Talk:Split infinitive/Archive 2

This is an archive of past discussions about Split infinitive. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Teddy bear example

In the R. L. Trask example, illustrating the potential locations for the word "gradually" in a sentence, isn't there another option? I may not be a professor of English or anything, so maybe my take on it is incorrect, but wouldn't, "She decided to get rid of, gradually, the teddy bears she had collected," do the job? It doesn't split the infinitive and it retains the meaning of the sentence in full, at the cost of a couple paltry commas. If I'm wrong someone please correct me. I just took objection to the matter-of-fact nature ("... it is also the only semantically sound possibility") of the conclusion when I saw another option. Lemonsawdust 08:36, 28 May 2006 (UTC)

Well, Trisk doesn't offer that one. I suppose it is just about possible but it is clumsy and no one would, naturally, put an adverb in commas in this way.Damiancorrigan 16:07, 28 May 2006 (UTC)

Well I don't know about that. It may not be "ideal," but I, personally, don't consider it a clumsy construction.Lemonsawdust 03:44, 29 May 2006 (UTC)

Is the solution not simply to add this possibility to the list and label it awkward like the others?--Lo2u 11:04, 29 May 2006 (UTC)

Maybe. I wasn't sure enough to change it myself, so I sought wiser counsel here. If the conclusion itself is a part of Trask's example, then I'm totally satisfied. It just wasn't clear to me if it was part of the example or a conclusion extrapolated from it, so I asked the question as if it was the latter. Thanks for the clarification, however. Lemonsawdust 11:58, 29 May 2006 (UTC)

Good point - if it's not part of Trask's example it really is not very good that it should have to be added. Besides which I think most people are slightly more likely to say "she decided that she would gradually get rid of the teddy bears that she had collected". I prefered the old example.--Lo2u 10:48, 30 May 2006 (UTC)

Is it just me who thinks that "to get rid of" sounds a bit odd in the first place? Most of my grammar is instinctive "that sounds wrong", so feel free to say "yes, it is just you". It just made me think: If the argument is based on a non-standard use of "to get" (to the extent that http://www.edufind.com/english/grammar/GET.cfm lists it under "other"), would that make it less valid? Is there a reason for why the website has listed it under "other", or should it really be in with one of the other patterns? --Alsuren 21:48, 29 July 2006 (UTC)

"She decided to gradually get rid of the teddy bears she had collected." seems awkward to me. I think the following has the same meaning, is less wordy and lacks the split infinitive:

"She decided to get rid of her teddy bear collection gradually."

Badocter 05:25, 30 July 2006 (UTC)

I came here with the same point raised by LemonSawdust, but after reading this discussion I feel that the comma-delimited adverb should be added. Though it should not be added as part of the Trask example. I have used a comma separated adverb in writing on many occasions. Most wouldn't say it. Wouldn't the sentence, "She decided that she will gradually rid herself of the teddy bears she had collected", be clear and avoid the split infinitive? (Hechz 23:55, 1 June 2007 (UTC))

I'll add the comma-separated example, but since it's not part of Trask's example and since it sounds far less natural to me than the split infinitive, I'll put it in a note. Others can change that if they want, and we'll try to arrive at a consensus.

There are already examples in the article similar to your second suggestion (with "that she will" and "rid herself of"), so my opinion is that we don't need that suggestion. Subjectively, I have to say that "rid herself of" sounds a lot stronger than "get rid of", in addition to its greater formality. —JerryFriedman 16:27, 2 June 2007 (UTC)

Counter example please

There are indeed many examples where an infinitive needs to be split to make the meaning clear, and especially to make the emphasis clear. There is also a difference in acceptability and clarity between what is written and what is spoken. That is, more care should be taken when the English is written. The article includes some of these examples. Nevertheless, there are many occasions where a little extra effort to avoid a split infinitive will greatly aid clarity or elegance. Indeed, if it were not so, there would be no controversy. Could we have some examples of these included near the beginning of the article? Another reason for taking care with these things is that little words shuffled in a range of ways in sentence are very difficult for speakers of other languages to understand. Witt y lama 13:24, 31 May 2006 (UTC)

Prescriptive Grammar is out of date

I think that this on the whole is an excellent article about a contentious topic. If the article were about say the structure of atoms, we'd have not compunction in stating the latest views of physicists as being the 'NPOV' viewpoint rather than say physicists of the 18th century. Similarly, we should take the viewpoint of linguists as regarding the meaning of grammar. Prescriptivism is essentially an idea of the past regarding grammar, and linguists (almost) universally describe language from a descriptive viewpoint. The term 'prescriptive grammar' is itself an anachronism, and really refers to 'style'. And herein lies the problem. You can't say something is stylistically right or wrong any more than you can say a painting is stylistically wrong as opposed to being stylistically poor. Most of the supporters of the Split Infinitive based their arguments on misunderstandings of how language works. A similar case is to be seen with the '2 positives makes a negative' in language, which is a thorough misunderstanding of the concept of scope.

This leads to the situation where the meaning of grammar is widened to encompass the 'style' element, but then within this meaning there is a non-sensical debate about whether it is right or wrong. Prescriptivists can't have it both ways. If the meaning is 'style' then it's simply a matter of opinion. If the meaning is 'grammar' from the scientific viewpoint then there is no question about whether it is right or wrong: people do it and that's all. So essentially the argument comes down to a definition of the word grammar.

[Of course, the easiest way to know that split infinitives are fine is to see what a trivial rule it is. The difference between 'a' and 'the' has been treated in papers running to over 1000 pages, yet native speakers never get it wrong. Somehow we are meant to imagine that a 6 year old child can pick up this '1000 page rule-book' without thinking yet be flummoxed by a rule that can be described in a few words. Very few people really understand grammar (Pinker, Chomsky etc are amongst the experts) yet the fact that the split infinitive in all its triviality is debated is more a commentary on class and society than on grammar.]

NPOV is all very well, but when you debate grammar you ask the experts: and the experts are linguists, and the linguists say 'Split infinitives are as relevant as the toga' Macgruder 08:37, 16 August 2006 (UTC)

Some facts:

For several centuries, split infinitives were extremely rare in English. (This, by the way, refutes your claim that the proscription is too simple to be a real rule. It was a real descriptive rule.) There are still a great many people who don't split infinitives with negatives. I'm one, and I'm sure I naturally said "I told you never to do that" long before I ever heard of any rules on the subject.

When split infinitives started to become common, some people prescribed against them, initially because of the above fact.

There are still people who dislike them and consider them inappropriate in formal speech and writing.

Prescriptive grammar has a very strong, observable influence on formal speech and writing (including Wikipedia). I think it's very hard to find linguists, even the greatest experts on scope, putting sentences such as "This construction ain't never had no analogues in none of the other Germanic languages" in their articles or lectures.

So prescriptive grammar is not the phlogiston of linguistics (if I may paraphrase). I agree that it's a style issue, that is, partly esthetic partly about clarity, but the esthetic issue needs both sides presented, the clarity issue needs facts such as sentences with split infinitives that can't be changed easily without losing clarity, the history needs to be covered, and the current status as a shibboleth needs to be described with reference to usage manuals.

