June 03, 2005

Step on a crack, break a grammar rule

A few days ago, I posted on metafilter a link to Stanley Fish's NYT Op-Ed "Devoid of Content". One of the mefi commenters observed that

I don't know if anyone noticed, but the article isn't grammatically correct, itself. He starts at least one sentence with "And..".

I'm a mere amateur in the bacteriology of prescriptivism, compared to professional bug-hunters like Geoff Pullum, Arnold Zwicky and Geoff Nunberg. Still, this struck me as a particularly interesting specimen.

The usage note in the American Heritage Dictionary's entry for and says that

It is frequently asserted that sentences beginning with and or but express “incomplete thoughts” and are therefore incorrect. But this rule has been ridiculed by grammarians for decades, and the stricture has been ignored by writers from Shakespeare to Joyce Carol Oates. When asked whether they paid attention to the rule in their own writing, 24 percent of the Usage Panel answered “always or usually,” 36 percent answered “sometimes,” and 40 percent answered “rarely or never.”

This is wild. There is nothing in the grammar of the English language to support a prescription against starting a sentence with and or but --- nothing in the norms of speaking and nothing in the usage of the best writers over the entire history of the literary language. Like all languages, English is full of mechanisms to promote coherence by linking a sentence with its discourse context, and on any sensible evaluation, this is a Good Thing. Whoever invented the rule against sentence-intitial and and but, with its a preposterous justification in terms of an alleged defect in sentential "completeness", must have had a tin ear and a dull mind. Nevertheless, this stupid made-up rule has infected the culture so thoroughly that 60% of the AHD's (sensible and well-educated) usage panel accepts it to some degree.

Imagine something like this happening in another domain. For example, someone decides that stepping on sidewalk cracks puts your spine out of alignment and tends to cause bad posture and muscular weakness. No evidence is provided, and the idea is widely ridiculed by biomedical researchers, but all the same, after a few years, 24% of a panel of family doctors say that they "always or usually" pay attention to this rule in their own perambulations, and 36% say that they "sometimes" do.

The only way I can make sense of this story is to see it as an infectious form of obsessive-compulsive disorder. But why are the cultural domains of grammar and linguistic usage so vulnerable to such epidemics, at least in modern times?

For anyone who thinks that there is a basis in historical patterns of usage for this "rule", a few examples can be found below.

The fastidious Henry James was fond of sentence-intitial And. According to the Hyper-Concordance of the Victorian Literary Studies Archive, Washington Square alone contains 128 instances in 65,392 words, for a rate of 1,957 per million words. The first two examples:

It was the way a young man might talk in a novel; every one looking at him, so that you wondered at his presence of mind. And yet Mr. Townsend was not like an actor; he seemed so sincere, so natural.

For a minute, if it had not been for the rumbling of the carriage, you might have heard a pin drop. "I don't know, Aunt Lavinia," said Catherine, very softly. And, with all his irony, her father believed her.

The elegant elitist Virginia Woolf was even fonder of this usage -- there are 175 examples in the 69,811 words of To the Lighthouse (a rate of 2,507 per million). For example:

When they talked about something interesting, people, music, history, anything, even said it was a fine evening so why not sit out of doors, then what they complained of about Charles disparage them -- he was not satisfied. And he would go to picture galleries they said, and he would ask one, did one like his tie? God knows, said Rose, one did not.

She had a dull errand in the town; she had a letter or two to write; she would be ten minutes perhaps; she would put on her hat. And, with her basket and her parasol, there she was again, ten minutes later, giving out a sense of being ready, of being equipped for a jaunt, which, however, she must interrupt for a moment, as they passed the tennis lawn, to ask Mr Carmichael, who was basking with his yellow cat's eyes ajar, so that like a cat's they seemed to reflect the branches moving or the clouds passing, but to give no inkling of any inner thoughts or emotion whatsoever, if he wanted anything.

The first example in Mark Twain's Tom Sawyer comes early in the first chapter:

The old lady reached out her hand and felt Tom`s shirt, and said: "But you ain`t too warm now, though." And it flattered her to reflect that she had discovered that the shirt was dry without anybody knowing that that was what she had in her mind.

Tom Sawyer has sentence-initial And 139 times in 73,844 words, for a rate of 1,882 per million.

Similarly in Charles Dickens' Bleak House, the first of the 629 examples come on line 134 out of 40,353:

The raw afternoon is rawest, and the dense fog is densest, and the muddy streets are muddiest near that leaden-headed old obstruction, appropriate ornament for the threshold of a leaden-headed old corporation, Temple Bar. And hard by Temple Bar, in Lincoln's Inn Hall, at the very heart of the fog, sits the Lord High Chancellor in his High Court of Chancery.

