Language Log

How long have we been avoiding the passive, and why?

A few days back, the Senior Writers' Lounge at Language Log Plaza was enlivened by an exchange about the passive voice in English.  Poser relayed a query about where the injunction against the passive originated.  Nunberg fixed on George Orwell's 1946 article "Politics and the English Language", where Orwell firmly instructs us: "Never use the passive where you can use the active."  I demurred, noting that the injunction was a commonplace in college writing handbooks in the 30s and 40s (in the U.S., anyway); and now I'm ready to show some of the evidence.

So Orwell isn't the originator.  But it's likely that his very influential essay brought Avoid Passive to a much larger audience than it had before; no doubt Strunk and White's equally influential Elements of Style (1st ed. 1959) helped spread the word in the U.S.  Eventually, Avoid Passive becomes a central element in the ideology of English writing style.

But where DID it originate? 


Fowler (1926) shows no animus against the passive, nor do the great American grammar ranters of the late 19th century, Richard Grant White and Alfred Ayres.  Hall's (1917) survey of disputed usages doesn't mention Avoid Passive or anything like it.  Something seems to have happened (possibly only in the U.S.) in the first two decades of the 20th century -- or maybe I'm just looking in the wrong places.  Stay tuned for further developments.

In any case, in the 30s we see handbooks characterizing the passive voice as a weakness to be avoided.  From Foerster & Steadman's Writing and Thinking: A Handbook of Composition and Revision (1931:380):


And from Kierzek's Macmillan Handbook of English (1939:65):


Ah, you will have been struck by how similar these passages are, even though they come from different authors and different publishers (Houghton Mifflin and Macmillan, respectively).  Such similarities are all over the place in the world of English handbooks.  Writers base their texts on the ones they learned from themselves, a fact that results in some continuity over generations.  Publishers want their texts to be authoritative and standard (we are, after all, talking about the "facts" about English, right?), and of course everybody reads the competition, so editors work to align their books with others, adding or revising material on their own, or farming the writing out to others (who may work for different publishers on different occasions).  All of this ensures a remarkable sameness to the texts. 

(It's not just handbooks of English, of course.  All kinds of textbooks are shaped by the same forces.  Some of you will have noticed the recent flap about American history texts, as reported, for example, in Diana Jean Schemo's "Schoolbooks Are Given F's In Originality", in the New York Times of 7/13/06, p. A1.)

Every so often, an author will perceive some infelicity in student writing which hasn't (so far as the author knows) yet been catalogued, and a new injunction will find its way into a handbook.  And probably then propagate to other handbooks.  This is quite likely what happened with Avoid Passive.  There are writers who are fond, even overfond, of the passive; they could use a warning.  A teacher or editor confronted with such a student naturally wants this warning to be couched not just as a critique of a particular sentence, but as general advice that can applied to future writing; as a result, the advice tends to get rigidified into a blanket injunction.  Which can then spread.

There are fashions in these proscriptions.  Some of them, like the passionate contempt for speaker-oriented hopefully voiced in some circles, seem inexplicable to me.  Others gain some power from associations with other bits of ideology, both linguistic and non-linguistic.  Look at Avoid Passive in this light.

The formulations above object to the passive on two grounds.  There's an esthetic objection: passives are felt to be less smooth, interesting, and natural than actives.  These are judgments of linguistic taste, much like the musical judgment that Beethoven's symphonies are more interesting and moving than Mozart's (I am not espousing that judgment myself, by the way).  But tastes differ, notoriously, and there are contexts in which a reasonable person might find a passive to be smoother, more interesting, and more natural than an active.  Advocates of Avoid Passive are then telling writers, "We have better linguistic taste than you do; learn to think the way we do."  (In fact, "Do as I say, not as I do", since advocates of Avoid Passive use passives themselves with some frequency.)