So I agree again that this is a pretty good article, though I think it needs a comprehensive edit. And I'd still like primary-source evidence for the claim that anybody ever condemned split infinitives because of an analogy with classical languages or French! —JerryFriedman 21:08, 23 August 2006 (UTC)

Classical Languages Argument

In the "argument from classical languages" section, the article says, "In Greek and Latin, it is impossible to split infinitives because these languages never use their infinitives together with a preposition." This doesn't make much sense to me; at best it's rather misleading. I think what the author meant is that Latin and Greek form their infinitives with an affix on the verb itself, as opposed to with a particle (not a preposition) as in English. Infinitives couldn't be split because they were single words. Masily box 15:37, 21 August 2006 (UTC)

Greek splits its infinitives as an oratorial device. That extract from the article is erroneous. Raifʻhār Doremítzwr 21:10, 21 August 2006 (UTC)

I'm ignorant of Classical Greek - how does it do this? What infinitival particles are used? Slac speak up! 23:28, 21 August 2006 (UTC)

Coïncidentally, it is the same as English’s: “to” (though pronounced tɒ, as opposed to the English tʊ). It was explained to me a long time ago by someone much more familiar with Classical Greek than I am; I can’t remember how it’s split, only that it is split. Raifʻhār Doremítzwr 11:04, 22 August 2006 (UTC)

"το" is the neuter definite article in Greek. It can be used with the infinitive to make a kind of gerund: "το λυειν" would mean "the (act of) loosening" or the like. I suppose it could be separated from the infinitive by an adjective or something, but it's not really the same thing as the English split infinitive. User:Angr 11:36, 22 August 2006 (UTC)

To boldly complain about the intro

I'm not a big fan of having the startrek reference in the intro of the article, but that's a minor issue compared to the entire paragraph discussing. The intro should not discuss alternate ways of beginning a startrek episode. Vicarious 11:45, 15 September 2006 (UTC)

The only real reason it's there is that it's a popular, extremely well-known example of a split infinitive. The fact that it's a Star Trek reference is essentially tangential. Thefamouseccles 00:03, 16 Oct 2006 (UTC)

reorg

Am I the only one who thinks this article needs serious reorganization? Right now I'd like to give it some, maybe in the next week or so. —JerryFriedman 21:04, 20 October 2006 (UTC)

History

The section on history, and the above discussions, and some things which have been written elsewhere in Wiki in the last day or two leave me thinking we need a comprehensive understanding of the history of this. I had a look in our library but didn't find anything. So I am going to attempt my own summary, based on my understanding of the various parts of the story I have picked up. This cannot go in the article as in this form it is original research; but I am sure the research has been done and it would be great if anyone can find authorities supporting or correcting my views. Anyway, the background as I understand it, for what it's worth:

When the Germanic languages broke away from Indo-European, one of the innovations was the formation of an infinitive out of the PIE verbal noun. Other Indo-European languages (eg. Latin, Greek) also have infinitives, but they are not from the verbal noun; they have a different origin and therefore they are not quite analagous, at least in the early phase.
Because the verbal noun really was a noun, it could be used with a preposition. This preposition survives into the modern languages (to do, zu tun). In the syntax of the modern languages these are no longer prepositions; they are best called "particles" or similar; but it is no co-incidence that they are identical in form to the prepositions "to", "zu", Dutch "te" etc; their original use was prepositional. Since the infinitive in Latin and Greek does not come from a verbal noun, it was never used with a preposition (or particle which looks like / is derived from a preposition).
Just as you can't normally put an adverb between a preposition and the following noun phrase, so this was not possible with the verbal-noun/infinitive; and in most Germanic languages it is still not possible: for example in German you cannot say *zu nicht tun (to not do). In other words, in Germanic you cannot split an infinitive. (Note that this is a slightly different situation from Latin, where there is no such 2-word construction to split!)
But then French enters the tale. French is descended from Latin, but has a lot of fetures which are not Latinate, and these are mostly borrowed from Germanic. One of the novelties in Old French which was not found in Latin was the use of the prepositions à and de before an infinitive. This will be an influence from the Germanic construction.
Because French is not Germanic, however, it processed this differently. We would need a Romance syntactician to explain this in detail, but basically where German beschließen zu tun (=decide to do) breaks down into beschließen + zu-tun, French décider de faire beaks down as décider-de + faire; i.e. the preposition hangs on more closely to the finite verb. This means that in French, the natural place for an adverb IS between the prepositon/particle and the infinitive: décider de ne pas faire. In other words, in French you CAN split an infinitive, though to call it splitting an infinitive is to view French Grammar through the lens of English, which is not quite kosher.
After the Norman Conquest, English borrowed many elements of syntax from French. It is no co-incidence that it is in the Middle English period - in the two centuries after the conquest - that the split infinitive appears in English.
Possibly the development of the "full infinitive" as the citation form in English encouraged English grammarians to do what neither French nor Germanic grammar had done before, namely to see the "to" as PART of the infinitive. Only at this point is the terminology "splitting" meaningful. And since the split infinitive was rare and non-standard, prescriptive authorities did not like it.
The rise of the colloquial split infinitive in the 19th century, which not only made it common but also extended it to other kinds of adverbs besides the negation, may just be an extension of this, but it was nevertheless something new, and I would be surprised if there was not some impuls which sparked it off, though I can't think what.

Any or all of the above may be wrong, but if it is right, it should be helpful. I would be grateful for feedback. --Doric Loon 18:56, 24 October 2006 (UTC)

A suggestion I've seen ~~(I can come up with a reference)~~ is that it's parallel to constructions with finite verbs and with modals plus bare infinitives. She gradually gets rid of her teddy bears and She will gradually get rid of her teddy bears lead to She has decided to gradually get rid of her teddy bears. "If the adverb should immediately precede the finite verb, we feel that it should immediately precede also the infinitive…" George Curme, "The Split Infinitive", American Speech 2:8 (May, 1927), available at JSTOR if you have access. ("precede also the infinitive"?)

What your account leaves out is that split infinitives increased for a few centuries, then decreased in the sixteenth and seventeenth, then came back in the eighteenth and especially the nineteenth. Very strange. Anyway, that's interesting stuff! —JerryFriedman 22:58, 24 October 2006 (UTC)

Thanks Jerry. I hope you don't mind, but you put your comments in the middle of mine and it wasn't clear what was whose, so I have moved your remarks to below. This will make it easy for other comments to follow from them. --Doric Loon 09:52, 25 October 2006 (UTC)

Feedback to point 1

This is all very interesting, but I'm not sure it belongs in the article, at least in the main part. Maybe it can be some kind of sidebar, if we get good references. I think we could begin the history of the article at Anglo-Saxon. (But maybe I'm prejudiced because I'm thinking the history belongs right after the lead.) - JerryFriedman 22:58, 24 October 2006 (UTC)

Well, probably it belongs in the article on the infinitive; but my thinking was that we can't really understand what a split infinitive is without understanding what the particle is, and why a Germanic infinitive is different from, say, Latin. --Doric Loon 09:52, 25 October 2006 (UTC)

