kind of adverb, to use the nearest SAE analogy. They are a formal part of speech by themselves, distinct from nouns, verbs, and even other Hopi “adverbs.” … One does not say “it’s a hot summer” or “summer is hot”; summer is not hot, summer is only WHEN conditions are hot, WHEN heat occurs. One does not say “THIS summer,” but “summer now” or “summer recently.” There is no objectification, as a region, an extent, a quantity, of the subjective duration feeling. Nothing is suggested about time except the perpetual “getting later” of it. And so there is no basis here for a formless item answering to our “time.”

The heart of Whorf’s argument is that the Hopi practice of conducting rituals today to affect the events of tomorrow derives from the analogies drawn from grammatical differences in how English and Hopi deal with the cyclical aspects of time.

Hopi “preparing” activities again show a result of their linguistic thought background in an emphasis on persistence and constant insistent repetition. A sense of the cumulative value of innumerable small momenta is dulled by an objectified, spatialized view of time like ours, enhanced by a way of thinking close to the subjective awareness of duration, of the ceaseless “latering” of events. To us, for whom time is a motion on a space, unvarying repetition seems to scatter its force along a row of units of that space, and be wasted. To the Hopi, for whom time is not a motion but a “getting later” of everything that has ever been done, unvarying repetition is not wasted but accumulated. It is storing up an invisible change that holds over into later events.11 As we have seen, it is as if the return of the day were felt as the return of the same person, a little older but with all the impresses of yesterday, not as “another day,” i.e. like an entirely different person. This principle joined with that of thought-power and with traits of general Pueblo culture is expressed in the theory of the Hopi ceremonial dance for furthering rain and crops, as well as in its short, piston-like tread, repeated thousands of times, hour after hour.

This seems extremely sensible to me. Does it mean that Hopi “can’t understand time”? Of course not. It just means that grammatical differences have had an influence on cultural differences. No more, no less.

UPDATE: Thanks to a comment by Tim May I fixed my discussion. I knew I was missing a step in the argument, but hadn’t spent the time re-reading Whorf before posting. I had mistakenly asserted that Hopi treated time as a “mass noun”, when – in fact – it has no mass nouns. This should be clearer now.

UPDATE: More discussion here, and here. Of specific relevance to this post is Mark Liberman’s complaint that people are “leaving Edward Sapir out of it.”