Yet another colleague has called for backup to respond to someone who insists that splitting infinitives is always and without exception wrong.
Siiigggghhhhh. Really, do these people never, ever look anything up? Why is it that people will cling fast to the authority of their grade-school English teacher when they would readily confess that their biology, chemistry, or physics teachers could not trump the knowledge of physicians, pharmacists, or engineers? Not just linguists (whom grammar cranks resolutely scorn – oh, please, let me never drive across a bridge built by someone with the same attitude towards expertise) but the authors of grammar references will tell you that there is no good basis for the belief that one should never “split an infinitive” – a belief which is a comparatively recent invention and imposition.
Here’s the Oxford Guide to Canadian English Usage (466–7):
English writers started splitting the infinitive at least as early as 1380 [my note: in Old English you couldn’t without hacking a word in two, as you will see in the history I give below]; condemnation of such constructions did not begin until the nineteenth century. And certainly it seems rather arbitrary that the infinitive in the sentence ‘To carefully revise this will take time’ is vehemently opposed when no one objects to the interjection of the adverb in quite similar structures, such as ‘This needs to be carefully revised’ or ‘This appears to have been carefully revised’. . . . Various structures in English suggest that there is no indissoluble bond between the to and the base verb in the infinitive. In the sentence ‘We were going to but we decided not to’, the base verb has been dropped altogether. . . . Theoretical arguments aside, it is not difficult to produce examples of sentences that require a split infinitive: ‘They gave her enough to more than satisfy her needs’; . . . ‘His key failure was to not require an estimate in advance’. In this last example, if not were placed in front of to, readers might assume that the key failure was something other than not insisting upon an estimate.
The Oxford Guide goes on to produce several examples of cases where trying to avoid a split infinitive produces awkward or ambiguous writing.
The New Fowler’s Modern English Usage has a lengthy section on the question (736–738). After a history of the usage, it notes that “it is clear that a rigid adherence to a policy of non-splitting can sometimes lead to unnaturalness or ambiguity. Gradations of these can be found in the following: (unnatural) In not combining to forbid flatly hostilities, (ambiguous) In not combining flatly to forbid hostilities; (unambiguous) In not combining to flatly forbid hostilities.” It then goes on to cite numerous examples of split infinitives in common and standard usage, and concludes “No absolute taboo should be placed on the use of simple adverbs between the particle to and the verbal part of the infinitive.”
The New York Public Library Writer’s Guide to Style and Usage (148) lines up on the same side (of course):
Split infinitives were first seen in the 1400s and were accepted until a grammar book published in 1864, The Queen’s English, strongly opposed placing words between to and the stem of the verb. This pronouncement became the voice of authority until 1926, when H.G. Fowler, in Modern English Usage, argued that personal style and clarity of meaning should take precedence. Yet almost 75 years later, many people, including teachers and grammarians, still believe that split infinitives are incorrect. Considering that split infinitives have been “wrong” for only about 60 years may help put the usage in perspective.
It then offers examples that are more readable with a “split” infinitive.
So all these people are hewing with such vehemence to something that was largely invented by a Victorian. I am happy to say that in other details of science and society, we have quite readily discarded the mistaken notions of benighted voices from earlier centuries. We do not make measurements in physics based on phlogiston or caloric; we do not calculate epicycles into astronomy. And we can certainly stop feeling bad about doing something that well-educated standard users of English do all the time.
Let me cap off by quoting myself:
In Old English, one couldn’t split an infinitive – infinitives were one word. The word “to” was used just before a specific “inflected infinitive” form that was used in a limited set of situations. But during Middle English, a lot of the inflectional endings were lost, including the suffix that denoted the infinitive, and the inflected infinitive form with “to” before it came to be standard. But the “to” doesn’t travel with the infinitive everywhere: we all should to know that. At any rate, certain self-styled English experts who just happened to be in positions of influence decided that if Latin didn’t split its infinitives (of course it doesn’t; they’re one word), then English shouldn’t either, and so the “to” and the root should never be separated. But does this improve the clarity of English? Does it allow for better communication? If we allow so-called “split infinitives,” we can allow different meanings for “to really do something” and “really to do something,” for instance. We can also avoid some annoyingly difficult phrasing on occasion. Here’s an example of a useful split I found in a document recently: “We will keep pushing you to constantly increase your limits.” Move the “constantly” and it can be read as modifying “keep pushing” rather than “increase.” The main value of the supposed rule is that it allows “those who know” to set themselves above “those who don’t.” But does that serve communication? Wouldn’t it be better to look at it case by case and decide where the insertion of the adverb after “to” is just lumpy and inelegant and where it’s elegant and nuanced?
Remember: language exists for communication, not communication for language. A usage is worth keeping if it lets you do more with the language. It is not worth keeping if it serves mainly to limit what you can do with the language. And it is certainly not worth keeping if it is baseless and goes against centuries of common practice!