On “Passive Aggressive” and Other Wrong-Headed Advice of Strunk & White’s Style Manual

Sep 9, 2015 by

ElementsOfStyleThis post is prompted by a recent Facebook discussion about the “50 Years of Stupid Grammar Advice” article, published by Geoffrey K. Pullum in The Chronicle of Higher Education in April 2009. Although I typically comment on more recent articles, the heated discussion of this essay and some specific comments that were made about it reveal that the issue is far from settled (or even properly understood by many people). Overall, I agree with Pullum’s dictum that The Elements of Style by William Strunk and E.B. White “has not improved American students’ grasp of English grammar; it has significantly degraded it”. In this post, I challenge three objections to Pullum’s critique that came up in the Facebook discussion.

The first objection concerns the purported style/grammar distinction. One of the discussants wrote that The Elements of Style “outlines basic style for writing “The elements”. Once you master the basics, you may progress.The problem with this objection is that it puts the proverbial cart in front of the horse: grammar is not some advanced calculus that one may or may not master at some advanced stage of one’s education. Rather, grammar is a glue that holds words together. Any competent speaker of a language (which you certainly are, unless you are Tarzan or a toddler) “does grammar” every day, every time words are strung together into utterances. Style, in language as in art or fashion, is the ability to speak and write—that is, to “do grammar”—effectively and with flair. If grammar is like mortar that keeps the bricks of language (aka words) together, one cannot build a beautiful, “stylish” brick building, whatever style—Gothic, Baroque, or modernist—it may be, without laying the bricks and mortar properly. Grammar, therefore, necessarily precedes style. In fact, we acquire most of the grammar of our mother tongue before we go to school in the first place. What our grammar teachers in school and writing coaches who pen books such as The Elements of Style can and should do is develop the students’ explicit awareness of the inner workings of English grammar, much like their explicit awareness of natural laws is developed by science teachers. Then and only then is it useful to talk about “style”, or how to speak and write in ways that are appropriate to the circumstances, effective in communicating one’s ideas, and appealing to the intended addressees.

A second and related objection (from the same commentator) is that The Elements of Style is not a book about grammar at all but one that is concerned purely with stylistic matters. He writes: “Does the author of this [= Pullum] not understand what a style guide is? He may bemoan the use of Elements of Style as a grammar text and he is very right in that. It is not intended as a grammar primer.” This is not a fitting objection because, despite its title, much of the “advice” in The Elements of Style concerns grammar. It operates with such grammatical terms as “noun”, “verb”, and “adjective” and purports to give advice about the proper use of such grammatical structures as “passive” (vs. active) and “positive form” (vs. negative form). The authors, fine writers that they may be, do not have a good grasp of these concepts or how they play out in English. Some of their suggestions about these issues are completely made up out of thin air and others are based on a mistaken idea that English grammar can and should be understood on the basis of Latin grammar. Consider, for instance, Strunk & White’s advice that statements should be “put in positive form”. Obviously, negation imparts meanings and therefore it cannot always be avoided in favor of rephrasing the sentence “in positive form”. Even double negatives are sometimes necessarily to convey a meaning that a “positive form” does not deliver. (Double negation in historical and regional forms of English is discussed in more detail in my earlier post.)

The assumption that the English grammar works (or should work) just as the Latin grammar did is also wrong-headed. Although English and Latin are distantly related, English grammar (and Germanic grammar, more generally) is quite different from that of Latin, which was a far more synthetic language than English is. One example of such ill-considered advice that resulted from molding the English grammar to that of Latin is the prohibition against split infinitives. As discussed in my earlier posts (see here and here), in Latin, the infinitive form of the verb is marked by a suffix which is an integral part of the verb word, whereas in English, the infinitive marker to is, like other verbal markers such as auxiliaries and modals, an independent word rather than a morpheme attached to the verbal root. In Latin, inserting an adverb between the verbal root and the infinitive marker would involve splitting a word, which is impossible, whereas in English doing so involves splitting a phrase, which is perfectly fine. Even if splitting infinitives is not, strictly speaking ungrammatical, in English, is it perhaps a sign of “bad style”? That view can be adopted only if one thinks that Shakespeare, Lord Byron, Robert Burns, Daniel Defoe, Benjamin Franklin, William Wordsworth, Abraham Lincoln, George Eliot, and Henry James were all writers with “bad style”: they have all used split infinitives. (The prohibition against split infinitives became a staple of style manuals only in the 20th century.) Moreover, rather than being a bad choice under any circumstances, splitting an infinitive may have a stylistic effect, that of de-emphasizing the adverb: consider Pullum’s example of The dean’s statements tend to completely polarize the faculty and The dean’s statements tend to polarize the faculty completely. But rather than explain the subtleties of their usage, Strunk & White outlaw split infinitives altogether.