The second objection to the passive is via an appeal to metaphorical values attached to the two voices: active as energetic, strong, and emphatic, passive as inert, weak, and reserved.  Now, it might well be that these metaphorical connections spring in part from taking the grammatical terminology "passive" and "active" rather too literally.  But they certainly arise from the (significantly mistaken) beliefs that sentences describe actions, that the subjects of active clauses denote the agents or doers of these actions, and that the subjects of passive clauses denote the receivers or recipients of these actions.  (These beliefs are part of a traditional, but seriously inadequate, conceptual framework for grammar in which syntactic concepts are, for the most part, collapsed with semantic ones.)

Next, note the unspoken (and questionable) assumption here that energetic action, strength, and emphasis (rather than reserve) are unalloyed goods. These are conventionally taken to be MASCULINE qualities, so that a bit of linguistic ideology (about the values of the active vs. the passive) plugs into a bit of non-linguistic ideology (about the values of the masculine vs. the feminine).  It's possible that the valuing of the active voice over the passive, and the more general favoring of a spare and forceful writing style in the early 20th century, is connected to wider social values of the time; but I'm not enough of a cultural historian to speak to the point.

Orwell's main objection to the passive seems to have been that it is wordy, a criticism we'll soon see from another source.  I've always found this objection baffling, because in real life most passive clauses are agentless (the first sentence of this posting is of a relatively rare type), so that they're generally SHORTER than their active counterparts.  Consider an active clause with subject X and direct object Y:  X V Y.  Its agentless passive counterpart is of the form:  Y be Ved.  If X is n words long, the passive is shorter than the active by n-1 words (or n words in some contexts).  When I wrote, above, "in the 30s we see handbooks characterizing the passive voice as a weakness" I could have saved a word by using the passive: "in the 30s we see the passive voice characterized as a weakness". 

Yes, I know, Orwell and the other wordiness critics are thinking about agentive passive clauses (with "by X" in them).  But most passive clauses that might be problematic are agentless.

Something that both Foerster & Steadman and Kierzek get more or less right, but tends to be downplayed in later advice about the passive, is that how "important" the referent of Y is plays a role in choosing the voice for a clause.  Unfortunately, they don't really have the vocabulary to talk about what's going on here, so I doubt that their talk about importance is at all comprehensible to their readers.

One piece of it pretty much everybody is clear on: use the (agentless) passive when the referent of X is unknown or irrelevant, or (at your peril) when you want to conceal that information.  But passives, including agentive passives, are also good when the referent of Y is topical in the discourse, old information, foregrounded, etc.  Passives (in which Y appears as subject) are good for this because subjects tend to denote things that are topical in the discourse, old information, etc.  There's now quite a considerable literature on the discourse functions of subjects and on related matters.  What's important here is that, in the face of what's known about discourse organization, a blanket prohibition against passives whenever an an active version is available is just bad advice.

But now for the really good stuff, from Jensen, Schmitz & Thoma's Modern Composition and Rhetoric (2nd ed., 1941; the 1st ed. was in 1935), which devotes two pages to the "passive style", starting on p. 437 (I've bold-faced my favorite parts, and flagged a bunch of specific points).  JST start by deploring wordiness but then shift into a hymn to action:

There's a lot more, in which passive-style variants are judged "less direct and less simple", while active-style variants are labeled "more forceful and more vigorous".  JST conclude (p. 439) with a stern rebuke to those inclined to the passive style:


If the index is to believed, this is ALL that JST have to say about passive clauses in the 654 pages of their book: Just Say No.  Papa H. would be proud.

Actually, it turns out that Hemingway only occasionally led with a punch.  Of the chapters in my 1938 Scribner's edition of the short stories, roughly half begin with stative sentences, half with what could (very) generously be labeled as sentences denoting one or more activities.  One of the hard-to-classify first sentences (in "A Way You'll Never Be") even has a passive in it:


[Note: Several people have suggested to me that overt opposition to passives is much less strong in the U.K. than in the U.S., and that in fact passives are more frequent in formal speaking and writing in the U.K. than in the U.S.  I have no evidence about either claim (though I do have a renewed appreciation of just how hard it is to count passives).]

zwicky at-sign csli period stanford period edu

Posted by Arnold Zwicky at July 22, 2006 08:11 PM