Feedback to point 4

This last (Germanic influence on prepositions before the infinitive) needs a citation, imho. - JerryFriedman 22:58, 24 October 2006 (UTC)

Absolutely. These are my own conclusions; we need authorities. --Doric Loon 09:52, 25 October 2006 (UTC)

Feedback to point 5

It might be a coincidence. A citation would be great; if we don't have one, I think we should call it speculation. What would be really awesome would be examples of how infinitives were treated in the French and German of those same centuries, or at least evidence that it was the same then as it is now. - JerryFriedman 22:58, 24 October 2006 (UTC)

I don't mind your moving my comments, DL, and I wish that while you were at it, you had read my mind and deleted the parenthesis I just crossed out :-) Anyway, it looks like we agree on a lot, and now I see that you did mention the need for authorities. —JerryFriedman 21:48, 25 October 2006 (UTC)

As you can see, I got started on the history with some things I have citations for (many of them thanks to User:150.203.177.218). —JerryFriedman 00:18, 26 October 2006 (UTC)

Thank you! This is SO much better! --Doric Loon 18:26, 27 October 2006 (UTC)

It's good to know I'm accomplishing something. —JerryFriedman 21:02, 27 October 2006 (UTC)

By the way, I don't feel done with the reorganization. For instance, I'd like the article to discuss moving the adverb or rewriting the sentence only once. And is there any way we can move ~~the ridiculous red herring about Latin~~ the argument from classical languages to some kind of sidebar? —JerryFriedman 21:11, 27 October 2006 (UTC)

I think I would want to keep the red herring, but in an aquarium where it belongs. Since people reading this will often have that myth in mind, it is better to confront it than to ignore it. --Doric Loon 21:21, 29 October 2006 (UTC)

Better reference

Anybody out there got Visser so we can refer directly to him instead of Nagle? —JerryFriedman 17:26, 26 October 2006 (UTC)

Removal of "The full infinitive as a prerequisite"

I took out the following text.

The full infinitive as a prerequisite

In English grammar the bare infinitive do is distinguished from the full infinitive to do, and to do is often used as the citation form. This is probably because the English infinitive lacks any distinctive inflection, in contrast to French and German, where the bare infinitive is recognisable as such. Therefore in English, the word to is conceptualized as part of the infinitive, whereas in French and German, à/de and zu are not. The difference is subjective, as the constructions are parallel, but this functional difference in English perception allowed for the subsequent admonition against split-infinitive constructions.

My reasons were

It's speculative.
Two of the first three anti-splitters quoted mention the bare infinitive as the infinitive and the to as a particle, in modern fashion. Thus if we had to speculate on their concepts, we'd say they came up with their admonitions without conceptualizing to as part of the infinitive. —JerryFriedman 21:09, 27 October 2006 (UTC)

Good point. A lot of this article was written before we had any sources. The more actual quotes appear, the more needs deleted. Frustrating, but this is progress too. --Doric Loon 21:19, 27 October 2006 (UTC)

Quotation for possible use

"One reason why the older generation feel so strongly about English grammar is that we were severely punished if we didn't obey the rules! One split infinitive, one whack; two split infinitives, two whacks; and so on." —A correspondent to the BBC on a program(me) about English grammar in 1983. Quoted by David Crystal, The Cambridge Encyclopedia of the English Language, p. 91 (searchable at Amazon). —JerryFriedman 22:41, 27 October 2006 (UTC)

That's edging on the "grammar is a big list of Dont's that you get taught in English class" fallacy, and I don't think that's a misconception we should risk propagating. Unfortunately the BBC (as with most other major media outlets) is really shoddy with linguistic issues. Strad 05:05, 12 November 2006 (UTC)

The quotation is from a correspondent, not the BBC, and as it's used to establish that the rule was firmly entrenched, not to discuss the origin of the rule, I think it's okay. I hope we've made the origin quite clear at this point. —JerryFriedman 23:27, 13 November 2006 (UTC)

Well, I copied it from here into the article over a week ago, and I think it reads well there.--Doric Loon 23:30, 13 November 2006 (UTC)

More removed text

I removed lots of text, with apologies to the authors and no desire to foreclose discussion of it. I removed this as unsourced "piling on" in regard to prescriptivism.

Stylistically, the careful placement of another word between to and the bare infinitive sometimes avoids ambiguity or ugliness. The old prohibition on split infinitives is particularly surprising when one observes that there are a number of expressions in English that are weakened considerably by avoiding the split infinitive.

And this as a pointless example of debate in the article instead of the discussion page.

In other instances, use of a split infinitive is for many people the most natural way to add certain kinds of emphasis in conversation:
Student A: "I'm going to do better next year."

Student B: "I'm going to really do better next year."

It is not necessary to split the infinitive in the above example, however. It could just as easily be said that "I'm really going to do better next year." The speaker may also have intended to say: "I'm going to do much better next year."

And this because the first sentence is superfluous and I don't see the relevance of the second. (If it has some, it needs to be rewritten, imo.)

The best way to avoid using split infinitives is usually via a change in lexical choices. However, in spoken language, phonetic stresses and timing is usually all that is needed for a sentence's actual implications to be understood.

And this because you can split an infinitive with nothing modal in sight, e.g., "The goal is to further exclude Arafat." (Quoted by Burchfield from U. S. News and World Report.) Indeed neither infinitive below ("to disentangle" and "to split") is accompanied by a modal.

It is probably not possible to disentangle this argument from the modality of English grammar. Typically, in a phrase such as "I am going to", the verbal construct "to be going to" acts as a modal verb, analogous to other standard modal verbs "will", "could", "can" etc. In this sense, it becomes apparent that the preposition 'to' does not belong to the infinitive verb, but rather to the modal verb. In this case, it becomes impossible to split an infinitive.

And this because it's POV (though I agree with it completely) and unnecessary

The rule of thumb should always be to make it easier for the reader or listener to understand.

—JerryFriedman 22:09, 28 October 2006 (UTC)

To do

Things that somebody, possibly me, could do:

Rewrite the paragraph about unwisely avoiding the splitting of non-infinitive compound using Fowler. This is where he said "bogy-haunted creatures", by the way. Also, I think the AP Style Guide recommends avoiding those splits, but I can't get my hands on it.
Check the quotation from Bill Walsh (end of the lead). If it doesn't say that no grammarians condemn the split infinitive, Burchfield quotes the Times from May 18, 1992: "The most diligent search can find no modern grammarian to pedantically, to dogmatically, to invariably condemn a split infinitive."
Find a source for the possibility that the 19th-century prohibitions became part of people's "natural" linguistic intuitiion. I have this weird feeling that I saw that somewhere recently.
Consider restoring the claim that for some people, "I don't want to see you any more" means that I don't care, different from "I want to not see you any more." I asked about this in alt.usage.english, and believe it or not, two or three people supported that claim.
Consider the implications of this statement that Burchfield quotes from Visser: "From about the beginning of the sixteenth century to about the last decades of the eighteenth century the use of the split infinitive seems to been as it were tabooed in the written language." There could conceivably have been a taboo we don't know about and, to put speculation on top of speculation, it could even have been influenced by Latin and Greek. —JerryFriedman 22:26, 28 October 2006 (UTC)
Translate the quotation from Layamon and find it for real instead of having to quote Burchfield. I know I just saw the complete text on Google Book Search (or some place), but I can't find it, and I can't do this any more today! —JerryFriedman 22:33, 28 October 2006 (UTC)

Finally! (Though that's not the version I remember, which had the two mss in parallel with notes at the bottom.) I don't know whether leaving out "cnihtes" was my mistake or Burchfield's. But that's not important; what's important is verifying your sources!