Another example of Strunk & White’s advice that is more harmful than helpful is their “war on passive”. Instead of defining the concept of “passive” clearly and explaining when passive structures are or are not effective, Strunk & White are “passive aggressive” about this grammatical construction, exhibiting negative attitudes and passive resistance to this very common phenomenon of the English language. The passive construction is evidently part of the English grammar’s repertoire; otherwise, nobody would have been using such structures, and Strunk & White would have had nothing to object to in the first place. Sometimes, passive is stylistically awkward and sometimes it is stylistically necessary: for example, what politician would prefer I will raise taxes instead of a noncommittal statement Taxes will be raised? Opponents of passive would claim that avoiding passive would force one to avoid vagueness or noncommittal (a good thing, in their view!), but that is not true either. Sentences such as Someone will raise taxes or They will raise taxes are just as evasive as the passive, although they employ active voice. (In many languages, including Spanish and Russian, one can avoid specifying the agent of an action by simply omitting the subject altogether, as in Podnimut nalogi, literally ‘will-raise taxes’, with the verb in the 3rd person plural form but no overt subject.)

A bigger problem with The Elements of Style, as Pullum points out, is that its authors do not have a good grasp as to what is or is not a “passive”: two out of three of their alleged examples of passives actually do not contain any passive. Sure, as one of the Facebook discussants pointed out, “…”he left college because his health was impaired.” sounds better than “The reason that he left college was that his health became impaired.”” But the problem with the latter sentence is not passive but its unnecessary verbosity (which may actually be advantageous in some contexts). As Pullum points out in his essay, the offending sentence does not contain any passive; impaired is a past participle being used here as an adjective, not a part of a passive structure. Moreover, become does not appear with passives (as also noted by Pullum), but it happens to be one of the most reliable tests for adjectives.

The bigger problem is that, instead of using structural criteria for concepts like “passive”, Strunk & White and their ilk all too often rely on semantic (i.e. meaning-related) criteria for defining grammatical concepts. This approach to grammar is completely wrong-headed as the same meanings can often be conveyed by differing grammatical means. But as a result of such “style advice”, as Pullum writes, typical “college graduates today … equate the grammatical notion of being passive with the semantic one of not specifying the agent of an action”. Yet, “agent-less sentences” is a semantic category that is too broad to map to a single grammatical construction. Not specifying the agent of an action happens not only with passives but also when the sentence involves one of a special sort of intransitive verbs, called “unaccusative”. (Note also that not specifying the agent cannot be the entire function of passives, as they exist also in languages like Spanish and Russian, which, as discussed above in connection with the Podnimut nalogi example, allow to not specify the agent by simply omitting the subject.) An unaccusative verb is found in Pullum’s example A bus exploded (which, he writes, many students would analyze as passive “because it doesn’t say whether terrorists did it”). Similarly, The boat sunk contains an unaccusative verb rather than a passive construction. Compare it with the sentence The boat was sunk. The presence of the auxiliary be (here, was) is not the only difference between these two sentences. There is a subtle semantic difference between them: the passive but not the unaccusative version implies an agent of an action. In contrast, unaccusative verbs such as exploded (in Pullum’s example), sunk (in The boat sunk), or melted (as in: The chocolate melted) denote changes of state that occur without an external causer, from within. This semantic property of unaccusative verbs correlates with their incompatibility with constructions that require an implied agent, such as a purpose clause or a purpose adverb. Hence, sentences such as *The boat sunk in order to collect the insurance or *The chocolate purposefully melted are ungrammatical (or semantically incongruous, depending on one’s analysis). Passives, in contrast, are compatible with purpose clauses and adverbs like intentionally or purposefully: witness The chocolate was purposefully melted (by the chef) in order to mix it into the batter. This sentence is perfectly fine with or without an explicit expression of the agent in the form of a by-phrase (here, by the chef).

One subclass of unaccusative verbs includes verbs of existence, such as exist, remain, thrive, and of course the locative be. (For a detail discussion of various subclasses of unaccusative verbs across languages, see Levin & Rappaport Hovav 1995; different types of be are discussed in Pereltsvaig 2001 and references cited therein.) Like the change-of-state unaccusative verbs discussed above, verbs of existence do not entail an agent, yet examples like There were a great number of dead leaves lying on the ground are not passive structures (as Pullum notes, citing the example from Strunk & White). Nor is it any more “passive” than (the less verbose) A great number of dead leaves were lying on the ground, which Strunk & White appear to advocate. The latter structure is not always preferable to the existential variant with there: consider Waiter! There’s a fly in my soup and A fly is in my soup. At least to me, the latter sounds more awkward than the former.