In this novel, the 629 examples of sentence-initial And in 361,977 words correspond to a rate of 1,738 per million.

As point of reference for these rates, consider that the word "think" occurs in the same texts at rates of 2,340 per million (Washington Square), 688 per million (To the Lighthouse), 650 per million (Tom Sawyer), and 1,392 per million (Bleak House).

[Update: a reader emailed to observe that Microsoft's Grammar Checker attempts to impose this rule by putting wavy lines under sentence-initial and and but, offering "moreover" or "in addition" as substitutions. I just checked, and this is sad but true. The same correspondent suggested that:

Some people may perceive the use of and and but as more appropriate for casual or informal writing/discourse (in terms of register), including fiction (which is probably forgiven or accepted because it's "art"), while academic and formal writing registers are perceived to require either more complex sentences in which single "thoughts" comprise a sentence (i.e., rewrite the sentences to form one complex - and probably much longer and harder to read - sentence) or longer, rarer, and more obscure alternatives should be used to the monosyllabic and and but.

Perhaps some people think that sentence-initial and is informal and should only be used in fiction -- the author of the note is apparently an existence proof -- but I doubt that this belief is held on the basis of any evidence. When I look at well-regarded works of history, political analysis, economics, philosophy and so on, I find a rate of sentence-initial and that is similar to the rates found in works of fiction.

Here are two examples from Gibbon's Decline and Fall:

The policy of the emperors and the senate, as far as it concerned religion, was happily seconded by the reflections of the enlightened, and by the habits of the superstitious, part of their subjects. The various modes of worship, which prevailed in the Roman world, were all considered by the people, as equally true; by the philosopher, as equally false; and by the magistrate, as equally useful. And thus toleration produced not only mutual indulgence, but even religious concord.


Among the innumerable monuments of architecture constructed by the Romans, how many have escaped the notice of history, how few have resisted the ravages of time and barbarism! And yet, even the majestic ruins that are still scattered over Italy and the provinces, would be sufficient to prove that those countries were once the seat of a polite and powerful empire.

Here are a couple of example from Jefferson's Notes on the State of Virginia:

A true answer to these questions would probably lighten their authority, so as to render it insufficient for the foundation of an hypothesis. How unripe we yet are, for an accurate comparison of the animals of the two countries, will appear from the work of Mons. de Buffon. The ideas we should have formed of the sizes of some animals, from the information he had received at his first publications concerning them, are very different from what his subsequent communications give us. And indeed his candour in this can never be too much praised. One sentence of his book must do him immortal honour. `J'aime autant une personne qui me releve d'une erreur, qu'une autre qui m'apprend une verité, parce qu'en effet une erreur corrigée est une verité.'


Of their eminence in oratory we have fewer examples, because it is displayed chiefly in their own councils. Some, however, we have of very superior lustre. I may challenge the whole orations of Demosthenes and Cicero, and of any more eminent orator, if Europe has furnished more eminent, to produce a single passage, superior to the speech of Logan, a Mingo chief, to Lord Dunmore, when governor of this state. And, as a testimony of their talents in this line, I beg leave to introduce it, first stating the incidents necessary for understanding it.

From Federalist Papers #1, by Alexander Hamilton:

Were there not even these inducements to moderation, nothing could be more ill-judged than that intolerant spirit which has, at all times, characterized political parties. For in politics, as in religion, it is equally absurd to aim at making proselytes by fire and sword. Heresies in either can rarely be cured by persecution.

And yet, however just these sentiments will be allowed to be, we have already sufficient indications that it will happen in this as in all former cases of great national discussion.

And from Federalist #5, by John Jay:

Considering our distance from Europe, it would be more natural for these confederacies to apprehend danger from one another than from distant nations, and therefore that each of them should be more desirous to guard against the others by the aid of foreign alliances, than to guard against foreign dangers by alliances between themselves. And here let us not forget how much more easy it is to receive foreign fleets into our ports, and foreign armies into our country, than it is to persuade or compel them to depart.

And from Keynes' General Theory of Employment, Interest and Money:

Obviously our quantitative analysis must be expressed without using any quantitatively vague expressions. And, indeed, as soon as one makes the attempt, it becomes clear, as I hope to show, that one can get on much better without them.

As an exercise for the reader, I recommend trying Microsoft's suggested "Moreover" or "In addition" as substitutions for the sentence-initial Ands in the passages above. Is this ever an improvement, or even equally good? I don't think so.

Posted by Mark Liberman at June 3, 2005 08:09 AM