Does "and he cleopede him to; alle his wise cnihtes. for to him reade;" mean something like, "and he called to him all his wise knights to advise him"?

Since the split infinitive is in the middle of the line, I took out the statement that it's for the rhyme. Is it really for the meter? —JerryFriedman 00:31, 31 October 2006 (UTC)

Is Layamon's split infinitive an inversion?

And on that subject, do we really know that Layamon wouldn't have written that in prose? Do we know what the inversion might have done for the meter? I agree, of course, that the Shakespeare example is pure poetic license. —JerryFriedman 00:45, 14 November 2006 (UTC)

I quote from the article: "This may be a poetic inversion for the sake of meter, and therefore says little about whether Layamon would have felt the construction to be syntactically natural" (emphasis mine). The article isn't saying with certainty that it's inversion, or that it benefits the meter, but rather that since we don't know whether that's the case, we can't assume that it isn't; and this uncertainty lessens the value of the example. Ruakh 01:24, 14 November 2006 (UTC) Never mind, I misunderstood what part of the article JerryFriedman was referring to. (See his comment below.) Ruakh 18:53, 14 November 2006 (UTC)

My mistake. —JerryFriedman 03:56, 15 November 2006 (UTC)

Nah, I think it's mostly my fault: I should have Ctrl-F'd the page for "Layamon" to make sure I read all the article had to say about him. Either way, all's well that ends well. :-) Ruakh 04:40, 15 November 2006 (UTC)

Remember also that Layamon splits the infinitive with an object pronoun. I know of no other example of this in either Middle English or Modern English, and it would certainly be unidiomatic today, even thought the split infinitive has never been more widely and more flexibly used than it is now. Note that this is an accusative pronoun which is the object of the infinitive, and is quite different from to all hurt one another, where all is an extension of the subject pronoun. Layamon's construction really is VERY odd, and therefore my default opinion is that it is poetic licence, unless it can be demonstrated to have parallels. --Doric Loon 06:48, 14 November 2006 (UTC)

I should have said which part of the article I was referring to. "In verse, poetic inversion for the sake of meter or of bringing a rhyme word to the end of a line often results in abnormal syntax, as with Layamon's and Shakespeare's split infinitives (cited above), in which the infinitive is split by a pronoun and a past participle respectively. However, clearly these would never have occurred in a prose text by the same authors." I think that may be a little too strong. I agree that Layamon's construction would be unidiomatic today (as you said above), but I'm not at all sure that split infinitives are now used more flexibly than ever before. Wycliff's prose "to this manner trete there brother" would certainly be impossible now. What I'd look for is personal pronouns as direct objects before any verb in prose contemporary with Layamon—but I don't even know where to find prose of that time, and I don't necessarily understand Middle English well enough to be sure of what I'm reading.

Well, here's one from about 1450, thanks to Hall (1882): "… if ther with al he kepe and fulfille al the lawe of God so miche and in the maner as it is longing to him forto it kepe and fulfille…" [emphasis mine]. From Pecock's Repressor, p. 102 of the Elibron Classics edition. Hall says Reginald Pecock used many split infinitives. —JerryFriedman 17:41, 14 November 2006 (UTC)

BTW, I think your translation of Layamon is right - put it in the article! --Doric Loon 06:54, 14 November 2006 (UTC)

Thanks! Consider it done. —JerryFriedman 17:41, 14 November 2006 (UTC)

Goold Brown

The 1851 edition of Goold Brown's Grammar of Grammars is at http://www.gutenberg.org/files/11615/11615-8.txt . I just looked through it and couldn't find anything about split infinitives. (There's a LOT, though, about how the infinitive and especially the to are to be interpreted.) It makes me wonder whether he didn't address the question till the 1873 edition. —JerryFriedman 00:51, 14 November 2006 (UTC)

As Emily Litella used to say, "Never mind." —JerryFriedman 17:41, 14 November 2006 (UTC)

...to boldly split infinitives that no man had ever split before!

Is there room for this Douglas Adams quote in the article? It is, of course, a spoof of Star Trek.

TRiG 23:49, 11 December 2006 (UTC)

Define "compound split infinitive"

At least do so earlier in the article. It appears at least twice without any explanation of what it means. Xiner 22:03, 13 December 2006 (UTC)

Good point. Done. —JerryFriedman 23:10, 14 December 2006 (UTC)

Source of the classical argument

I'm not entirely satisfied with this sentence from the article:

It has been claimed that the dislike of the split infinitive is based on a comparison with classical languages.

I especially don't like this sentence in combination with the concluding sentence:

The argument from classical languages may be a straw man constructed by a defender of the split infinitive and repeated as "part of the folkore of linguistics".

I don't like the first sentence partly because it uses weasel words and partly because it makes too weak a statement about too strong a claim.

Too weak: It has been claimed
Too strong: is based on

First, to address the "too strong" point. "is based on" implies something deeper than what I think many defenders of the split infinitive really think or mean (perhaps even the authors of American Heritage Book of English Usage, although I don't have the book so I am missing some context). If the article said "The dislike of the split infinitive is based on a comparison with classical languages", I (and many other readers, I imagine) would read either "Without a comparison with classical languages, there would be no dislike of the split infinitive" or "All dislike of the split infinitive was caused by a comparison with classical languages (perhaps indirectly through some initial comparison with classical languages)", or some combination of the two. This is a very strong claim, which I don't think even American Heritage Book of English Usage is making (perhaps someone who has the book could clarify if they are making an argument based on what is (allegedly) a straw man or simply distinguishing the classical languages argument, which they see as a valid argument with a faulty premise, from other arguments, which they see as irrational). However, even if AHBoEU makes this claim, the correct way to weaken it is not to qualify it with "It has been claimed" but to replace it with a claim more representative of those who cite the classical argument to defend the split infinitive.

Second, to address the "too weak" point, starting the section with "It has been claimed" sets up the defense based on the (possibly unattested) classical argument to sound bad from the start, and between this and the "straw man" comment at the end the whole section reads as a criticism of the anti-"classical argument argument" defense. I would argue that the first sentence is every bit as much a straw man as the anti-"classical language argument" defense much of this section currently attacks.

