Historians are always born too late: Alexander the Great is dead, and so is Charlemagne, so that descriptions of past events have to be reconstructed from chance remarks, unreliable accounts, deliberate lies, and semi-obliterated traces. Language historians are in a similar situation. All of a sudden (always all of a sudden) texts begin to suggest odd pronunciations, new auxiliaries, loss of endings, or previously unheard-of syntax. Why did people in 1324 or 1657 change the norm (the dates are, of course, imaginary) and why did the innovation spread? If we could be there and observe! Alas and alack: too late. But every now and then we are allowed to witness such shifts. Curiously, it turns out that phenomena happening before our eyes are often as hard to explain as those of epochs past.
Several years ago, I read the following sentence in a student newspaper: “Pedestrians are warned to not cross the street, etc.” I am sure I missed the “birth” of this process, but I do not think that to not cross the street was so common, let alone universal, in the seventies and even in the eighties. Few questions about grammar are less exciting than the use of the split infinitive. The topic was made famous by H.W. Fowler (A Dictionary of Modern English Usage, 1926), who divided the English speaking world into five categories: 1) those who neither know nor care what a split infinitive is; 2) those who do not know but care very much; 3) those who know and condemn; 4) those who know and approve; and 5) those who know and distinguish. He addressed mainly the second group. What followed that discussion has been a long and trivial anticlimax, though Fowler’s joke about people who would as soon be caught putting their knives in their mouths as splitting an infinitive has made several generations of language lovers smile. Since 1926 variations on Fowler’s entry have been repeated in every book with the words English usage in their titles, and every author arrived at approximately the same conclusions: split if you must, but do it with understanding. The essence of the change that has occurred in American English and that constitutes the subject of this post can be captured in the formula: “Split, split, split.” Splitting has become gratuitous.
There is, apparently, no gain in substituting to not cross the street for not to cross the street. From a theoretical point of view, it would be easy to justify either variant, for measurements proving that the particle to is attached more closely to the verb than to the negation do not exist. A phrase like to not cross suggests a great cohesiveness of not cross, whereas not to cross points in the opposite direction. At one time, to (now a formal marker of the infinitive) was the same to that we know today as a preposition. The infinitive could be declined, and originally its dative preceded by to indicated purpose (“in order to”), but quite early this construction began to be used more or less mechanically in contexts in which reference to purpose could not be detected. No words ever stood between the preposition to and the dative infinitive. Splitting became necessary after English had lost most of its endings and the rigid word order deprived the language of its former flexibility. However, with regard to infinitives, tradition remained stable and favored not to be over to not be, unless special emphasis and a particularly artificial mode of expression were required (“How sad it is to not be when flowers bloom and all around make love!”). Needless to say, the new usage (to not cross the street) does not imply emphasis. Here is an ad: “The military is experimenting with a program that uses meditation and yoga to not only treat stress, but helps troops prepare for the rigors of combat.” Why to not only treat instead of not only to treat? The entire text is awkward. The editor knew little about the rigors of writing good English but unwittingly, almost instinctively (a common case in language change) imitated what had become the most recent norm.
The split infinitive creeps in (unless the writer is ready to restructure the whole sentence) when the choice is between delaying an adverb too long and thrusting it between to and the verb (for example, to fully understand the use of such an intricate rule in a few days is impossible). But in the following sentence it would have been easy not to split. “All we expect them to do is to quietly withdraw from regions [where] they have never been before…” To withdraw quietly from regions would have served its purpose equally well and not made the period bulky. “As president of the United States I would order the secretary of the Treasury to immediately buy up the bad home loan mortgages in America…” The speaker (John McCain) wanted to say immediately together with buy, not after America, and I would immediately order to buy up is not quite the same as I would order to immediately buy up, but to immediately do something can now be seen everywhere. It has become a stock phrase (compare: they wanted us to immediately discontinue the experiment). In most cases switching the adverb would have been easy, as in: “It is common to relocate or remove buildings that have proven to frequently be in the wake of storms, flooding or other disasters” (= that have frequently proven). Splitting to be strikes me as especially ugly. From students’ papers: “…again allowing French words to easily be borrowed at will” (= to be easily borrowed); “This made them more likely to also be very familiar with… Latin” (= this also made them). I have a rich collection of similar sentences.
Little can be added to what Fowler and others have said about the split infinitive. When used reasonably, it need not irritate anyone, though the commonsense rule (do not split if you can help it) has lost none of its value. In …a nationalism that allowed the English language to once again outlive the language of a foreign conqueror (from a paper), that once again allowed would, to my mind, have been preferable, but let her rip. As an observer, I find it interesting that someone somewhere began to incautiously split infinitives, and now everybody splits them, as though they were firewood. No other shifts in syntax, morphology, or phonetics seem to have triggered the new fashion. Did some celebrity speak so and the public follow the splitter? Do videogames promote this usage? Language historians are always on the lookout for causes. Does anyone know the cause of this change?
Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to [email protected]; he’ll do his best to avoid responding with “origin unknown.”
As to celebrity, there’s the famous Star Trek catchphrase “To boldly go where no man has gone before”.
I think it’s Fowler’s group (1) which has grown rather than group (4). Personally, negatives are the only things I find jarring as splitters of infinitives. Among the examples you cite, “have proven to frequently be” seems to me to be slightly but definitely distinct from “have frequently proven to be”.
Maybe the frequency of occurrence of split infinitives correlates with that of gapping constructions (”I don’t want to!”, “Do we have to?”, etc) where the verb is not merely separated from the to-particle but omitted altogether. Some anti-splitting traditionalists would oppose these as “ending a sentence with a preposition”.
I remember reading somewhere that at the end of the 19th century there was a fad for publishing books of “correct” usage of the English language and that many of these were written not by linguists but by ‘experts’ in good manners! So all sorts of “rules” got invented then, such as not splitting infinitives, ending sentences with prepositions, and all sorts of other nonsense up with which I shall not put; to fittingly quote Winston Churchill!
There are instances where there doesn’t seem to be an alternative to splitting the infinitive eg ‘We expect to more than double our sales this year’ (unless ‘more than double’ is assumed to be a verb in its own right).
‘He failed to completely satisfy his critics’ is different in meaning from ‘He failed completely to satisfy his critics’, so you’d have to split the infinitive to convey a particular sense.
Sometimes it is difficult to phrase a sentence without splitting an infinitive eg ‘We expect to more than double our sales this year’.
Some phrases would mean something different if the infinitive wasn’t split eg:
‘He failed completely to satisfy his critics’
and
‘He failed to completely satisfy his critics’.