From “English and Nootka” by Benjamin Lee Whorf in “Reading About Language” edited by Charlton Laird, pp. 64-66.

The Indo-European languages and many others give great prominence to a type of sentence having two parts, each part built around a class of word —substantives and verbs—which those languages treat differently in grammar. As I showed in the April 1940 Review, this distinction is not drawn from nature; it is just a result of the fact that every tongue must have some kind of structure, and those tongues have made a go of exploiting this kind. The Greeks, especially Aristotle, built up this contrast and made it a law of reason. Since then, the contrast has been stated in logic in many different ways: subject and predicate, actor and action, things and relations between things, objects and their attributes, quantities and operations. And, pursuant again to grammar, the notion became ingrained that one of these classes of entities can exist in its own right but that the verb class cannot exist without an entity of the other class, the "thing" class, as a peg to hang on. "Embodiment is necessary," the watchword of this ideology, is seldom strongly questioned. Yet the whole trend of modern physics, with its emphasis on "the field," is an implicit questioning of this ideology. This contrast crops up in our mathematics as two kinds of symbols—the kind like a, 2, x, and the kind like +, —, /, though in view of 0, 1/2, pi, and others, perhaps no strict two-group classification holds. The two-group notion, however, is always present at the back of the thinking, although often not overtly expressed.

Our Indian languages show that with a suitable grammar we may have intelligent sentences that cannot be broken into subjects and predicates. Any attempted breakup is a breakup of some English translation or paraphrase of the sentence, not of the Indian sentence itself. We might as well try to decompose a certain synthetic resin into Celluloid and whiting because the resin can be imitated with Celluloid and whiting.

When we come to Nootka, the sentence without subject or predicate is the only type. The term "predication" is used, but it means "sentence." Nootka has no parts of speech; the simplest utterance is a sentence, treating of some event or event-complex. Long sentences are sentences of sentences (complex sentences), not just sentences of words. In “tl'imshiya'isita'itlma” we have a simple Nootka sentence. The translation, "he invites people to a feast," splits into subject and predicate. Not so the native sentence. It begins with the event of "boiling or cooking" tl'imsh; then comes -ya ("result") = "cooked"; then -‘is "eating" = "eating cooked food", then -ita ("those who do") = "caters of cooked food; then -‘itl ("going for"); then -ma, sign of third-person indicative, giving tl'imshiya'isita'itlma, which answers to the crude paraphrase. "he, or somebody, goes for (invites) eaters of cooked food."

We are constantly reading into nature fictional acting entities, simply because our verbs must have substantives in front of them. We have to say "It flashed" or "A light dashed," setting up an actor, "it" or "light," to perform what we call an action, "to flash." Yet the flashing and the light are one and the same! The Hopi language reports the flash with a simple verb, rehpi: "flash (occurred)." There is no division into subject and predicate, not even a suffix like -t of Latin tona-t "it thunders." Hopi can and does have verbs without subjects, a fact which may give that tongue potentialities, probably never to be developed, as a logical system for understanding some aspects of the universe. Undoubtedly modern science, strongly reflecting western Indo-European tongues, often does as we all do, sees actions and forces where it sometimes might he better to see states. On the other hand, ‘state’ is a noun, and as such it enjoys the superior prestige traditionally attaching to the subject or thing class; therefore science is exceedingly ready to speak of states if permitted to manipulate the concept like a noun. Perhaps, in place of the ‘states’ of an atom or a dividing cell, it would be better if we could manipulate as readily a more verblike concept but without the concealed premises of actor and action.