Poisoning our minds

Scott Simon, watch out! When a language scold e-mails to criticize you for calling a rattlesnake poisonous -- as you did on this morning's "Weekend Edition" -- just ignore it. When you asked the rattlesnake bagger "Are they poisonous?" you were using the word correctly.

I'm issuing this preemptive counterattack because just a few weeks ago, another public radio stalwart mistakenly apologized for calling snakes poisonous. After "All Things Considered" ran a story about snake milking last month, host Robert Siegel gave air time to a letter from a listener about the same "mistake": 

My 4H Club students in the elementary schools where I give lessons on insects and arachnids would never forgive me if I did not point out that snakes and other animals that bite and inject a toxin, are venomous, not poisonous.

Now it is true that the snake's toxin is called venom. But venom is a kind of poison -- the kind produced by animals and insects -- not a substance distinct from poison. Samuel Johnson certainly knew what venom was, but in his renowned Dictionary, he referred to "a poisonous serpent" and "a poisonous insect."

A few of the Victorian gentlemen engaged in the Great Language Tidy-Up of their time did hope to completely separate poison from venom. But they had to concede that the distinction was fuzzy, given the Bible's "poison of asps" and "adders' poison."

The OED's definition of poisonous is "Containing, or of the nature of, poison; having the properties of a poison; venomous." And its examples include "teeth ... by which they ejaculate their poyson" (1661), "poisonous vipers (1665), and "poison fangs" (1796).

The only current style guide I've seen that insists on the distinction is Paul Brians's "Common Errors in English" site. He says:

Snakes and insects that inject poisonous venom into their victims are venomous, but a snake or tarantula is not itself poisonous because if you eat one it won’t poison you. A blowfish will kill you if you eat it, so it is poisonous; but it is not venomous.

Brians's formulation acknowledges that venom is a poison; his objection seems to be that "poisonous snake" or "poisonous spider" could be misunderstood as "unfit to eat." But after more than four centuries of using poisonous for venomous animals, I think we can handle the ambiguity.