Some points that I think are worth considering are:

Hardly anyone would claim that A. modern prescriptivists that suggest avoiding the split infinitive are making their arguments based on a faulty comparison with classical languages (hence why I don't think the defense is a straw man), although they might suggest that B. it is due to an appeal to the authority of early prescriptivists, who they believe (likely without reliable evidence, and possibly incorrectly) made their argument based on a faulty comparison with classical languages. If there are credible experts who make the second claim, and it is necessary to keep this in the section, making it clear (by being more explicit) that it is the second claim they are making, not the first, would help to not set up the defense as sounding bad to begin with.
Although, as far as I can tell from the discussion page and based on Bailey's article, no one seems to be able to find a citation of anyone making the classical languages argument, this is a very different thing from saying that this argument has never actually been made, as the section strongly implies. In fact, even if no citations are ever found, I find it very difficult to believe that this argument was never made by early prescriptivists, many of whom believed that English grammar was inferior to the more perfect Latin and classical Greek (i.e. English had "decayed") and made other prescriptions based on this misconception.
An argument that attacks an opposing argument that was never made is not necessarily a straw man. It can only be a straw man if the argument uses this attack to avoid the opponents' real arguments, which does not appear to be the case here. Even then, the term "straw man" carries a normative judgment that I don't think is appropriate if the person making the attack was simply mistaken about whether the opponents' argument had ever been made, in the same way that I don't think it is appropriate to call a spoken untruth a "lie" if it is based on mistaken information.

Here are my suggestions:

Introduce the section with a more representative claim, something like "Some lingists and modern grammarians have suggested that early proscriptions of the split infinite may have been based on a faulty comparison with classical languages". This can easily be cited (and should be) with statements from recognized experts in linguistics and modern prescriptivists who make this weaker claim. AEBoEU could be called out as an extreme example early in the section, or it could be merely cited at the beginning, with the real content staying where it is in the section (and after some better, more NPOV exposition). I am of the opinion that it is probably not necessary to mention claim B from above, even if it can be well cited, since the "due to an appeal to the authority of early prescriptivists" part of the claim is already well covered earlier in the article.
"Many of those who accept splitting ascribe such an argument to their opponents" needs to go, unless there are more examples than AHBoEU that appear to do this.
"Straw man" possibly should be removed as well, although I could be persuaded otherwise. If "straw man" stays, I think it should be separated enough from Bailey's comment on "folklore of linguistics" to make it clear that the comment on the anti-"classical languages argument" defense being a straw man is not citing Bailey (no such comment is made in Bailey's article).
If anyone can find where the misattribution of Robert Lowth that used to be in the article comes from, and it was not simply mistaken original research of the editor, this would certainly be an interesting tidbit of information, and might give a good explaination of why the classical languages argument is so often cited even if mistakenly so.

I'll wait a while for discussion before making any changes to the article. CyborgTosser (Only half the battle) 10:38, 10 January 2007 (UTC)

Thank you, CyborgTosser, for this very thoughtful comment.

The background to this is that before we started a major (and I think generally very valuable) rewrite of this article in the second half of last year, Wikipedia itself was making this claim. See for example the version of the article which (amazingly) was featured two years ago: Wikipedia:Today's featured article/September 28, 2004. I have also heard this claim many times in conversation, but read it only once, in Bill Bryson's book Mother tongue, which is hardly heavyweight linguistics, but is valuable precisely because it reflects what relatively-well informed amateurs are thinking about language. Somebody had by that stage added the American Handbook quote, which looks rather more serious than Bill Bryson, so I took it that this was standard wisdom. As we started to search for sources, however, we drew a complete blank, much to my surprise. The discovery of the Bailey quote, indicating that linguistic research has also looked into this and drawn a blank, suggests strongly that the Latin connection was never a serious issue. Hence some editors wanted to remove it, but I felt it should be left in and debunked.

I still feel that, but am completely open on the question of how to do it. I do think a lot of people believe the Latin comparison is the ONLY argument against the split infinitive, and since we all now recognise that such an argument is fundamentally weak, I think it is often thrown up to knock down modern prescription without a debate of other issues. But I am not suggesting anyone is lying, so straw man may be the wrong term.

It is important to remember that prescription first criticised the split infinitive as recently as the 1830s. The idea that English was a degenerate language, unlike the glories of Latin and Greek, is a humanist concept that surely was long since dead in the age of the British empire and jingoism. So I would not be surprised if the comparison with Latin was never an issue in this debate.

You are right about Lowe - I would love to know where that came from.

--Doric Loon 23:14, 10 January 2007 (UTC)

Non-splitters ascribing the argument from Latin to splitters: A search of Google Books will find more citations of this than we can use from grammar books (as I recall, a lot of them are for people studying to be English teachers). I'll add some within a week (maybe sooner) if nobody does it first.

An interesting example is from Ask Oxford:

It is still widely held that splitting infinitives is wrong. This dislike is long-standing but is not well founded, being based on an analogy with Latin. In Latin, infinitives consist of only one word (e.g. crescere `to grow'; amare `to love'), which makes them impossible to split: therefore, so the argument goes, they should not be split in English either. But English is not the same as Latin.…

By "interesting" I mean that they invite queries, so someone could query them on their source for the statement about the analogy. Who knows? They could have one. However, they only take queries by snail mail, so it would be especially convenient for someone in Britain to do it. I don't want to put pressure on anyone :-) but just in case anyone wants to write them, the address is

Oxford Word and Language Service

Oxford English Dictionary Department

Oxford University Press

Great Clarendon Street

Oxford OX2 6DP

UK

As my two cents, I agree with DL that we should leave this in and debunk it, since it's so widely believed (pending a recheck of Google Books).

"Straw man": I didn't put that phrase in, but I didn't change it because I thought the meaning was weaker than it was. However, the two dictionaries I checked say that it's ascribed to an opponent purposely in order to be confuted (interesting that Oxford and Merriam-Webster use that same uncommon word), which I doubt is the case here, so I'd be happy to see it go and might take it out myself.

Lowth: You can find this claim other places, for instance with a search of Google Groups. Looking at Google Books might be interesting too. I suspect this one would be harder to find in print than the one about Latin, but I haven't checked. I also suspect it's pure academic legend.

Rewriting: I'll be happy to participate in discussions and rewriting, but right now I don't have time to even really think about it. Start without me if you want. —JerryFriedman 17:03, 11 January 2007 (UTC) (fully organic, not to be tossed)

I put in three references. Do we need more?

I did find someone who blamed Lowth in print, and somebody who blamed Lindley Murray. I'm not sure we care that much, though.

Book to look for: Hugh Sykes Davies, Grammar without Tears (1953), may have a history as well as a debunking of the argument from Latin.