The choice between there- and bare existential sentences depends in large part on the nature of the noun phrase (here, a fly): as discussed by Pullum, phrases denoting new information in English (and cross-linguistically) tend to come later rather than sooner, while phrases denoting old information come sooner rather than later. (In English this distinction correlates, albeit imperfectly, with the indefinite/definite distinction.) Thus, Waiter! There’s a fly in my soup can be followed by The fly is at the bottom of the bowl, but the reverse (i.e. using a definite noun phrase in a there-sentence) is not grammatical: *A fly is in my soup. There is the fly at the bottom of the bowl. The same principle affects the choice between active and passive. For example, in a context where we are talking about a certain girl, the passive The girl was kissed by a boy is a better choice than its active counterpart, A boy kissed the girl.

A third and last objection to Pullum’s critique that I will discuss here is that his criticism is anachronistic. Another fan of Strunk & White’s manual wrote in the Facebook discussion:

“There are a lot of landmark things from long ago that are basically useless or reversed by now. That doesn’t mean they weren’t once significant. I found Strunk & White useful once upon a time; that I wouldn’t recommend it now doesn’t negate the value it once had for me. The dude seriously needs to let it go.”

This objection, however, is also wrong: advice dished out by Strunk & White was as ill-considered, internally inconsistent, unhelpful, and even harmful to the students from half a century ago, when the book was first published, as it is today. The only way that The Elements of Style could have been useful, let alone significant, in the 1950s is if English had changed in the relevant respects between then and now. That is not the case, however: as Pullum’s examples and the discussion in my earlier posts (linked to above) amply shows, people—including respected, even beloved writers—used split infinitives, passives, double negatives, and other “no-no”s according to Strunk & White’s for centuries. “Style gurus” such as Strunk & White and their predecessors (and, sadly, many that followed them) continue to prohibit such uses, labeling them as “clumsy”, “illogical”, “confusing”, and “monstrous”—seemingly, to no effect. Ironically, such prescriptivists themselves commit the errors they warn their readers about, as Strunk & White do by violating four of their own dictums in one sentence, “contravening “positive form” and “active voice” and “nouns and verbs””, as well as separating a relative clause from the noun which it describes (see Pullum’s article for a detailed discussion). Similarly, Robert Lowth, an 18th-century grammarian violated his own prohibition against stranding a preposition—while discussing how wrong it is to do so:

“The Preposition is often separated from the Relative which it governs and joined the verb at the end of the Sentence … as, ‘Horace is an author, whom I am much delighted with.’ … This is an Idiom which our language is strongly inclined to; it prevails in common conversations, and suits very well with the familiar style if writing; but the placing of the Preposition before the Relative is more graceful, as well as more perspicuous; and agrees much better with the solemn and elevated style.” (emphasis mine)

Such slip-ups underscore a larger point: it is impossible to change the grammar of a language by decree. Language just does not work that way. No matter how “clumsy”, “illogical”, “confusing”, or “monstrous” a given grammatical structure may be labeled today, there is no telling if it will not only survive but prevail in the language of tomorrow. For example, this is exactly what happened with passive progressives. Samuel Pepys wrote in his diary: “I met a dead corpse of the plague, just carrying down a little pair of stairs…” (emphasis mine). Today, this sentence appears so confusingly wrong to us that you may be surprised to find that “style gurus” of Pepys’ day castigated the then-new being carried down and its ilk, yet the latter has eventually vanquished the old structure.

Thus, while prescriptivists assault this or that grammatical construction, language continues to develop according to its internal logic, which is all too poorly understood by those attempting to fight it. The only thing defeated by such “style gurus” is the sanity of those who are hoping to learn from their advice. Pullum puts it so well that I would like to close with a passage from his essay:

“It’s sad. Several generations of college students learned their grammar from the uninformed bossiness of Strunk and White, and the result is a nation of educated people who know they feel vaguely anxious and insecure whenever they write “however” or “than me” or “was” or “which,” but can’t tell you why. The land of the free in the grip of The Elements of Style.”

Freeing generations of English speakers from the tyranny of “idiosyncratic bumblers who can’t even tell when they’ve broken their own misbegotten rules” and freeing room for better-informed writing instruction is long overdue.




Levin, Beth, and Malka Rappaport Hovav (1995) Unaccusativity: At the Syntax-Lexical Semantics Interface. Linguistic Inquiry Monograph. Cambridge, MA: MIT Press.

Pereltsvaig, Asya (2001) On the Nature of Intra-Clausal Relations: A Study of Copular Sentences in Russian and Italian. Ph.D. dissertation, McGill University.



Related Posts

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:

  • Maybe the biggest problem with the “you have to teach the basics first” camp is that so many people view The Elements of Style as the be-all and end-all of composition instruction. It can’t be both a style primer and the only style book you’ll ever need.

    • That’s a great point, Jonathon!

      The more I think about it, the more I wonder how can scholars who advocate critical thinking, argumentation and analysis in their fields be so uncritical, obstinate and unreasonable about writing. If I were to ask the same people who are fans of Strunk & White and their ilk if it’s okay to preach something that is not true, never was true, has no factual support of any kind, and is just handed down from someone who isn’t even an expert in that area—just because that’s how their teachers taught it that way—I think I’d get an overwhelming “no”. But when it comes to “writing style”, it seems to be okay. I wonder if it’s some sort of subconscious “victim syndrome”, same as victims of domestic violence tend to later become perpetrators of it…

  • nachasz

    Another instance of molding grammar of modern language on Latin one can find in Polish. Nie is usually descibed as a negative particle, though considering placement of stress in negated verbs, it seems to be rather negative prefix.

    • That’s interesting. Can you please give me more details? Can anything come between “nie” and the verb, for example?

      • nachasz

        Can anything come between “nie” and the verb, for example?

        I was going to answer that no, but I found in Lem’s Cyberiadsomething like this:

        Król, któremu poniekąd uszedłem, zwie się w samej rzeczy Grólem, a to z tej przyczyny, że nie Państwem rządzi, lecz Baństwem…

        But in spite of appearance, this is not the kind of construction you were looking for, because here “nie” negates noun, not the verb.

        Prescriptive orthography, as it is taught in schools, says that “nie” can be written jointly with adjectives and nouns, but separately with verbs. How to treat participles, then? Several years ago they could go either way. Non-smoking could be translated as nie palący or niepalący. There was hardly any distinction between them, so now niepalący is the only recommended form

        • Thanks, nachasz! Let’s not put sentential and phrasal negation into one category as that only confuses things. With respect to sentential negation, can anything separate that “nie” from the verb? I would think not. A personal pronoun would be a good thing to test: would something like “She not this/him/it remembers” work as sentential negation? In Russian it doesn’t. Still, I don’t know if that tells us that the negation marker is a prefix. It may be a clitic. Similar issues arise with respect to -sja in Russian at the other end of the verb…

  • Vatroslav Herceg

    In Croatia we have also similar talks about prescriptivism. The book Čiji je jezik? from 2011. by Mate Kapović attacked prescriptivistic positions and strongholds.

    • Thanks for your comment, Vatroslav! There’re definitely prescriptivist traditions in other languages/cultures as well (e.g. in Russian).

  • Michael Morse

    I’m at a loss to understand this discussion. The attacks on S&W remind me of Frank Zappa’s famous comment on censoring Rock lyrics: treating dandruff with decapitation. If the undeniable fallibilities of S&W are worth critique–and who would deny it?–what, in the case of a wholesale abandonment of “prescriptivism”–as hoary a piece of oversimplification through jargon as I can imagine–should replace it?
    The contrast between “style” and “grammar” presented here is in one crucial respect a pseudo-debate. The role of logic cannot be bracketed or ignored. Time and again S&W point us to lazy or muddled or, hell yes, ill-informed constructions that generate immediately baffling ambiguities that undermine the sense the speaker is attempting to convey. In the brave new, Steven Pinker-y, post-prescriptivist world, we’re gonna spend an awful lot of time disentangling what is meant from what is said. Unless y’all have a better plan? And what *is* it, exactly?

    • For someone who complains about not understanding the discussion, you write in a rather twisted manner yourself…

      Anyway, yes I do have a better plan, and it would include teaching grammar correctly (which would involve teaching done by teachers who are themselves well-trained and well-informed in the matter they are teaching) AND separate teaching of style or, better still, effective writing, with the latter being based on informed understanding of grammar rather than the lack thereof.


      • Michael Morse

        But it sounds like your stated aim is entirely consonant with S&W’s, if not identical, even. That your proposed instructors be better qualified grammarians is, surely, a technical distinction rather than a philosophical one. And you leave a crucial philosophical question unaddressed, a question that S&W frankly admitted probably couldn’t be answered: the “separation” of grammar, style, and effective writing. They argued that there are no hard and fast lines here. You disagree; but, again, if you have explained how these distinctions obtain, I didn’t see it.
        I suspect the real conflict, for you if not for Pullum, is not so much that S&W were lackadaisical grammarians, but that they had no knowledge of nor interest in linguistics. In a sense, that’s what interests me about the discussion. Is a knowledge of scientific linguistic theory a prerequisite for teaching the art of clear prose and thinking? I wouldn’t give that an unqualified “yes,” by any means; but I do think your case against S&W is premised on discounting their plainly stated premises, and is to that extent at least not proven.

        • Whatever S&Ws aims are, they certainly do not speak of grammar cogently or in an informed way. Even if one can teach effective writing or style without recourse to grammar (which I am not at all convinced of), S&W certainly aren’t doing that. And as a matter of general principle, if one writes about X, one should be informed (better still: knowledgeable) about X. They write about grammar and they are not well-informed about it.