Sighting in the wild of the argument from Latin (and German and French): John B. Opdycke, Get it Right! A Cyclopedia of Correct English Usage (1941) has this on p. 174: "Tho [sic] the split infinitive is generally used in current as well as in earlier literature, it is preferably to be avoided for the reason that it is hardly logical. The word to is a part of the infinitive, that is, to go is logically one term, just as aller (to go) is one word in French, gehen (to go) in German, ire (to go) in Latin. To separate the to from the go by the insertion of an adverb is very much like separating these three foreign words before the last syllable of each for the insertion of a modifier—al-rapide-ler, geh-schnell-en, ir-celeriter-e. In only a few cases does the split infinitive add anything by way of grace or euphony or rhythm to expression. In most cases it is awkward and inchoate. To emphatically understand and to thoroly [sic] investigate may be accepted. But to hurriedly achieve, to distinctly and unmistakably concur, and to beautifully decorate condemn themselves. Macaulay's free use of the split infinitive has apparently provided the liberalists in English grammar with irrefutable argument in its behalf." —JerryFriedman 19:48, 16 January 2007 (UTC)

Argument from other languages

I rewrote it in view of the above. Much of what we were discussing became irrelevant. Now to see how everyone else revises it. I'm about to write to Bailey, by the way. —JerryFriedman 22:36, 16 January 2007 (UTC)

Sorry, Jerry, but this is NOT the argument from classical (or even "other") languages. It is the argument from the full infinitive. Opdycke's key argument is "the word to is a part of the infinitive, that is, to go is logically one term" - an argument from the inherent structures of English. The fact that he goes on to draw illustrative parallels with other languages is neither here nor there. There is nothing reprehensible about such supporting analogies. There is something entirely reprehensible, though, about taking other languages AS THE STARTING POINT and judging English by their standards. THAT is the issue at stake in the classical languages debate.

The argument from classical languages says that a construction is wrong in English if it is wrong in Latin, because Latin is the standard which other languages have to match up to. This argument is very commonly ascribed to prescriptivists, but - your Opdycke quote notwithstanding - we STILL have not found any prescriptivist who uses it. The point of the section on the argument from classical languages is to make that clear. So I am going to revert your changes. But the Opdycke quote is useful for illustrating the point made in the section above, and I think it might be worth adding there. --Doric Loon 23:08, 16 January 2007 (UTC)

Well, I disagree slightly with Doric Loon -- sure, the arrogance of assuming that what was right for Latin is right for English was annoying and false; but the illogic was a key point too: split infinitives aren't illegal in Latin; they're (according to that conception) impossible. Even if you do worship Latin the analogy makes no sense.

That said, though, I agree with you in being unimpressed by the Opdycke argument and its relevance to a "argument from other languages" paragraph. Plus, I am highly doubtful that he even knows much about Latin grammar (which might explain his syllabification of ire). Doops | talk 23:18, 16 January 2007 (UTC)

True, they can hang together. But we have dealt with the English logic argument separately from the veneration of Latin argument, and I think it is good to keep them separate. Anyway, I have tried to find an alternative way to use Opdycke. See if you like it. If not, feel free to revert. --Doric Loon 23:34, 16 January 2007 (UTC)

I'm not at all sure the English-logic argument should be kept separate from the veneration-of-Latin argument. I think the argument goes: "To go" is really one word in English (full infinitive), as "proved" because it's often translated by a single word in other languages, especially that most venerable of all languages, Latin. By the way, the quotation from Bailey may need some attention, as the argument from a purist he can't find is "Many people who ought to know better—'authorities' on English—declare that the objection to separating to from the infinitive verb that follows is based on Latin (or some other language) where infinitives are single words." This is precisely Opdycke's argument, but note the mention of Latin.

While I'm at it, I think we should restore the AHDEU's "English is not Latin" to source our ~~"non sequitur"~~ "methodological error" argument. But I'm going to have to get back to this later. —JerryFriedman 23:41, 16 January 2007 (UTC)

The more I think about it, the more I think the supposed argument from Latin is an attempt to bolster the argument from the full infinitive, so they should be treated together (and my mistake was not doing so). But we'll have to look at splitters who mention this argument and see what they ascribe to their opponents. —JerryFriedman 00:17, 17 January 2007 (UTC)

I'm completely with Jerry Friedman on this one. Even if Opdycke doesn't come out explicitly to give the argurment outlined above, his choice of analogy makes it clear that he had an idea of what was an "infinitive" in English that misses the subtle differences between an English to make and a German machen, and given other parts of the paragraph, I think it is clear he is implicitly making that argument. Of course, others may disagree with me on how clear this is, but, assuming this is the first known record of this argument, it is also probably the source of the "folklore of lingistics", even if that's all it is. I think there is enough here to dismiss the "straw man" comment as Wikipedian speculation (I will admit, I am guilty of this sometimes myself), unless there is a recognized authority out there that is making a similar comment. Since the authoritative sources that mention the classical languages argument hint that it is more prevalent or given more weight than it seems we can support here (at least for now) with the one source, I think it may be worth mentioning that when the classical languages argument is mentioned by splitters, it is perhaps exaggerated. However, one sentence beyond a quote from Bailey ought to do it for that, unless there is a serious debate going on about this subject beyond the halls of Wikipedia.

I favor putting Opdycke back into the article, with the possible exaggeration by the sources that defend the split infinitive on the weakness of the classical languages argument qualified as such. There is hardly a difference between the argument AHBoEU mentions and Doric Loon's interpretation of Opdycke for purposes of organization of the article, which is to say that even allowing that there may be significant differences, these are distinctions that are not widely recognized as differentiating the two as separate arguments. Put another way, you won't find a paper in The Annals of Split Infinitive Proscription attacking the "(Mis-)cited by AHBoEU argument" and "Doric Loon's interpretation of Opdycke's argument" as separate pro-splitting views. CyborgTosser (Only half the battle) 07:09, 17 January 2007 (UTC)

OK, it may seem like I'm splitting hairs, and I concede I'm still groping towards the point here. But I think the difference is that Opdycke (despite his silly parody of what splitting would be in Latin, French and German) is approaching this in a perfectly respectable manner: look at the structures of English, use analogies with other languages to help you understand them, and draw conclusions about English usage. Like both of you, I think he misunderstands the English infinitive, but I don't think Latin is the cause of this misunderstanding: rather it comes from the fact that English convention includes the marker in the citation form where other languages, even ones like German which have the marker, omit it from the citation form. Perhaps I am being too generous to him, but I think he is using other languages much the way I would use them - to shed extra light on what we find in English by noting parallels and contrasts, which in itself is perfectly respectable. Because drawing lessons from comparative linguistics is a good thing.

I think you're being a little too generous. It looks to me like he's using three languages as supposed proof that "to go" is "logically" one word. —JerryFriedman 22:43, 17 January 2007 (UTC)

On the other hand, the position which is frequently attributed to prescriptionists, what we have been calling the Argument from Classical Languages, is one which assumes that the classics are the highpoint of human civilisation and all virtue lies in imitating them - a renaissance/humanist position which obviously has no place in modern linguistics. We have called this a straw man, and whether we choose to stick with that phrase or not, it is definitely true that people who want to knock prescription (e.g. Bill Bryson) often use this as a very easy way to do it: accuse prescritpionists of Renaissance thinking and they appear absurdly behind the times. Because prescribing for one language on the basis of another is a bad thing.

Now I think we have pretty well established that SERIOUS linguists opposed to splitting (of which there were quite a number in the 19th century) never used the second argument. What is still unclear is whether it is a complete straw man, or whether some second-rate English teachers in past generations were stupid enough that they really did use it. If they were, then it is possible that Opdycke is one of them, and that I really am being too generous to him. But I would like to see at least ONE unambiguous example before we interpret unclear cases in such a way.

Of course, if I am the only person here who sees a substantial difference between these two arguments... --Doric Loon 14:49, 17 January 2007 (UTC)

Here's a quotation that goes about as far as possible in the direction of attributing veneration of Latin (plus a little historical error). "The rule forbidding a split infinitive comes from the time when Latin was the universal language of the world. All scholarly, respectable writing was done in Latin. Scientists and scholars even took Latin names to show that they were learned. In Latin, infinitives appear as a single word. The rule which prohibits splitting an infinite [sic] shows deference to Latin and to the time when the rules which governed Latin grammar were applied to other languages." Writing Science Through Critical Thinking by Marilyn F. Moriarty, p. 253.

Like everyone else, she's unclear about the "absurd" (as she says later) argument she thinks anti-splitters made. It might look something like this:

It's an undeniable rule of Latin grammar that you can't say things like i-celeriter-re.
Because we old-school prescriptivists defer to Latin, we apply that rule to English.
The closest English equivalent to ire is to go.
Therefore in English you can't or shouldn't say "to speedily go".

I'm not sure why I'm putting this here, but maybe it will be food for someone's thought. —JerryFriedman 22:43, 17 January 2007 (UTC)

Yup, that (both the Moriarty quote and your very clear dissection of it) is precisely the argument from classical languages. I'm not quite clear, though, whether Moriarty is pro or con. If she is pro-splitting, characterising anti-splitters in this way, then it is the best example yet of the straw man. If she is anti-splitting herself, and seriously using this argument, then that turns all my conclusions so far upside down. Either way it's a useful quote. --Doric Loon 11:10, 18 January 2007 (UTC)

First of all, this is a very good discussion, and thank you especially to JerryFriedman for doing some more research on the topic. Given the sources that we now have, here are some new suggestions:

Given that we have some sources on both sides, I don't think it would be controversial to start the section with something like "An often cited argument for prohibition of the split infinitive is that the construction is not possible in (Latin/classical languages/certain other languages)". This is verifiable, and neutral in that it doesn't specifically imply either that 1. many prescriptivists use this argument or 2. the pro-splitters just made it up, either of which would compromise the neutrality of the section if given the prominent position of the first sentence.
Early in the section, perhaps the second sentence, we could have something like "Although it is not clear that this argument has ever been common among prescriptivists (with Richard Bailey, a professor of English, supposing it is "part of the folklore of linguistics"), <insert here something that in a neutral way points out that at least a similar argument has been made by at least one anti-splitter, and that it is repeatedly mentioned by pro-splitters>".

Now we have some context of where the argument comes from. This would be a good time to explain the argument fully, in a moderate form, and to explain the linguists' dissatisfication with the argument, without getting into the (potentially speculative) debate of why the argument seems to be attacked by pro-splitters more often and more regularly than it is actually used by prescriptivists. This can wait until after the argument is presented and refuted (and here I assume that the only disagreements we or the experts have is whether this argument was ever common, not whether it might be a valid argument). Another thing that should probably be moved further back in the section is the attribution of this argument when it is given to the humanist idea of the purity of the classics. Opdycke certainly doesn't mention this idea. I think it would be good to keep at least a brief mention of this rather than pull it completely, since it can give some insight into 1. why someone might make the classical languages argument (speculative, I admit, but tied to my second point) and more importantly 2. why pro-splitters (who often are keenly aware of this humanist idea as applied to other prohibited constructions by early prescriptivists) might (probably unintentionally) mischaracterize or exaggerate the prevelance of the argument. (a wishlist item for this point would be a good source to cite that discusses the humanist element of early prescriptivism in general)

I think the source that blames Lowth in print might be useful if we want to mention misattribution of the argument. Stating as fact that pro-splitters have made some mistakes in attributing the classical languages argument is in my opinion a far cry better than hinting that they might have made the whole thing up.

I also think we should try to work Moriarty in there somehow, no matter which side of the debate she is on. I can't imagine a source that more clearly explains the argument, even if in an extreme version. CyborgTosser (Only half the battle) 20:25, 18 January 2007 (UTC)

Yes, that or something like it is the right way to go. Don't think I have much to add at the moment. --Doric Loon 16:33, 19 January 2007 (UTC)

Sorry to take so long. Moriarty goes on to say that the argument she attributes to Latinists is "absurd" and that there's no reason not to split except the possibility that readers will disapprove, which writers might want to consider.

The book that mentioned Lowth has:

The prohibition of that practice was created in 1762 by one Robert Lowth, an Anglican bishop and self-appointed grammarian. Like Dryden's antiterminal-preposition rule, Lowth's anti-infinitive-splitting injunction is founded on models in the classical tongues. But there is no precedent in these languages for condemning the split infinitive because in Greek and Latin (and all the other romance languages) the infinitive is a single word that is impossible to sever.

Richard Lederer, A Man of My Words: Reflections on the English Language, St. Martin's Press, 2003, ISBN 0-312-3175-9, p. 248. —JerryFriedman 19:09, 23 January 2007 (UTC)

Another Lowth example:

This is the most famous 'split infinitive' in history (another of Bishop Lowth's prohibitions, on the grounds that infinitives cannot be split in Latin). However, the alternatives ('Boldly to go…', 'To go boldly…') sound much more clumsy and destroy the powerful rhythm of the iambic pentameter.

Peter Stockwell, Sociolinguistics: A Resource Book for Students, Routledge, 2002, 0-415-23452-2, p. 98. This is a citation for the claim that the Star Trek quote is the most famous split infinitive—if this guy is more reliable on his topic than on the history of prescriptive grammar. —JerryFriedman 19:48, 23 January 2007 (UTC)

Trying again

Okay, I gave it another shot. See what you think. I took out parts that say things we now have quotes for. —JerryFriedman 00:04, 25 January 2007 (UTC)

Jerry, I am less uncomfortable with this than with what you did last time. BUT - you leave Moriaty looking like an authority. It needs to be pointed out that she is wrong (attestably, factually) when she says that the prescriptive rule dates from a time when all serious writing was in Latin. That ceased to be the case in the 16th century, but opposition to the split infinitive begins in the 19th. I think you need to say much more strongly that like all of the others who make this claim she cites no sources and is obviously utterly unable to find a SERIOUS writer who has ever argued this way. (I still doubt that ANY prescriptionist has ever argued in quite this way, but since I don't want to reopen debate on subtleties, I will content myself with the statement that the only writer we have sofar found who argued in anything remotely resembling this manner was (as I think we have all agreed) pretty second-rate.) I find that the last sentence of a section carries particular weight, and I would like to see the last sentence of this section coming back to the point that anti-prescriptivist assersions about the ACL are as yet entirely unproven. --Doric Loon 10:31, 25 January 2007 (UTC)

I think you're right, and I tried, but now it may be too long. I welcome other contributions. —JerryFriedman 18:42, 25 January 2007 (UTC)

No, that's good. Well done. --Doric Loon 16:47, 28 January 2007 (UTC)

I agree. Good NPOV, well cited, and not too long in my opinion. CyborgTosser (Only half the battle) 12:07, 29 January 2007 (UTC)

Split infinitives in Latinism context (additive reference)

This is in reference to the impossibility of 'splitting' infinitives. To be sure: English(analytic) was modeled after latin (synthetic) by the earliest prescriptive pedants enthralled with Latin who failed to see that Latin infinitives consist of one word and are impossible split. (Q.v. Sir Thomas Smith's cortex-withering DeRecta et Emendata Linguae Anglicae Scriptione Dialogus of 1568.)

Sorry, could you make that a bit clearer? If the Smith quote is relevant, please give us it. But don't deride 16th-century writers for the crime of thinking 16th-century thoughts. There was nothing pedantic about valuing Latin at a time when all serious literature was written in the language, and there was nothing simple-minded about modelling English on Latin at a time when philology had not yet come to realise either the nature of English or the very modern principle that each language is as good as any other. Looking down on past generations is an anachronistic approach - probably more unscholarly than the faults you criticise them for. So please help us to clarify what Smith and others thought, but do it with clear citations (in Latin if he wrote Latin) and objective analysis, not with value judgments. --Doric Loon 16:44, 28 January 2007 (UTC)

Chicago Manual of Style

In my experience, The Chicago Manual of Style has generally been cited as the definitive source when it comes to contemporary American English. Now, I don't know much about the text myself, nor whether it would cover the split infinitive, but it certainly seems to me as if, in an article whose subject is primarily the controversy over a grammatical construction, the lack of any reference to the Manual is very strange. Is there a reason for this, or is it because no one's gotten around to digging through it yet? Ourai ^тʃ_с 00:54, 2 April 2007 (UTC)

I suspect the latter. Why don't you do it? I think you will probably find it makes no mention, since "style" here means commas and points rather than grammar, but check. --Doric Loon 07:27, 2 April 2007 (UTC)

I used to work for a publication that went by Chicago, and I feel certain the manual has something to say about split infinitives. However, I think it's most widely used in academic publishing, where it's rivaled by the APA and MLA. I suspect it has little influence on journalism or literature. But by all means, another citation from a prestigious source would be helpful. —JerryFriedman 22:24, 2 April 2007 (UTC)

It does indeed. Here's the start of their discussion of split infinitives in their on-line Q&A:

Dear Grammar Geekess, CMOS has not, since the thirteenth edition (1983), frowned on the split infinitive. The fifteenth edition now suggests, to take one example, allowing split infinitives when an intervening adverb is used for emphasis (see paragraphs 5.106 and 5.160).

—JerryFriedman 05:29, 7 April 2007 (UTC)

No such thing

Ok, I'd like to get people's opinion on something. The Collins Concise English Dictionary sitting on my bookshelf has a special supplement on English usage and is quite explicit in it's treatment of split infinitives. It states that there is no such thing as a split infinitive in the English language as there is, strictly speaking, no such thing as an infinitive in (modern) English. It goes on to state that confusion arises as English uses the preposition "to" followed by a base form of a verb in a way that is similar to a Latin infinitive. Further, that grammatical rules against splitting this form are merely a prejudice but it should still be avoided as it is a clumsy construction.

Seemingly those working at Collins take the stance that the infinitive can't be split as it simply doesn't exist. I'm not suggesting the wikipedia entry should be rewritten but it seems to have completely omitted this point (unless I've gone blind). How do people think we should reconcile the two viewpoints, if at all? As the one of the highest circulated dictionaries in the UK I'm inclined to believe Collins - although I'd feel happier if it was the OED - perhaps I'll check that too. AulaTPN 22:12, 6 April 2007 (UTC)

This looks to me like a mere difference in nomenclature. The argument that the Collins folks are making looks like the same as Jespersen's in the "History of the term" section of the article, only Collins has changed "infinitive" to "base form". (I think there's a tendency for the linguists' pendulum to swing between discarding the musty old Latin terms and reviving the useful and widely understood Latin terms. I'm probably paranoid in believing this comes from a desire to stay one up on non-linguists.) Possibly Collins should be mentioned as lining up with Jespersen. But "split infinitive" is what everyone calls it, and we should stay with common usage unless it changes.

I think the article amply demonstrates that the "to amply demonstrate" construction (I'm pretending to be NPOV about nomenclature) has an interesting history and is not merely the kind of feature that doesn't need encyclopedic discussion, the way "will amply demonstrate" doesn't. In other words, there is such a thing. It also demonstrates that rules against splitting are not necessarily prejudice, though in the last century and a half prejudice may have contributed to them.

What the Oxford folks say on line about split infinitives is here. I quoted it above, so I'll just point out that they don't criticize the term. Also, they say the prohibition is based on an analogy with Latin. Since they take queries on paper, I still think someone in Britain should ask them for their source. Maybe they have one. Not to nag, but any volunteers?

Oxford Word and Language Service

Oxford English Dictionary Department

Oxford University Press

Great Clarendon Street

Oxford OX2 6DP

UK

—JerryFriedman 05:21, 7 April 2007 (UTC)

There is certainly such a thing: the question is what to call it. The article does acknowledge the Collins perspective: in the section on history of the term, there is a paragraph saying that some linguists regard the phrase "split infinitive" as misleading for precisely the reasons your Collins quote suggests. But it also notes that no alternative terminology has been proposed, and therefore we stick with this phrase. We have to call it something, because the issue is frequently discussed. --Doric Loon 08:58, 7 April 2007 (UTC)

Compound Split Infitive

Does anybody know how old this term is, or where it originated? I wanted to add a note on this to the History of the Term section, where we have earliest attested references for other related forms, but I couldn't find anything on it. I suspect it is relatively recent. I hope it was not coined for our Wikipedia article! --Doric Loon 11:51, 3 May 2007 (UTC)

Origins

In the last paragraph of the section entitled "Origins", the sentence "She will gradually get rid of her teddy bears" is given as an example of the tendency toward placing adverbs imediately precedent to an infinitive verb. While this sentence is indeed exemplary of such a tendency; being itself a case of an indivisable infinitive, it isn't very helpful in explaining incidences of split infinitives. I think it should be less confusing if the article were to use an example with a "to" infinitive. Considder "She wishes to gradually get rid of her teddy bears." or "She is going to gradually get rid of her teddy bears."--Jr mints 18:57, 21 May 2007 (UTC)

But I think that's the point. The sentence She will gradually get rid of her TBs may or may not contain an infinitive, depending on how you analyze it. Will get is the future tense. Historical linguists will see this tense as being formed from a modal + infinitive construction, but synchronic linguists simply see the whole phrase as a two-part finite verb. But even synchronic linguists may see an analogy to an infinitive construction. So this sentence is not meant to be a split infinitive - it is, rather, an example of a finite verb being split by an adverbial, which is very common in English. The theory is that this then was transferred to the infinitive proper, producing a split infinitive. To make sense, the argument has got to start from an example which is finite. --Doric Loon 21:57, 21 May 2007 (UTC)

Rhetoric

I did read somewhere (unfortunately I cannot remember where) that the split inifinitive should be considered a rhetorical error, not a grammatical error, and that people often confused the two. Esthameian 22:43, 27 May 2007 (UTC)